Attenuation of Hyperoxic Lung Injury in Newborn Thioredoxin-1-Overexpressing Mice through the Suppression of Proinflammatory Cytokine mRNA Expression

The role of thioredoxin-1 (TRX), a small redox-active protein with antioxidant effects, during hyperoxic lung injury in newborns remains undetermined. We investigated TRX impact on hyperoxic lung injury in newborn TRX transgenic (TRX-Tg) and wildtype (WT) mice exposed to 21% or 95% O2 for four days, after which some mice were allowed to recover in room air for up to 14 days. Lung morphology was assessed by hematoxylin/eosin and elastin staining, as well as immunostaining for macrophages. The gene expression levels of proinflammatory cytokines were evaluated using quantitative real-time polymerase chain reaction. During recovery from hyperoxia, TRX-Tg mice exhibited an improved mean linear intercept length and increased number of secondary septa in lungs compared with the WT mice. Neonatal hyperoxia enhanced the mRNA expression levels of proinflammatory cytokines in the lungs of both TRX-Tg and WT mice. However, interleukin-6, monocyte chemoattractant protein-1, and chemokine (C-X-C motif) ligand 2 mRNA expression levels were reduced in the lungs of TRX-Tg mice compared with the WT mice during recovery from hyperoxia. Furthermore, TRX-Tg mice exhibited reduced macrophage infiltration in lungs during recovery. These results suggest that in newborn mice TRX ameliorates hyperoxic lung injury during recovery likely through the suppression of proinflammatory cytokines.


Introduction
Bronchopulmonary dysplasia (BPD) is one of the most severe complications affecting premature infants [1]. Despite the remarkable improvements in neonatal-perinatal medicine, the overall incidence of BPD has not changed substantially. Recent studies have investigated the long-term consequences of BPD on the respiratory health of older children and adults. In extremely low-birth-weight survivors, who were born weighing less than 1000 g, lung function values at 8 and 12 years of age were lower than the reference values, especially in children with a history of severe BPD and in those exhibiting to produce either TRX heterozygous or WT littermates. Genotyping was performed using polymerase chain reaction (PCR) analysis of tail biopsies.

Neonatal Hyperoxic Exposure and Recovery
Newborn pups (< 12 h old) were randomly assigned to the normoxia (normal room air) or hyperoxia (95% O 2 ) condition. Exposure to hyperoxia was performed for 96 h in a chamber (BioSpherix, Redfield, NY, USA) that allowed for the continuous monitoring and regulation of O 2 and CO 2 . The dams were switched between hyperoxia and normoxia every 24 h. The inside of the chamber was maintained at atmospheric pressure, and the mice were exposed to a 12 h light/dark cycle. After hyperoxic exposure, some mice were allowed to recover in normoxia until they were 14 days old.

Lung Tissue Collection
Mice were anesthetized with intraperitoneal injection of pentobarbital (50 mg/kg) after determining the body weight. The pulmonary artery was perfused with phosphate-buffered saline via the right ventricle, and the right lung was excised and snap-frozen in liquid nitrogen for RNA and protein analyses. The left lung was inflated through the trachea with 10% neutral-buffered formalin (Sigma-Aldrich, St. Louis, MO, USA) at 25 cm gravity pressure and allowed to fix in situ for 1 min; the trachea was tied, and the left lung was then removed and fixed further overnight at 4 • C. The left lung tissue was paraffin-embedded, and 5-µm-thick sections were mounted onto glass slides for further analyses.

Lung Histology and Morphometry
Computer-aided morphometric analysis was performed using ImageJ software version 1.49 (NIH, Bethesda, MD, USA) to determine distal airspace maturation, and mean linear intercept length (Lm) and the number of secondary septa were determined.
Paraffin-embedded lung tissue sections were stained with hematoxylin and eosin. Lm, defined as the mean length of line segments on random test lines spanning the airspace between intersections of the line with the alveolar surface, was obtained using light microscopy by dividing the total length of a line drawn across the lung section by the total number of intercepts, until the number of intercepts reached 50 for each field [28]. Lm was assessed in six nonoverlapping fields of lung parenchyma in one tissue section per animal, and six animals per condition at each time point were examined.
Elastin staining was assessed using an elastic stain kit (Abcam, Cambridge, MA, USA) according to the manufacturer's instructions. The number of secondary septa, where elastin was detected, was manually counted in six nonoverlapping fields of lung parenchyma in one tissue section per animal, and six animals per condition at each time point were examined.

