We examined tracheal aspirates from infants admitted to the C.S. Mott Children’s Hospital Newborn Intensive Care Unit, as approved by the University of Michigan Institutional Review Board. Entry criteria included gestational age at birth ≤ 32 weeks, mechanical ventilation for respiratory distress, and age ≤ 7 days. Aspirates were collected during routine tracheal suctioning of mechanically ventilated premature infants in the first two weeks of life as described (49). Specimens were centrifuged (1,200 X g for 5 min at 15°C) and supernatants were stored at -80°C.

Tracheal aspirates were assayed for extracellular F-actin and gelsolin present in the supernatant using ELISA (F-actin ELISA from MyBioSource, gelsolin ELISA from LSBio).

The animal experiments were performed in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals recommendations. The protocol was approved by the University of Michigan Committee on Use and Care of Animals. Two day-old wild type C57BL/6J, Batf3^-/- (B6.129S(C)-Batf3^tm1Kmm/J) or Clec9a^{gfp -/-} (B6.Cg-Clec9a^tm1.1Crs/J) mice (Jackson Laboratories, Bar Harbor, ME) were exposed to air or 75% oxygen for up to 14 days using a polypropylene chamber coupled to an oxygen controller and sensor (BioSpherix, Lacona, NY) (50). Pups were produced by breeding homozygous pairs of each genotype. C57BL/6J, as the inbred strain is an approximate control for Batf3^-/- and Clec9a^{gfp -/-} mice. Dams were exchanged between air and hyperoxia daily. In selected experiments, neonatal mice were treated with 10ul of PBS (Sigma D6662) or recombinant human plasma gelsolin (Cytoskeleton Inc., HPG6) 0.5mg/kg intranasally daily during exposure to hyperoxia. Intranasal administration of the liquid to anesthetized mice allows for inhalation into the airway. Starting on day of life (DOL) 2, the hyperoxic exposure was continued for 10 days. On DOL 14 the mice were inoculated with 30 µl of RV1B (3 x 10⁸ PFU/ml). Lungs were analyzed immediately after hyperoxia or 2 days after sham or RV infection. Neonatal mice of both sexes were used in all experiments.

RV1B (from American Type Culture Collection, Manassas, VA) was grown in HeLa cells, concentrated, partially purified, and titered as described (51). Viral titers were measured by plaque assay (52).

Neonatal mice were exposed to normoxia or hyperoxia for 14 days. On day 14 of exposure, bronchoalveolar lavage (BAL) was performed with 0.3 mL cold PBS. BAL cells were separated from supernatant by centrifugation. F-actin and gelsolin levels in BALF supernatants were measured by ELISA. BALF supernatant (100 μl) was incubated with or without human gelsolin (5 µg) in vitro in room temperature for 10 min and F-actin levels were measured by ELISA. In selected experiments, 14-day-old wild type mice were inoculated with hyperoxic BALF 15 µl intranasally, followed by 15 µl of RV1B (3 x 10⁸ PFU/ml). Lungs were analyzed 2 days after RV infection.

Lung CD11c+ immune cells were purified using CD11c microbeads kit (Miltenyi Biotec, Auburn, CA). Briefly, single cell suspension from five lungs of wild type or Clec9a^{gfp -/-} adult mice were each pooled, incubated with anti-CD11c antibody and purified with the microbeads and columns. The purified lung CD11c+ immune cells were divided into different treatment groups with 1.5X10⁶ cells per condition each time. The purified lung CD11c+ cells were co-cultured with culture medium (10%FBS-DMEM), culture medium containing 20% of hyperoxic BALF, or hyperoxic BALF and RV1B (15 µl of 3 x 10⁸ PFU/ml) overnight in 37°C incubator. After co-culture, the cells were stained with flow antibodies of F4/80, CD11c, CD103, CD11b and IL-12 (BioLengend, San Diego, CA). The fixable live/dead staining is applied to distinguish the live cells (31). The cells were analyzed by flow cytometry machine (BD LSRFortessa) and results were analyzed by FlowJo software.

