R213G polymorphism in SOD3 protects against bleomycin-induced inflammation and attenuates induction of proinflammatory pathways

Abstract

Extracellular superoxide dismutase (EC-SOD), one of three mammalian SOD isoforms, is the sole extracellular enzymatic defense against superoxide. A known human single nucleotide polymorphism (SNP) in the matrix-binding domain of EC-SOD characterized by an arginine-to-glycine substitution at position 213 (R213G) redistributes EC-SOD from the matrix into extracellular fluids. We previously reported that knock-in mice harboring the human R213G SNP (R213G mice) exhibited enhanced resolution of inflammation with subsequent protection against fibrosis following bleomycin treatment compared with wild-type (WT) littermates. Herein we set out to determine the underlying pathways with RNA-Seq analysis of WT and R213G lungs 7 days post-PBS and bleomycin. RNA-Seq analysis uncovered significant differential gene expression changes induced in WT and R213G strains in response to bleomycin. Ingenuity Pathways Analysis was used to predict differentially regulated up- and downstream processes based on transcriptional changes. Most prominent was the induction of inflammatory and immune responses in WT mice, which were suppressed in the R213G mice. Specifically, PKC signaling in T lymphocytes, IL-6, and NFΚB signaling were opposed in WT mice when compared with R213G. Several upstream regulators such as IFNγ, IRF3, and IKBKG were implicated in the divergent responses between WT and R213G mice. Our data suggest that the redistributed EC-SOD due to the R213G SNP attenuates the dysregulated inflammatory responses observed in WT mice. We speculate that redistributed EC-SOD protects against dysregulated alveolar inflammation via reprogramming of recruited immune cells toward a proresolving state.

INTRODUCTION

Inflammation and fibrosis are conserved biologic responses that are central to diverse lung and pulmonary vascular diseases, including acute respiratory distress syndrome, pulmonary hypertension (PH), and interstitial lung diseases (5, 26, 35). A central feature of inflammatory and fibrotic diseases is the contribution of oxidative stress, which can be defined as a disruption in redox signaling and can arise in the setting of impaired antioxidant capacity (4, 6, 16, 27, 28, 30, 34).

The antioxidant enzyme extracellular superoxide dismutase (EC-SOD), which catalyzes the dismutation of superoxide to hydrogen peroxide and oxygen, is abundant in the lung and vasculature (24). Low EC-SOD activity is implicated in human lung diseases including idiopathic pulmonary fibrosis and PH, and numerous animal models characterized by inflammation and fibrosis (9, 10, 19–22, 29). Not only does the expression level of EC-SOD in the lung influence the severity of illness, but so does its tissue distribution (7–9, 19). The other prominent feature of EC-SOD is its positively charged COOH-terminal heparin binding domain (HBD), which dictates the localization of EC-SOD to the negatively charged extracellular matrix of the lung and vasculature. A naturally occurring human nonsynonymous single nucleotide polymorphism (SNP), rs1799895, within the HBD of EC-SOD results in an arginine-to-glycine amino acid substitution at position 213 (R213G), which decreases the matrix binding affinity of EC-SOD without altering enzyme activity and redistributes active EC-SOD from the lung into the plasma and epithelial lining fluid (9). Humans expressing the R213G variant of EC-SOD demonstrated an increased risk of cardiovascular disease but decreased risk of lung diseases, including COPD and asthma (7, 9, 12, 13, 19). Similar observations have been observed in studies utilizing the mice expressing the knock-in of the identical R213G SNP (7, 9, 19).

The HBD sequence is conserved between humans and mice, allowing us to test how alterations in the distribution of EC-SOD impact disease progression in a mouse strain genetically engineered to express a knock-in of the identical human SNP. Parallel to human studies, the R213G mice exhibit an increased risk of chronic hypoxic PH but attenuated intratracheal (IT) LPS-induced alveolar inflammation (9). Relevant to this current project, we also recently demonstrated that the R213G mice are protected against IT bleomycin (bleo)-induced lung fibrosis (19). The IT bleo model is a well-established model results in early alveolar inflammation, fibrosis, and PH secondary to fibrosis (11). Bleo-treated R213G mice demonstrated a robust initial release of pro- and anti-inflammatory cytokines into the bronchoalveolar fluids (BAL), as well as an early 7-day inflammatory cell influx (19). In contrast to the wild-type (WT) mice, the R213G mice were able to fully resolve the inflammatory response and prevent ongoing fibrosis and PH (19). In this study, we used an unbiased approach to interrogate the pathways responsible for the proresolving protective phenotype in the R213G mice, with RNA sequencing (RNA-Seq) with bioinformatic analyses to gain novel insight into the molecular pathways and biological processes that differed between these two strains and predicted upstream regulators that could account for these different disease responses.