Immunoprecipitation
Lung 12,000× g supernatants from the hyperoxia-exposed WT or TRX-Tg mice were incubated with 2 µg hTRX antibody (ab133524; Abcam, Cambridge, UK) or nonimmune antibodies (normal rabbit IgG; Wako, Osaka, Japan) for 1 h at 4 • C. Next, 50 µL nProtein A Sepharose 4 Fast Flow suspension (GE Healthcare, Uppsala, Sweden) was added, and the mixture was incubated for 1 h at 4 • C. After centrifugation at 12,000× g for 20 s, the supernatants were removed and analyzed to detect MIF by Western blotting. Sample inputs were simultaneously loaded.

Statistical Analysis
All values were expressed as means ± standard error of the mean. Comparisons between groups were performed using the statistical package JMP ver. 14 (SAS Institute, Cary, NC, USA) by one-way analysis of variance, followed by Tukey's post hoc test for parametric data or the Kruskal-Wallis test followed by the Mann-Whitney U test for nonparametric data. P-values less than 0.05 were considered to indicate statistical significance.

Neonatal Hyperoxia Does Not Negatively Affect Body Weight in TRX-Tg Mice
The effect of neonatal hyperoxia on body weight was assessed in the WT and TRX-Tg mice. On day 4 after neonatal hyperoxia, no significant decrease in body weight was observed in the WT or the TRX-Tg mice compared with their normoxic counterparts. However, neonatal hyperoxia led to a reduction in body weight in both the male and female WT mice on day 14 during recovery from hyperoxia, whereas no change in body weight was observed in the TRX-Tg mice on day 14 following neonatal hyperoxia (Figure 1).

Figure 1.
Changes in body weight after neonatal hyperoxic exposure. The changes in body weight were evaluated in 4 and 14-day-old WT and TRX-Tg mice following normoxic or hyperoxic exposure (n = 6 per group). Data are shown as means ± SEM. Comparisons between groups were performed using one-way ANOVA followed by Tukey's test. ** p < 0.01. WT: wildtype; TRX: Thioredoxin-1; Tg: Transgenic; Air: Normoxia; O2: Hyperoxia; SEM: Standard error of the mean; ANOV: Analysis of variance.

Alveolar Development Is Impaired After Neonatal Hyperoxia in Both WT and TRX-Tg Mice
In normal room air, the mice developed well-organized terminal airways. In contrast, the exposure of newborn mice to hyperoxia for 96 h led to the impairment of alveolar development, resulting in alveolar expansion and simplification ( Figure 2A). Compared with the normoxic control lungs, a longer Lm was observed after neonatal hyperoxic exposure in the lungs of both the WT and TRX-Tg mice on day four ( Figure 2B). Neonatal hyperoxia also led to a reduction in the number of secondary septa in the lungs of both the WT and TRX-Tg mice ( Figure 2C,D).

Alveolar Development Is Impaired After Neonatal Hyperoxia in Both WT and TRX-Tg Mice
In normal room air, the mice developed well-organized terminal airways. In contrast, the exposure of newborn mice to hyperoxia for 96 h led to the impairment of alveolar development, resulting in alveolar expansion and simplification ( Figure 2A). Compared with the normoxic control lungs, a longer Lm was observed after neonatal hyperoxic exposure in the lungs of both the WT and TRX-Tg mice on day four ( Figure 2B). Neonatal hyperoxia also led to a reduction in the number of secondary septa in the lungs of both the WT and TRX-Tg mice ( Figure 2C,D).

Alveolar Development Is Impaired After Neonatal Hyperoxia in Both WT and TRX-Tg Mice
In normal room air, the mice developed well-organized terminal airways. In contrast, the exposure of newborn mice to hyperoxia for 96 h led to the impairment of alveolar development, resulting in alveolar expansion and simplification ( Figure 2A). Compared with the normoxic control lungs, a longer Lm was observed after neonatal hyperoxic exposure in the lungs of both the WT and TRX-Tg mice on day four ( Figure 2B). Neonatal hyperoxia also led to a reduction in the number of secondary septa in the lungs of both the WT and TRX-Tg mice ( Figure 2C,D).

Impaired Alveolar Development Is Mitigated in TRX-Tg Mouse Lungs during Recovery from Hyperoxia
On day 14, the Lm was significantly longer and the number of secondary septa were significantly lower in the neonatal WT mice that were exposed to hyperoxia for four days followed by recovery in normal room air for ten days, compared with their normoxic counterparts. In contrast, despite the impaired alveolarization after neonatal hyperoxia on day 4, the TRX-Tg mice had a significantly improved Lm ( Figure 3B) and exhibited an increase in the number of secondary septa ( Figure 3D) after recovery in normal room air on day 14 compared with the WT mice that were exposed to the identical conditions.