Mouse whole-lung RNA was prepared using TRIzol (Invitrogen, Carlsbad, CA). Gene mRNA expression (IL-12p40, IFN-γ, TNF-α) was quantified using SYBR green real-time quantitative PCR technology. Primer sequences are listed in Supplemental Table I. The level of gene expression was normalized to mRNA of β-actin or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as indicated, using the 2^-ΔCT algorithm. For graphic representation, normalized mRNAs for the target genes were plotted multiplied by 10^-3.

Whole mouse lung homogenates in PBS were centrifuged and supernatants analyzed for pro-inflammatory cytokines. IL-12p40, IL-12p70, TNF-α and IFN-γ were measured by ELISA (all from R&D Systems, Minneapolis, MN).

Lungs were perfused with PBS containing EDTA (0.5 mM), minced, and digested with Liberase TM (100 µg/mL; Roche, Indianapolis, IN), together with collagenase XI (250 µg/mL), hyaluronidase 1a (1 mg/mL), and DNase I (200 µg/mL; Sigma, St. Louis, MO) for 1 hour at 37°C (53). Cells were filtered and washed with RBC lysis buffer (BD Biosciences, Franklin Lakes NJ) and kept on ice in media containing 10% serum. Dead cells were stained with Pac-Orange Live/Dead fixable dead staining dye (Invitrogen). Lung cells were then stained with fluorescent-labeled antibodies against various leukocyte surface markers (CD45, CD11b, CD11c, F4/80, CD103, and IL-12). Appropriate isotype-matched controls and Fluorescence Minus One (FMO) controls were used in all experiments. Antibodies were purchased from EBiosciences (San Diego, CA) or Biolegend (San Diego, CA). Cells were fixed and analyzed on a Fortessa (Becton-Dickinson, San Jose, CA) or FACSAria II (BD Biosciences) flow cytometer. Results were analyzed using FlowJo software (Tree Star, Ashland, OR). For analysis of intracellular IL-12 or IFN-γ, fresh aliquots of digested lung tissue were stimulated for 4 h at 37°C with Cell Stimulation Cocktail plus protein transport inhibitors (40.5 µmol/L PMA, 670 µmol/L ionomycin, 5.3 mmol/L brefeldin A, and 1 mmol/L monesin [Invitrogen]), fixed, permeabilized with Cell Permeabilization Buffer (Invitrogen), and incubated with anti-mouse IL-12 clone C17.8 (BioLegend).

Lungs were perfused with 5 mM EDTA, inflated to 30 cmH2O pressure with 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO), and paraffin embedded. Five μm-thick paraffin sections were stained with Hematoxylin and eosin (H&E). To assess alveolarization, alveolar chord length was determined as described (54). In summary, alveolar chord length was calculated as the mean length of line segments on random test lines spanning the airspace between intersections of the line with the alveolar surface was calculated. Four random images from two sections for each animal were photographed at X200.

Unless otherwise noted, data are represented as mean ± standard error. Statistical significance was determined by unpaired two-tailed t-test or one-way analysis of variance, as appropriate. Statistical significance was defined as P < 0.05.

We have previously shown that hyperoxic exposure of neonatal mice (a model of BPD) increases the number of activated lung IL-12-producing, Clec9a+CD103+ DCs, induces lung pro-inflammatory responses and airway hyperreactivity following RV infection (31). We have also shown that CD103+ DCs and Clec9a signaling are required for hyperoxia-induced inflammatory responses to RV (32). In addition, we have found that hyperoxia induces airway cell death and increases F-actin levels in BALF supernatant (32). Since gelsolin depolymerizes F-actin decreasing its binding to Clec9a (47), we examined the effects of gelsolin on neonatal hyperoxia-induced inflammatory responses to RV infection.

We exposed 2-day-old C57BL/6J mice to normoxia or 75% oxygen for 10 days and administered human gelsolin or control intranasally. Two days after the exposure completed, mice were inoculated with RV intranasally. We have previously published extensive data on sham controls for normoxia- and hyperoxia-exposed neonatal mice (32). Our goal for this study was to compare the effects of gelsolin on normoxia- and hyperoxia-exposed, RV infected mice, therefore we chose not to include a sham control for this study. We found that in RV-infected mice, prior hyperoxic exposure induced type 1 cytokine expression, including Il12, Ifng and Tnfa mRNA. This response was blocked by gelsolin treatment (Figure 1A). Protein expression of IL12p70, IFN-γ and TNF-α were also blocked by gelsolin treatment compared with PBS control (Figure 1B). These results indicate that gelsolin blocked hyperoxia-induced pro-inflammatory response to RV.