MATERIALS AND METHODS

Mouse model.

Animal studies were approved by the University of Colorado Denver Institutional Animal Care and Use Committee. Male and female mice (8–12 wk of age) homozygous for the R213G SNP (rs1799895) in EC-SOD on the C57BL/6 mouse background (R213G) (11) and C57BL/6 WT mice were treated with a single dose of IT bleo (50 μl at 1 U/ml) or PBS and euthanized with isoflurane inhalation and thoracotomy on day 7. The lungs were flushed with 10 ml PBS via the right ventricle to remove blood, snap-frozen in liquid nitrogen, and stored at −80°C.

RNA isolation and RT-quantitative PCR.

Lungs were pulverized under liquid nitrogen with a mortar and pestle and homogenized in 2 ml Qiazol reagent (Qiagen) with a Tissue Tearor (Biospec Products). Total RNA was isolated and purified using the miRNeasy Mini Kit (Qiagen) and resuspended in RNase-free water. DNA was eliminated using the RNA-free DNase set (Qiagen).

For quantitative (q)PCR experiments, cDNA was generated with iScript cDNA Synthesis Kit (Bio-Rad). TaqMan probes (Life Technologies) were used for qPCR with the following genes assessed Ccl2 (Mm00441242_m1), Ccr2 (Mm99999051_m1), Itgam (Mm00434455_m1), Cybb (Mm01287743_m1), Mmp12 (Mm00500554_m1).

RNA-Seq.

1X150 directional paired-end mRNA sequencing was performed on an Illumina hiSEQ 4000 system at the Genomics and Microarray Core at the University of Colorado Denver - Anschutz Medical Campus. Samples were demultiplexed and aligned to reference genome mm10 using TopHAT2 (14). Transcripts for each of our 14 samples were quantified with Cufflinks and annotated with GenCode (32). Transcripts were normalized using the geometric method (17). Transcript quantities <1 fragment per kilobase of transcript per million mapped reads in all samples were removed, resulting in 14,193 detected transcripts. One WT+PBS was excluded from analysis because of failure of library prep. Mouse lung samples chosen for RNA-Seq analysis included both males and females; n = 2 WT+PBS, n = 4 WT+bleo, n = 4 R213G+PBS, n = 4 R213G+bleo.

RNA-Seq data analysis.

All hierarchical clustering and principal component analyses were performed with R (R Foundation for Statistical Computing, Vienna, Austria). To investigate molecular pathways, biological processes and toxicity functions associated with differentially regulated genes, Ingenuity Pathway Analysis (IPA, Qiagen), Metascape analysis, and Gene Ontology categorization using PANTHER were performed.

Statistics.

qPCR data were analyzed with Prism (GraphPad Software, La Jolla, CA) by two-way ANOVA, with post hoc analysis by Holm-Sidak. Data are expressed as means ± SE. Significance was defined as P < 0.05.

RESULTS

Whole lung transcriptome profiling of WT and R213G mice revealed significant gene expression changes and dysregulated canonical pathways in response to bleo.

To determine bleo-induced transcriptional changes in the whole lung, we conducted next-generation RNA-Seq on WT and R213G mice +/− bleo. RNA-Seq of age-matched WT and R213G mice +/−bleo identified >14,000 genes, 154 of which were uniquely differentially expressed ≥1.5-fold in WT mice compared with WT+bleo mice and 88 of which were uniquely differentially expressed ≥1.5-fold in R213G mice compared with R213G+bleo (Fig. 1B), Additionally, 1,462 bleo-responsive genes were differentially expressed ≥1.5-fold in both WT and R213G mice but exhibited a more profound fold change (≥1.5) in one strain compared with the other after bleo. Of these bleo-responsive genes, 667 were ≥2-fold different after bleo in WT vs. R213G mice, while 795 bleo-responsive genes were ≥2-fold different after bleo in R213G vs. WT mice (Fig. 1B). Thus, a total of 821 differentially expressed genes (154 unique plus 667 with ≥2-fold between strains) in WT animals were chosen for further analysis (501 upregulated and 320 downregulated), while 883 (88 unique plus 795 with ≥2-fold between strains) differentially expressed genes in R213G animals were chosen for further analysis (309 upregulated and 574 downregulated). Unsupervised hierarchical clustering using all 14,193 detected transcripts, and principal component analysis using only the 1,704 differentially expressed transcripts separated WT, R213G, WT bleo, and R213G bleo-treated groups (Fig. 1, A and C).