Impaired Alveolar Development Is Mitigated in TRX-Tg Mouse Lungs during Recovery from Hyperoxia
On day 14, the Lm was significantly longer and the number of secondary septa were significantly lower in the neonatal WT mice that were exposed to hyperoxia for four days followed by recovery in normal room air for ten days, compared with their normoxic counterparts. In contrast, despite the impaired alveolarization after neonatal hyperoxia on day 4, the TRX-Tg mice had a significantly improved Lm ( Figure 3B) and exhibited an increase in the number of secondary septa ( Figure 3D) after recovery in normal room air on day 14 compared with the WT mice that were exposed to the identical conditions.

Hyperoxia-Induced Increases in Proinflammatory Cytokine and Chemokine mRNA Expression Levels Disappeared During Recovery from Hyperoxia in TRX-Tg Mouse Lungs
TRX was reported to exert chemotaxis-modulating functions and to suppress leukocytic infiltration to sites of inflammation [20]. Therefore, the lung mRNA expression levels of proinflammatory cytokines and chemokines, including Il-6, Mcp-1, Il-1β, Tnf, Cxcl1, and Cxcl2, were determined using quantitative real-time PCR. Compared with the normoxic controls, Il-6, Mcp-1, Il-1β, Cxcl1, and Cxcl2 mRNA expression levels were significantly increased in the lungs of both the WT and TRX-Tg mice after hyperoxic exposure on day four ( Figure 4A). However, after a 10-day recovery in normal room air, the mRNA expression levels of Il-6, Mcp-1, and Cxcl2 in the lungs of TRX-Tg mice exposed to neonatal hyperoxia returned to basal levels and were significantly lower than those in the hyperoxia-exposed WT mice ( Figure 4B). TRX was reported to exert chemotaxis-modulating functions and to suppress leukocytic infiltration to sites of inflammation [20]. Therefore, the lung mRNA expression levels of proinflammatory cytokines and chemokines, including Il-6, Mcp-1, Il-1β, Tnf, Cxcl1, and Cxcl2, were determined using quantitative real-time PCR. Compared with the normoxic controls, Il-6, Mcp-1, Il-1β, Cxcl1, and Cxcl2 mRNA expression levels were significantly increased in the lungs of both the WT and TRX-Tg mice after hyperoxic exposure on day four ( Figure 4A). However, after a 10-day recovery in normal room air, the mRNA expression levels of Il-6, Mcp-1, and Cxcl2 in the lungs of TRX-Tg mice exposed to neonatal hyperoxia returned to basal levels and were significantly lower than those in the hyperoxia-exposed WT mice ( Figure 4B).

Macrophage Infiltration Is Reduced in the Lungs of TRX-Tg Mice after Neonatal Hyperoxia
Based on the observed suppression of macrophage-related cytokine mRNA expression levels in the lung of TRX-Tg mice, we assessed the extent of macrophage infiltration in the lungs by immunohistochemistry using the macrophage marker F4/80. Compared with their normoxic counterparts, an increase in the number of macrophages was observed in the lungs of WT mice after hyperoxic exposure on day four and during recovery from hyperoxia on day 14. In contrast, in the lungs of TRX-Tg mice, although the higher basal macrophage numbers were observed under normoxic conditions, the hyperoxia-induced macrophage infiltration was inhibited on day 4 and the number of macrophages was decreased during recovery from hyperoxia on day 14 compared with the lungs of WT mice ( Figure 5D).

Macrophage Infiltration Is Reduced in the Lungs of TRX-Tg Mice after Neonatal Hyperoxia
Based on the observed suppression of macrophage-related cytokine mRNA expression levels in the lung of TRX-Tg mice, we assessed the extent of macrophage infiltration in the lungs by immunohistochemistry using the macrophage marker F4/80. Compared with their normoxic counterparts, an increase in the number of macrophages was observed in the lungs of WT mice after hyperoxic exposure on day four and during recovery from hyperoxia on day 14. In contrast, in the lungs of TRX-Tg mice, although the higher basal macrophage numbers were observed under normoxic conditions, the hyperoxia-induced macrophage infiltration was inhibited on day 4 and the number of macrophages was decreased during recovery from hyperoxia on day 14 compared with the lungs of WT mice ( Figure 5D).