Based on our previous findings that CD103+ DCs expand and mediate inflammatory gene expression during hyperoxia (31), we examined the effects of gelsolin treatment on lung CD103+ DCs. Recent reports describe two phenotypically and functionally distinct lung migratory CD103+ DC populations following respiratory viral infection, CD103lo and CD103hi DCs (55, 56). Both populations are absent in Batf3-/- mice, that lack CD103+ DCs in lungs and other organs, and unlike CD103hi DCs, CD103lo DCs express lower levels of lineage and maturation markers, including costimulatory molecules, suggesting they are phenotypically immature and functionally limited (55, 56). We looked for evidence of CD103hi and CD103lo DC populations in the lungs of neonatal mice during hyperoxia. Using flow cytometry, we examined lungs of control-treated hyperoxia- and air-exposed neonatal mice after a 4- and a 10-day exposure. These timepoints were designed to assess the acute and chronic responses. After a 4-day exposure, hyperoxia-exposed lungs had a significantly higher number of CD103lo DCs and no change in CD103hi DCs (Figures 2A–C). After a 10-day exposure, hyperoxia-exposed lungs had a significant increase in both CD103lo and CD103hi DCs (Figures 2A–C). The CD103lo and CD103hi DC populations were not present in lungs of age-matched Batf3-/- mice, confirming that both CD103lo and CD103hi DC populations are Batf3-dependent (Figure 2D). Next, we examined the effect of gelsolin on the two lung CD103+ subpopulations during normoxia or hyperoxia. Gelsolin attenuated the effect of hyperoxia on CD103lo DC expansion both after 4- and 10-day exposure, but it did not influence the hyperoxia-induced expansion of CD103hi DCs after 10-day exposure (Figures 2B, C). Taken together, these results show that the frequency of CD103hi and CD103lo DCs in neonatal lungs increases during hyperoxia and gelsolin treatment blocks the effect of hyperoxia on CD103lo DCs.

We examined the effects of gelsolin on hyperoxia-induced proinflammatory gene expression. In PBS treated mice, hyperoxia induced the mRNA expression of Il12p40, Myd88, Cd207, Cd103 and Clec9a (Figure 2E). Gelsolin treatment attenuated the effect of hyperoxia on the above pro-inflammatory gene mRNA expression (Figure 2E). This could reflect either that CD103lo DCs are the primary mediators of hyperoxia-induced proinflammatory responses, or that gelsolin dampens the proinflammatory activation of CD103hi DCs without affecting the increase in frequency during hyperoxia. Alternatively, gelsolin treatment during hyperoxia may modulate the functional properties of other cells that exert anti-inflammatory properties.

We have previously shown that hyperoxic exposure of neonatal mice increases F-actin levels in BALF supernatant (32). To assess the capacity of the actin-scavenger system in the airways, we evaluated the effect of hyperoxia on the balance of F-actin and gelsolin protein levels in the BALF supernatant. To do this we measured F-actin and gelsolin levels in BALF supernatants and calculated the ratio of F-actin to gelsolin levels. The ratio of F-actin to gelsolin levels was significantly higher in hyperoxic BALF (Figure 3A). Next, we conducted an in vitro experiment to assess if gelsolin can decrease the concentration of F-actin in hyperoxic BALF. Incubation of hyperoxic BALF with gelsolin decreased F-actin concentration (Figure 3B). We also designed an in vivo experiment and inoculated neonatal mice with hyperoxic BALF and RV, followed by gelsolin, all administered intranasally on DOL14. Whole lung mRNA expression of Il12p40, IFN-γ and TNF-α was assessed 2 days later. The expression of Il12p40, IFN-γ and TNF-α was most significantly upregulated by the combined treatment with hyperoxic BALF and RV and gelsolin attenuated this effect (Figure 3C). All other treatments had very small, nonsignificant effects. These results demonstrate that hyperoxia-induced F-actin in BALF plays a key role in mediating proinflammatory responses, especially in the presence of RV.

