Comments
Abstract
Ozone (O3)-induced respiratory toxicity varies considerably within the human population and across inbred mouse strains, indicative of gene-environment interactions (GxE). Though previous studies have identified several quantitative trait loci (QTL) and candidate genes underlying responses to O3 exposure, precise mechanisms of susceptibility remain incompletely described. We sought to update our understanding of the genetic architecture of O3 responsiveness using the Collaborative Cross (CC) recombinant inbred mouse panel. We evaluated hallmark O3-induced inflammation and injury phenotypes in 56 CC strains after exposure to filtered air or 2 ppm O3, and performed focused genetic analysis of variation in lung injury, as reflected by protein in lung lavage fluid. Strain-dependent responses to O3 were clear, and QTL mapping revealed two novel loci on Chr (Chromosomes) 10 (peak, 26.2 Mb; 80% confidence interval [CI], 24.6–43.6 Mb) and 15 (peak, 47.1 Mb; 80% CI, 40.2–54.9 Mb), the latter surpassing the 95% significance threshold. At the Chr 15 locus, C57BL/6J and CAST/EiJ founder haplotypes were associated with higher lung injury responses compared with all other CC founder haplotypes. With further statistical analysis and a weight of evidence approach, we delimited the Chr 15 QTL to an ∼2 Mb region containing 21 genes (10 protein coding) and nominated three candidate genes, namely Oxr1, Rspo2, and Angpt1. Gene and protein expression data further supported Oxr1 and Angpt1 as priority candidate genes. In summary, we have shown that O3-induced lung injury is modulated by genetic variation, identified two high priority candidate genes, and demonstrated the value of the CC for detecting GxE.
Ozone (O3) is a potent oxidant gas and ground-level air pollutant. Acute O3 exposure causes temporary decrements in lung function, respiratory inflammation, and tissue injury, as well as aggravating symptoms of common chronic lung diseases including asthma and chronic obstructive pulmonary disease (1). Short-term O3 exposure is also associated with an increased risk of respiratory tract infections and hospitalization (2). Importantly, these adverse outcomes have been linked with the pathogenesis of respiratory diseases (3). Thus, because of its involvement in disease incidence and exacerbations (4), identifying the molecular circuitry by which O3 exposure causes pulmonary inflammation and injury is a critical public health goal.
Controlled exposure and candidate gene studies in humans have provided strong evidence that responses to O3 vary widely and reproducibly across individuals, partially attributed to genetic variation (5–9). Studies using panels of inbred rodent strains have further demonstrated genetic effects on ozone response, including evidence of heritability, and led to the identification of multiple loci that control hallmark respiratory responses such as airway neutrophilia and injury (10, 11). These foundational studies proposed important candidate genes including Tlr4, Nos2, and Tnf, whose roles in O3 respiratory toxicity have since been thoroughly studied (12–17). Here, we sought to further characterize the genetic architecture of response to acute O3 exposure, focusing on injury and inflammation, thereby illuminating novel gene-by-environment interactions (GxE) for future experimental validation and investigation.
For the current study, we used the CC genetic reference population, a multiparental inbred strain panel derived by funnel inbreeding of five classical (A/J, C57BL/6J, 129S1SvImJ, NOD/ShiLtJ, and NZO/H1LtJ) and three wild-derived mouse strains (CAST/EiJ, PWK/PhJ, and WSB/EiJ). Nearly all previously published studies investigating genetic contributions to O3 responses focused on responses in classical laboratory strains, failing to capture the full breadth of genetic diversity available within the Mus musculus species. The CC captures over 90% of circulating genetic variants (>40 million SNPs and several million indels and small structural variants), owing in large part to the inclusion of wild-derived strains from three different Mus musculus subspecies (M. m. domesticus, castaneus and musculus) (18). Moreover, the breeding strategy used to generate CC lines created novel allelic combinations, thus enhancing the range of phenotypic variation observed compared with that across the founder strains alone (19). This population has already been used to identify genomic regions associated with various respiratory disease phenotypes including allergic inflammation in the airways (20, 21) and susceptibility to virus-induced respiratory disease endpoints (22). Hence, this resource is well suited for toxicogenomic discovery and identification of genes and variants contributing to O3-induced responses in the lung.