Lung Mif mRNA and Protein Levels Are Not Different Between WT and TRX-Tg Mice after Hyperoxic Exposure
An in vitro study previously demonstrated the regulatory involvement of TRX in MIF internalization and signaling, such as the augmentation of TNF-α production, by the direct binding of TRX with MIF [30]. Therefore, we used quantitative RT-PCR and Western blotting to assess mRNA and protein expression levels of MIF and performed immunoprecipitation to assess the direct binding of TRX with MIF in our in vivo model. There were no differences in the mRNA and protein expression levels of MIF between the lungs of WT and TRX-Tg mice exposed to hyperoxia ( Figure 6A-D), although the MIF mRNA and protein expression levels were increased in the lungs of TRX-Tg mice during recovery from hyperoxia on day 14 ( Figure 6D). No direct binding was observed between TRX and MIF by immunoprecipitation followed by Western blotting ( Figure 6E). An in vitro study previously demonstrated the regulatory involvement of TRX in MIF internalization and signaling, such as the augmentation of TNF-α production, by the direct binding of TRX with MIF [30]. Therefore, we used quantitative RT-PCR and Western blotting to assess mRNA and protein expression levels of MIF and performed immunoprecipitation to assess the direct binding of TRX with MIF in our in vivo model. There were no differences in the mRNA and protein expression levels of MIF between the lungs of WT and TRX-Tg mice exposed to hyperoxia ( Figure 6A-D), although the MIF mRNA and protein expression levels were increased in the lungs of TRX-Tg mice during recovery from hyperoxia on day 14 ( Figure 6D). No direct binding was observed between TRX and MIF by immunoprecipitation followed by Western blotting ( Figure 6E).

Discussion
In the present study, the newborn TRX-Tg mice tended to attenuate hyperoxia-induced lung injury compared with the newborn WT mice. During recovery from hyperoxia, the lungs of newborn TRX-Tg mice exhibited a shorter Lm and a greater number of secondary septa compared with the newborn WT mice, indicating that TRX overexpression mitigated the arrested alveolar development caused by neonatal hyperoxia. Furthermore, the mRNA expression levels of proinflammatory cytokines and chemokines, such as Il-6, Mcp-1, and Cxcl2, and the macrophage infiltration in the lungs of TRX-Tg mice exposed to neonatal hyperoxia were significantly suppressed on day 14 compared with the WT mice, suggesting that recovery from hyperoxic lung injury observed in the TRX-Tg mice might be due to the anti-inflammatory activity of TRX in the lungs of newborn mice. However, we did not observe an interaction between TRX and MIF in the lungs of TRX-Tg mice exposed to normoxia or hyperoxia.
Several animal models have shown the anti-inflammatory properties of TRX. We previously reported that recombinant human TRX attenuated the rate of lipopolysaccharide-induced preterm delivery in mice by preventing the elevation of proinflammatory cytokines including TNF-α, interferon-γ, MCP-1, and IL-6 in the maternal serum [23]. Using a mouse model of COPD exacerbation induced by cigarette smoke, Tanabe et al. demonstrated that TRX ameliorated neutrophilic inflammation by suppressing the release of granulocyte-macrophage colony-stimulating factor, thereby preventing the progression of emphysema, indicating the potential of TRX as a novel therapeutic agent that might counteract COPD exacerbation [31]. In a mouse model of irritant contact dermatitis induced by croton oil, topically applied recombinant human TRX significantly suppressed the inflammatory response by inhibiting the production of cytokines and chemokines, such as Tnf-α, Il-1β, Il-6, Cxcl1, and Mcp-1, in the skin tissues [32]. Furthermore, exogenous administration of recombinant human TRX significantly improved the survival rate and attenuated the histological changes in lungs in a mouse model of influenza pneumonia by diminishing the production of TNFα and CXCL1 in the lungs observed in influenza-inoculated mice [33]. In agreement with these animal models, we observed the anti-inflammatory properties of TRX, including the suppression of mRNA