Fig. 1.

Transcriptome analysis of wild-type (WT) and R213G mouse lung in response to bleomycin (bleo) treatment. A: heat map of the 14,182 detected transcripts (Log₂ FPKM). Unsupervised hierarchical clustering separates WT, R213G, WT+bleo, and R213G+bleo groups. B: the Venn diagram indicates select genes chosen for further analysis. In yellow are the 154 genes uniquely regulated in WT compared with WT+bleo (≥1.5-fold) and unchanged in R213G mice (<1.2-fold). In blue are the 88 genes uniquely regulated in R213G compared with R213G+bleo (≥1.5-fold) and unchanged in R213G compared with R213G+bleo (<1.2-fold). In purple are the 1,462 common differentially expressed genes (>1.5-fold) in both WT and R213G groups +/−bleo, but whose fold change was ≥1.5 between strains; 667 of these genes were ≥ 1.5 fold different in WT compared with R213G in response to bleo, while 795 of these genes were ≥1.5-fold different in R213G compared with WT in response to bleo. C: unsupervised principal component analysis (PCA) of differentially expressed genes illustrates segregation of each group between the principal components.

Pathway analysis performed with Ingenuity Pathway Analysis (IPA) revealed multiple dysregulated canonical pathways associated with bleo treatment in WT mice. Most prominently, canonical pathways involved in: recognition of bacteria/viruses, inflammation, triggering receptor expressed on myeloid cells (TREM1), leukocyte extravasation, and production of nitric oxide (NO) and reactive oxygen species (ROS) signaling were significantly enriched (P < 0.01), and activated (positive activation z-score) in bleo-treated WT mice relative to PBS-treated WT controls (Fig. 2A). PANTHER was used to further categorize the 821 differentially expressed transcripts in WT animals in response to bleo, based on implicated biological processes. The top three categorizations indicate a majority of these genes are involved in cellular processes (40%, e.g., cell communication, cell cycle, and movement of cellular components), metabolic processes (29%, e.g., primary metabolism and nitrogen compound metabolism), and response to stimuli (20%, e.g., stress response and immune response) (Fig. 2A).

Fig. 2.

Transcriptome profiling indicates differential induction of canonical pathways related to inflammatory and innate immune responses in WT and R213G mice in response to bleo treatment. A: top 5 most significantly dysregulated canonical pathways in response to bleo treatment (ranked by P value, Fisher’s exact test) with predicted activation z-score in WT mice identified with Ingenuity Pathways Analysis (IPA) and PANTHER categorization of Gene Ontology (GO) annotations for biological processes indicated in response to bleo treatment in WT animals. B: top 5 most significantly dysregulated canonical pathways in response to bleo treatment (ranked by P value, Fisher’s exact test) with predicted activation z-score in R213G mice identified with IPA and PANTHER categorization of GO annotations for biological processes indicated in response to bleo treatment in R213G animals. C: predicted activation z-scores of the top 10 most significantly dysregulated inflammatory pathways in WT (red) and R213G (blue) mice in response to bleo treatment.

IPA analysis of R213G lungs revealed unique canonical pathways regulated in response to bleo. Most prominently, canonical pathways involved in leukocyte extravasation, integrin-linked kinase (ILK), calcium, protein kinase A (PKA), and regulation of actin-based motility signaling were significantly enriched (P < 0.01) in bleo-treated R213G mice relative to PBS-treated controls (Fig. 2B). With the exception of PKA signaling, all of the top five most significantly regulated canonical pathways in R213G mice in response to bleo were predicted to be repressed (negative activation Z-score) (Fig. 2B). PANTHER was used to further categorize the 883 distinct expressed transcripts in R213G animals in response to bleo, based on implicated biological processes. The top three categorizations indicate a majority of these genes are involved in cellular processes (37%, e.g., cell communication, cell cycle, and movement of cellular components), metabolic processes (28%, e.g., primary metabolism and nitrogen compound metabolism), and biological regulation (11%, e.g., regulation of homeostasis, biological and molecular function) (Fig. 2B).