We measured inflammatory and injury responses in mice from 56 CC strains after a 3-hour exposure to filtered air (FA) or 2 ppm O3. The analyses here are primarily focused on total protein concentration in the bronchoalveolar lavage fluid, a marker of lung injury, for which we identified two novel genetic loci (one significant and one suggestive) associated with variation in responsiveness. Using a weight of evidence approach along with available genome sequence data, we prioritized candidate genes within the significant locus, located on Chr15. Finally, we confirmed protein expression of these candidates in lung tissue using immunohistochemistry.
Some of the results of these studies have been previously reported in the form of a preprint (bioRxiv, [21 May 2021] https://doi.org/10.1101/2021.05.20.445039).
Additional details about the methods used for this study, including data and code availability, are provided in Supplemental Methods and Supplemental Files.
We obtained female and male mice from 56 Collaborative Cross strains from the UNC Systems Genetics Core Facility between August and November 2018, housed in groups of three or more under normal 12-hour light/dark cycles with ad libitum food (Envigo 2,929) and water on Teklad 7070C bedding (Envigo). All studies were in compliance with University of North Carolina Institutional Animal Care and Use Committee guidelines in animal facilities approved and accredited by Association for Assessment an Accreditation of Laboratory Animal Care International.
We performed experiments when mice were between 9 and 12 weeks old. We randomly assigned mice to groups within a pair or trio, and randomized all pairs within a strain over the course of the study to mitigate batch effects (Figure E1 in the data supplement). We exposed mice to FA or 2 ppm O3 as described previously (23) with highly precise, stable concentrations of O3 across all batches (Figure E2). Further details are included in Table E1.
We returned mice to their normal housing upon cessation of exposure for 21 hours, then sacrificed the animals by lethal urethane injection to collect two BAL fractions, as described previously (23). We quantified total BAL protein using the Quant-iT Protein Assay kit (Thermo Scientific).
We prepared tissue sections from the main axial airway using samples from our previously published study involving FA- and O3-exposed adult, female C57BL/6J mice (23). We stained paraffin sections with antibodies raised against ANGPT1 (Invitrogen, cat# PA5–96841, polyclonal, 1:2000 dilution) or OXR1 (Invitrogen, cat# PA5–50079, polyclonal, 1:1000 dilution).
We isolated total RNA from banked lung tissue representing two strain pairs per haplotype using the Qiagen RNeasy Mini Kit. In triplicate reactions, we used 5 ng cDNA as input and measured expression of Rspo2 (IDT Assay ID: Mm.PT.58.31856570), Oxr1 (set 1: Mm.PT.58.6081589; set 2: Mm.PT.58.11085074), and Angpt1 (Mm.PT.58.9340436) normalized to the housekeeping gene Rps20 (ABI Assay ID: Mm02342828_g1).
To estimate broad-sense heritability (H2), we employed a Bayesian linear mixed model approach as implemented in R/INLA (24).