To further confirm that the effect of hyperoxic BALF on proinflammatory responses to RV is dependent on Clec9a, the receptor for F-actin, we isolated lung CD11c+ cells from wild type and Clec9a^{gfp -/-} mice. Subsequently, we incubated the CD11c+ cells with hyperoxic BALF and RV. The population of CD11c+ cells in the lung includes both populations of conventional lung DCs, CD103+ DCs and CD11b+ DCs, and resident alveolar macrophages. After 24 hours the cells were subjected to flow cytometry to quantify activation and IL-12 production in lung CD103+ DCs (Figures 4A–C). In wild type CD11c+ cells, hyperoxic BALF increased the frequency of IL-12+CD103+ DCs and addition of RV further increased the frequency of these cells. In Clec9a^{gfp -/-} CD11c+ cells, the effect of hyperoxic BALF on IL-12+CD103+ DCs was smaller and nonsignificant and the IL-12 responses of these cells to RV were also attenuated (Figures 4B, C). These results indicate that Clec9a expression on CD103+ DCs is required for maximal proinflammatory responses to RV in the presence of hyperoxic BALF, though the remaining response suggests that other signaling pathways contribute to proinflammatory activation of the CD103+ DCs under these conditions.

Inflammation and hypoalveolarization are key histopathologic features that occur in parallel in infants with BPD (10, 11, 57). In neonatal mice, exposure to hyperoxia in early life recapitulates the inflammatory and structural changes associated with human BPD (31, 58). The fact that gelsolin dampened proinflammatory gene expression during hyperoxia prompted us to examine the effects of gelsolin on hyperoxia-induced hypoalveolarization. We administered recombinant human plasma gelsolin (Cytoskeleton,Inc) 0.5 mg/kg intranasally daily during hyperoxic exposure. Hyperoxia was continued for 10 days, starting on DOL 2. Mouse lungs were assessed on DOL 16. In the BSA-PBS (control)-treated mice, as shown previously (58), hyperoxic exposure caused the development of fewer and larger air spaces compared to air-exposed mice (Figure 5A). Air exposed gelsolin-treated mice showed a normal alveolarization pattern. Unlike BSA-PBS-treated mice, hyperoxia-exposed gelsolin-treated mice did not show larger air spaces (Figure 5A). Differences in alveolarization were quantified using alveolar chord length measurements (Figure 5B). These results show that gelsolin treatment during hyperoxia preserves alveolarization.

We previously demonstrated that Clec9a+CD103+CD11c+ DCs are present in tracheal aspirates of mechanically ventilated preterm infants in the first week of life, and that tracheal aspirate CLEC9A mRNA levels positively correlated with the proinflammatory cytokine IL12B mRNA expression (32). Since F-actin is a ligand for Clec9a, and F-actin levels are regulated by gelsolin (47), we investigated the relationship between F-actin and gelsolin concentrations in human preterm infant tracheal aspirate supernatants, and the FiO2 on the day of sample collection. We obtained tracheal aspirates from 29 premature infants receiving mechanical ventilation for respiratory distress in the first week of life (Table 1). We found that the ratio of F-actin to Gelsolin protein levels positively correlated with FiO2 on the day of sample collection (Figure 6). These results demonstrate that F-actin levels relative to Gelsolin levels are increased in preterm infants receiving higher levels of supplemental oxygen.

Cumulative supplemental oxygen exposure during the first two weeks after a preterm birth is a risk factor for BPD development (59). This suggests a critical exposure window that modulates disease development and may involve deficiency of protective mechanisms. We examined changes in gelsolin concentrations in tracheal aspirate supernatants during the first two weeks of life in premature infants. We found that gelsolin levels decrease between week one and week two of mechanical ventilation (Figure 7). These findings demonstrate that prolonged mechanical ventilation of preterm infants is associated with relative deficiency in extracellular gelsolin and development of F-actin-dominated disbalance in the actin-scavenger system in the airways.