A detailed IPA analysis of canonical pathways related specifically to inflammatory and immune responses revealed differential induction of multiple inflammatory and immune responses in WT vs. R213G mice in response to bleo. For example, TREM1 signaling, interferon signaling, complement system, T-helper cell signaling, phagocytosis in macrophages and monocytes, and production of NO and ROS in macrophages are all significantly differentially regulated (P < 0.05) and predicted to be activated (positive activation z-score) in WT mice in response to bleo (Fig. 2C). However, these same pathways either have no significant dysregulation or are predicted to be repressed (negative activation z-score) in R213G mice in response to bleo (Fig. 2C).

Transcriptome profiling indicates induction of inflammatory and immune responses in WT mice in response to bleo.

A similar IPA analysis of specific biological processes and associated functional pathways based on differentially expressed genes in WT mice revealed that the most significantly dysregulated processes in response to bleo are indeed related to activation of inflammatory and immune responses (Fig. 3A). For example, biological processes such as cellular movement, inflammation, and immune cell trafficking are all predicted by IPA to be activated (positive activation z-scores) in WT mice in response to bleo (Fig. 3, A and B). Additionally, significantly dysregulated functional processes (P ≤ 0.00001), including: recruitment of cells (recruitment of leukocytes and phagocytes), activation of cells (leukocyte and macrophage activation), engulfment of cells, and extravasation are all predicted to be activated in WT animals in response to bleo (Fig. 3C).

Fig. 3.

Transcriptome profiling indicates differential induction of biological processes related to inflammatory and innate immune responses in WT and R213G mice in response to bleo treatment. A: heat map of implicated biological processes altered in response to bleo in WT mice. Upregulated processes (positive z-score) are depicted in orange, and downregulated processes (negative z-score) are depicted in blue; squares are sized based on predicted z-scores in IPA. Highlighted in black boxes are the biological processes related to inflammatory response, immune cell trafficking, and cellular movement. Below are the related pathway wheels containing differentially expressed genes involved in inflammation and innate immunity processes (e.g., extravasation, activation of macrophages, and innate immune response). B: heat map of implicated biological processes altered in response to bleo in R213G mice. Upregulated processes (positive z-score) are depicted in orange, and downregulated processes (negative z-score) are depicted in blue; squares are sized based on predicted z-scores in IPA. Highlighted in black boxes are the biological processes related to inflammatory response, immune cell trafficking and cellular movement. C: predicted activation z-scores of the top 30 most significantly dysregulated biological processes in WT (red) and R213G (blue) mice in response to bleo treatment.

Transcriptome profiling indicates suppression of inflammatory and innate immunity biological processes in R213G mice in response to bleo.

The same analysis was used to determine differentially regulated biological processes associated with R213G mice. This revealed that many of the most significantly dysregulated processes in response to bleo include suppression of the inflammatory and innate immune responses (Fig. 3). For example, biological processes such as cellular movement, inflammation, and immune cell trafficking, which are activated in WT animals in response to bleo (Fig. 3A), are all predicted to be repressed (negative activation Z-scores) in R213G mice in response to bleo (Fig. 3B). More specifically, significantly dysregulated functional processes, including: recruitment of cells (leukocytes and phagocytes), activation of cells, and leukocyte extravasation are all predicted to be repressed in R213G animals in response to bleo (Fig. 3C).

Transcriptome profiling of WT and R213G mice identifies upstream regulators of gene expression in response to bleo.

To identify upstream regulators of bleo-responsive transcripts, further analysis was conducted. Cytokines, transcriptional activators, kinases and phosphatases were predicted upstream regulators of the differential gene expression profile in bleo-treated WT and R213G animals (Fig. 4, A–D). Importantly, a number of functionally relevant upstream regulators were predicted to be significantly activated in WT animals, but not in R213G animals in response to bleo. For example, proinflammatory cytokines including IFNG, TNF, and IL1B are predicted to be significantly activated in WT mice but repressed in R213G mice in response to bleo (Fig. 4E). Similarly, key transcription factors involved in the innate immune response including IRF7, IRF 3, and STAT1 are predicted to be highly activated in WT mice but not in R213G mice in response to bleo (Fig. 4E). Immune-responsive serine kinases IKBKB, IKBKG, and IKBKE are also predicted to be activated in WT but not in R213G animals in response to bleo (Fig. 4E). Finally, phosphatase PTPRJ is predicted to be activated, while suppressor of cytokine signaling (SOCS) is predicted to be repressed, in WT animals but not in R213G animals in response to bleo (Fig. 4E).

Fig. 4.

Predicted upstream regulators of gene expression in response to bleo in WT vs. R213G mice. Cytokines (A), transcriptional activators (B), kinases (C), and phosphatases (D) predicted to be activated (orange) or suppressed (blue) in WT vs. R213G mice in response to bleo ranked by P value. E: summary table including activation z-scores and P values of functionally relevant upstream regulators in response to bleo in WT vs. R213G mice.