Discussion
In the present study, the newborn TRX-Tg mice tended to attenuate hyperoxia-induced lung injury compared with the newborn WT mice. During recovery from hyperoxia, the lungs of newborn TRX-Tg mice exhibited a shorter Lm and a greater number of secondary septa compared with the newborn WT mice, indicating that TRX overexpression mitigated the arrested alveolar development caused by neonatal hyperoxia. Furthermore, the mRNA expression levels of proinflammatory cytokines and chemokines, such as Il-6, Mcp-1, and Cxcl2, and the macrophage infiltration in the lungs of TRX-Tg mice exposed to neonatal hyperoxia were significantly suppressed on day 14 compared with the WT mice, suggesting that recovery from hyperoxic lung injury observed in the TRX-Tg mice might be due to the anti-inflammatory activity of TRX in the lungs of newborn mice. However, we did not observe an interaction between TRX and MIF in the lungs of TRX-Tg mice exposed to normoxia or hyperoxia.
Several animal models have shown the anti-inflammatory properties of TRX. We previously reported that recombinant human TRX attenuated the rate of lipopolysaccharide-induced preterm delivery in mice by preventing the elevation of proinflammatory cytokines including TNF-α, interferon-γ, MCP-1, and IL-6 in the maternal serum [23]. Using a mouse model of COPD exacerbation induced by cigarette smoke, Tanabe et al. demonstrated that TRX ameliorated neutrophilic inflammation by suppressing the release of granulocyte-macrophage colony-stimulating factor, thereby preventing the progression of emphysema, indicating the potential of TRX as a novel therapeutic agent that might counteract COPD exacerbation [31]. In a mouse model of irritant contact dermatitis induced by croton oil, topically applied recombinant human TRX significantly suppressed the inflammatory response by inhibiting the production of cytokines and chemokines, such as Tnf-α, Il-1β, Il-6, Cxcl1, and Mcp-1, in the skin tissues [32]. Furthermore, exogenous administration of recombinant human TRX significantly improved the survival rate and attenuated the histological changes in lungs in a mouse model of influenza pneumonia by diminishing the production of TNF-α and CXCL1 in the lungs observed in influenza-inoculated mice [33]. In agreement with these animal models, we observed the anti-inflammatory properties of TRX, including the suppression of mRNA expression levels of proinflammatory cytokines and chemokines, as well as the suppression of macrophage infiltration, in the hyperoxia-induced BPD model in newborn mice.
MIF, a proinflammatory cytokine, is released by T lymphocytes and can inhibit the random movement of macrophages [34]. In addition to T lymphocytes, MIF is secreted by a variety of other cells, including epithelial cells, endothelial cells, and macrophages [35]. MIF has been classified as a powerful cytokine, capable of inducing TNF-α, IL-1β, IL-6, and IL-8 and amplifying lipopolysaccharide-driven cytokine responses [36]. Son et al. demonstrated the specific affinity of TRX to MIF within cells, as well as culture supernatants of ATL-2 cells, which led to the MIF internalization into the ATL-2 cells; the authors also showed that MIF mediated the augmentation of TNF-α production from RAW264.7 macrophage cells [30]. In the current study, to examine the role of MIF in the observed differences between the WT and TRX-Tg mice during recovery from neonatal hyperoxia, we assessed the Mif mRNA and protein expression levels, direct binding of TRX with MIF, and Tnf gene expression levels in the lungs. In contrast to the previously reported in vitro findings, MIF and its downstream target TNF did not appear to be involved in the protective effects of TRX in the hyperoxic lung injury model used in the present study.
In clinical settings, extracellular concentrations of TRX were measured in various conditions characterized by oxidative stress and inflammation, including autoimmune diseases, ischemia-reperfusion injury, sepsis, viral infection, and acute lung injury [37][38][39][40]. Yu et al. reported that serum TRX might be a biomarker of cardioembolic stroke severity and that increased TRX levels might be a useful tool for predicting favorable prognosis in patients with acute ischemic stroke [41]. In patients with coronary artery disease accompanied with hyperhomocysteinemia, decreased serum TRX levels were closely correlated with the extent and severity of coronary artery disease [42]. TRX levels in human neonates, especially very premature babies, who often suffer from BPD, have not yet been investigated. To consider the utility of TRX as a biomarker or a novel therapeutic agent for BPD, the association of TRX levels in human samples, such as serum and tracheal aspirates, with the development and/or severity of BPD should be investigated. Furthermore, preclinical studies in animal models of BPD are warranted to assess the efficacy of treatment with recombinant human TRX.

Conclusions
Hyperoxic lung injury was attenuated in the newborn TRX-Tg mice compared with the newborn WT mice. The mechanism for the observed beneficial effect of TRX on lung development might occur through the suppression of proinflammatory cytokine gene expression after neonatal hyperoxic exposure, in the absence of MIF involvement.

Conflicts of Interest:
The authors have no affiliations with or involvement in any organization or entity with any financial interest, or nonfinancial interest in the subject matter or materials discussed in this manuscript.