We used a variant of the Haley-Knott regression termed regression on probabilities (ROP) to identify loci associated with protein response, as implemented in R/miqtl (25). First, we fit the null model,
| (1) |
| (2) |
We utilized a design in which mice within a given CC strain were exposed to FA or ozone (O3) in sex- and litter-matched pairs or trios. We observed an induction of inflammatory cell recruitment, predominantly defined by neutrophilia that was highly variable across strains after O3 exposure (Figures 1 and E3–E5). We detected significant main effects of strain and exposure on percent and number of BAL neutrophils (P < 10−10 for all), with no significant effect of sex (Figure E4). Additionally, we observed a highly significant strain-by-exposure interactive effect for both phenotypes (P < 10−16). Using data from O3-exposed animals only (because most strains have no neutrophils after FA exposure), we estimated the broad-sense heritability (H2) to be 0.47 (95% CI, 0.36–0.60) and 0.52 (95% CI, 0.23–0.78) for percentage and total number of neutrophils in BAL, respectively. As a quantitative measure of lung injury, we measured total protein concentration in BAL fluid (Figure 2A). This is a nonspecific marker of serum contents (primarily serum albumin [28]) that leak into the airspace upon damage to the alveolar epithelium and increased permeability of the underlying capillary beds. Previously, we documented a positive correlation between serum albumin and total protein levels across several CC strains (29). We observed evidence of modest strain effects in FA-exposed animals and clear O3 exposure effect across strains. For FA- and O3-exposed animals, we estimated H2 to be 0.26 (95% CI, 0.16–0.40) and 0.53 (95% CI, 0.41–0.64), respectively. Not only were strain and exposure effects evident in the protein concentration measurements (P < 10−16), but also interactive effects between strain and exposure (P < 10−10). For instance, while CC080 and CC071 (11th and 12th strains from the left in Figure 2A) had similar, low levels of protein in FA-exposed mice, their post-O3 exposure protein levels were highly divergent: CC080 with a modest increase compared with CC071, which had the highest O3-induced protein levels across all strains, Moreover, higher baseline levels of protein were not necessarily correlated with higher post-O3 exposure levels: CC040 and CC058 (first two strains on the right) had high levels of protein in FA-exposed animals with only slight increases after O3 exposure. To capture these patterns of response and examine gene-by-O3 exposure effects, we defined the total protein response as the difference in total protein concentration between the O3-exposed animal(s) and the FA-exposed animal for a given pair or trio within a strain. The range of total protein response values were nearly normally distributed, indicative of a polygenic, complex trait (Figure 2B), and H2 for this response trait was estimated at 0.39 (95% CI, 0.26–0.54).
Figure 1. Variation in O3-induced neutrophilia across 56 Collaborative Cross (CC) strains. (A) Percent and (B) total number of neutrophils in bronchoalveolar lavage fluid were measured in mice after filtered-air (FA; open circles) or 2 ppm O3 exposure (filled squares). Data are represented as strain means and standard error. The same data plotted by sex are visualized in Figure E4 (n = 4/FA group and 6/O3 group, on average).
[More] [Minimize]
Figure 2. Total protein concentration varies by CC strain and after O3 exposure, and within-pair protein concentration difference is a complex trait. (A) Total protein concentration is modestly influenced by CC strain in control, FA-exposed animals (open circles) and highly variable after O3 exposure (closed squares), with indications of strain-by-exposure interactions (n = 4/FA group and 6/O3 group per strain, on average). (B) To examine strain-by-exposure interactions, we defined a protein response phenotype as the difference in total protein concentration between O3-exposed mice and FA-exposed mice within a matched pair or trio. The distribution of this phenotype is roughly normally distributed, which suggests that it is a complex trait influenced by multiple genetic loci (n = 2 pairs and 2 trios per strain, on average).
[More] [Minimize]To identify genomic regions associated with variation in O3-induced neutrophilia and total protein response, we performed QTL mapping using a haplotype-based regression procedure accounting for the effects of sex and batch. While we did not detect significant loci associated with neutrophilia (Figures E6A and E6B), two QTL were associated with total protein response: one locus each on Chr 15 (Oipq1, ozone-induced protein response QTL 1) and 10 (Oipq2), the former surpassing the genome-wide significance threshold of α = 0.05 and the latter nearly surpassing the more permissive α = 0.1 threshold (Figure 3A). These loci explained roughly 27 and 13% of variation in the phenotype, respectively.