Validation of targets.

We sought to validate a series of genes implicated in bleo-induced inflammation that were differentially expressed by RNA-Seq analysis. We selected genes established in the recruitment of innate immune cells such as Ccl2, Ccr2, and Itgam (CD11b), activation of leukocytes Cybb (Nox2), and the metalloprotease Mmp12. The fold change expression in each of these genes was significantly greater in response to bleo in WT compared with R213G (Fig. 5, B–F).

Fig. 5.

Validation of differentially expressed proinflammatory genes. A: list of relevant inflammatory genes from RNA-Seq analysis that were differentially regulated between WT and R213G mice postbleo and their respective fold change that were selected for quantitative PCR validation. B–F: in similar fashion to RNA-Seq data, all validated genes were significantly reduced in the R213G+bleo group when compared with WT+bleo. n = 3–4, Two-way ANOVA, P < 0.05.

DISCUSSION

IT instillation of bleo has been well established and commonly used to model pulmonary fibrosis, as well as acute lung injury and PH associated with lung disease. Bleo initially causes direct increases in ROS and subsequent alveolar epithelial cell injury (2). The injured alveolar epithelial cells and the oxidative stress within the alveolar compartment drive recruitment and activation of inflammatory cells, leading to increased ROS levels. The dysregulated inflammatory response involves a vast expansion of recruited innate immune cells including neutrophils and monocytes/macrophages, which, if unresolved, drives subsequent lung fibrosis (18).

We have previously demonstrated that the release of EC-SOD into the serum and BAL, seen in the genetically engineered mouse strain harboring R213G EC-SOD, protects against bleo-induced inflammation and fibrosis (19). The protection conferred by the R213G variant is observed despite low EC-SOD lung content. This is the first study to interrogate transcriptome changes in the whole lung in response to bleo in WT mice and assess pathways altered in the R213G strain that could account for the observed protection in bleo-induced lung injury. We uncovered marked disparate differences in gene regulation between WT and R213G strains. Bioinformatic analysis reveals key differences between pathways and molecules in WT and R213G that provide insight into potential mechanisms responsible for protection of the R213G strain in bleo-induced lung injury.

Transcriptome profiling of these data sets provides an unbiased assessment of the differences in response to bleo-induced lung injury in WT vs. R213G strains. As expected, increases in inflammatory responses and immune cell trafficking at 7 days postbleo are elevated in WT mice. Overall, the same biological processes are starkly attenuated in the bleo-treated R213G strain. When examining the specific canonical pathways regulated in the two strains in response to bleo, we observe different patterns. Certain well-established canonical pathways are elevated in WT mice, such as TREM1 signaling and the production of NO and ROS (23, 36), but have no predicted activation state in R213G mice. In another subset of pathways, the WT and R213G strains share canonical pathways such as leukocyte extravasation signaling, but these pathways are differentially altered with a predicted activation in WT mice and attenuation in R213G mice. R213G mice also exhibit uniquely altered pathways such as inhibition of ILK signaling and activation of PKA signaling, which are unchanged in WT mice. ILK signaling has been previously implicated as a proinflammatory regulator in colitis and a driver of chemotaxis in vitro, while PKA/cAMP signaling has been shown to be anti-inflammatory (1, 3, 31). As a result, this analysis begins to identify the differences between the susceptible WT strain and protected R213G strain, and those pathways unique to the R213G stain open up new avenues worthy of exploration given to their implication in alleviating dysregulated inflammation.

Underlying the canonical pathways are alterations in cellular movement, inflammatory responses, and immune cell trafficking. All three processes are significantly upregulated in the WT mice and starkly attenuated in the R213G mice. The three biological functions that are most significantly dampened in the R213G strain include the extravasation, recruitment, and activation of leukocytes. Upon examining upstream regulators of biological processes, we found from our data that the R213G strain has a suppressed Th1-response with the attenuation of several upstream Th1 cytokines (e.g., IL-1β), transcription factors (e.g., IRF3), and kinases (e.g., IKK-γ). These data both reaffirm the importance of dysregulated inflammation in settings of lung injury and the importance of dampening the associated oxidative stress to resolve the inflammation and subsequently the injury.

The R213G EC-SOD variant does not impact its overall activity but instead alters its binding affinity and thus distribution. We, therefore, propose the increased unbound EC-SOD in the circulation and alveolar compartments protects against alveolar inflammation in the setting of bleo due to its effect on cells in these specific sites. In the circulation, increased oxidative stress can prime neutrophils and monocytes toward a proinflammatory state (25, 33). Additionally, injured and activated endothelial cells serve as a binding site for the recruitment of monocytes and neutrophils to the lung by increasing adhesion of proinflammatory leukocytes (15). Thus, increased circulating EC-SOD could result in lowering the oxidative burden and priming of recruited innate immune populations and vascular endothelial cells. Within the alveolar spaces, increased unbound EC-SOD can also attenuate the oxidative stress that is accompanied by the recruitment of leukocytes to the alveolar spaces. Thus, given the established increase in oxidative stress observed (alveolar and circulating) in response to bleo, we propose that the redistribution of EC-SOD into the fluids is a driving force behind the redox-dependent protection imparted in the R213G strain. This conclusion would be consistent with our prior observation that the R213G strain is also protected against early LPS-induced inflammation but is not protective in the setting of tissue oxidative stress as observed in chronic hypoxic PH (9). These findings are in parallel with the impact of the SNP in human disease risk and provide an opportunity to understand the role of redox-regulated mechanisms in different disease processes.

To our knowledge, this is the first study to date utilizing RNA-Seq in an IT bleo-model of lung injury and fibrosis. The unbiased nature of this RNA-Seq experiment allows us to conclude that the whole lung transcriptional changes in R213G mice are primarily related to modulation and downregulation of inflammatory pathways, with little contribution from other pathways. More specifically, pathways and biological processes related to leukocyte extravasation, macrophage activation, and T-helper cell signaling among others were overrepresented in the downregulated genes. Therefore, these findings serve as confirmation that differential expression of transcripts specifically related to inflammatory/immune modulation likely play a major role in the protective effects elicited by this particular SNP in the context of lung injury and warrants further study. Intriguingly, principal component analysis suggests that the strains at baseline have altered transcriptome profiles. Whether underlying differences at baseline reprogram the lung and immune responses resulting in the contrasting differences postbleo remains unclear. We speculate that changes in recruitment and programming state of recruited monocytes/macrophages and neutrophils could be driving the accelerated resolution observed.

The current study has several limitations. Whole lungs were analyzed rather than specific cell types. As a result, it remains unknown what cell populations in the R213G strain contribute to the changes seen in innate immune pathways. While the study included both male and female mice, the study was not powered to compare sex differences within each strain. Since others have reported both sex- and age-dependent changes in the bleo model, this will be a future area of study (11). Additionally, the exact cellular sources of the R213G EC-SOD in the circulation or alveolar compartment remain unknown. As a result, the intricate cell and molecular mechanisms at play remain to be elucidated. This current study serves as a strong foundation for ongoing inquires to determine the role of innate immune cell populations in the protection imparted in the R213G strain.

Taken together, these data suggest that the redistribution of EC-SOD is capable of attenuating recruitment and activation of inflammatory responses. Bleo lung injury and fibrosis have been shown to be mediated via increases in oxidant-mediated injury and subsequent dysregulated inflammation. Although tissue levels are low, the site of the oxidant injury in the IT-bleo model is the alveolar spaces. Thus the redistribution of EC-SOD into the alveolar microenvironment could be sufficient to accelerate the resolution of inflammation.

GRANTS

This work was funded by National Heart, Lung, and Blood Institute Grants R01 HL-126928-03S1 (A. M. Garcia), T32 HL-007171 (A. Allawzi), R01 HL-119533 (E. Nozik-Grayck and C. C. Sucharov), R01 HL-086680 (E. Nozik-Grayck), and HL-126928 (C. C. Sucharov).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.M.G., A.A., K.S., and G.M. performed experiments; A.M.G., A.A., P.T., L.H.-L., K.S., A.K.-F., and C.C.S. analyzed data; A.M.G., A.A., P.T., L.H.-L., K.S., R.P.B., A.K.-F., C.C.S., and E.N.-G. interpreted results of experiments; A.M.G., A.A., and P.T. prepared figures; A.M.G., A.A., and E.N.-G. drafted manuscript; A.M.G., A.A., P.T., L.H.-L., K.S., G.M., R.P.B., A.K.-F., C.C.S., and E.N.-G. edited and revised manuscript; A.M.G., A.A., P.T., L.H.-L., K.S., G.M., R.P.B., A.K.-F., C.C.S., and E.N.-G. approved final version of manuscript; C.C.S. and E.N.-G. conceived and designed research.