Figure 3. Quantitative trait locus (QTL) mapping uncovers a locus on Chromosome 15 significantly associated with O3-induced protein response. (A) QTL scan highlighting regions where variation in total protein concentration difference (i.e., protein response) is associated with haplotype. The red dashed line indicates the 95% significance threshold (red), and one locus on Chromosome (Chr) 15 surpasses this threshold (marker: JAX00061625_to_UNC25526264, 47.12 Mb). A second locus, on Chromosome 10, surpasses the 90% significance threshold (blue). (B) An 80% confidence interval for the locus on Chr 15 spans ∼15 Mb (40.17–54.88 Mb), indicated by the gray shading. (C) Haplotype dosage intensities at the Chr 15 peak marker for strain pairs, ordered by increasing protein response along the x-axis. Darker ticks indicate higher probabilities of a given haplotype at the peak marker. Strain matches with A/J or WSB/EiJ haplotypes are associated with low protein response, while those with C57BL/6J or CAST/EiJ are associated with high protein response. (D) Haplotype effects and confidence intervals at the peak marker within the Chr 15 QTL.
[More] [Minimize]Oipq1 mapped to a region on Chr 15 extending ∼15 Mb (80% CI, 40.17–54.88 Mb) which contained 31 protein-coding genes (Figure 3B). We initially inspected the founder haplotype effects pattern to identify whether there were functionally distinct haplotypes that could be used to partition the interval and prioritize regions of interest. Upon examining founder haplotype probabilities at the peak marker within Oipq1 (JAX00061625_to_UNC25526264: 47.12 Mb), we saw that CC strain pairs with C57BL/6J or CAST/EiJ founder haplotypes had high total protein response to O3, while strain pairs with A/J or WSB/EiJ haplotypes had low total protein response (Figure 3C). In accordance with these findings, haplotype substitution effects modeled using Diploffect (26) also showed a strong positive effect of the C57BL/6J and CAST/EiJ haplotypes on total protein response (Figure 3D). It should be noted that estimated allele effects for CAST/EiJ are less certain than for C57BL/6J due to lower representation of and less confident haplotype inference for this founder haplotype at the peak locus (Figure 3C). Interestingly, when visualizing the subspecies origin of the CC founder haplotypes within the QTL interval using the Mouse Phylogeny Viewer (30), we discovered an ∼2 Mb region of Mus musculus domesticus intersubspecific introgression into the genome of CAST/EiJ (which is of M. m. castaneus origin) from ∼43–45 Mb (Figure E7). This suggests that CAST/EiJ and C57BL/6J may share a haplotype within this region of Mus musculus domesticus origin. We then used the statistical approach TIMBR (27) to infer the allelic series (i.e., the number of functional alleles into which the CC founder haplotypes group). The results indicated a high likelihood of two functional alleles at this QTL, with greatest weight for a grouping of C57BL/6J and CAST/EiJ founder haplotypes versus the rest (Figure E8; Tables E3 and E4). We note that, as we observed with Diploffect, there was a similar level of uncertainty for CAST/EiJ haplotype effects (Figure E8). Hence, while there is some evidence to suggest that C57BL/6J and CAST/EiJ founder haplotypes are functionally distinct from other CC founder haplotypes, this haplotype grouping should be interpreted with caution. Thus, in subsequent analyses, we focused on variants with strain distribution patterns for which C57BL/6J was unique or where C57BL/6J and CAST/EiJ shared a common variant.
Oipq2 spanned an ∼19 Mb region (80% CI, 24.58–43.61 Mb) encompassing 90 protein-coding genes on Chr 10 (Figure E9A). We note that this locus overlapped a QTL we identified when using BAL protein data from O3-exposed mice only (Figures E6C and E6D), which surpassed the 95% significance threshold. Thus, accounting for protein levels in FA-exposed mice reduced the statistical significance of Oipq2. We examined the founder haplotype probabilities at the peak marker (UNC17621935: 26.204 Mb), which were sorted into a less defined pattern than at Oipq1. Strain pairs with WSB/EiJ haplotype at this marker had high total protein responses on average, but all other haplotypes were spread evenly across the phenotypic spectrum (Figure E9B). Intriguingly, within much of this region (including at the peak marker) in the CC strains surveyed, there is essentially no or low-confidence representation of the PWK/PhJ haplotype, and low representation of the other two wild-derived strain haplotypes (CAST/EiJ and WSB/EiJ) (18). Previous work has established that all three wild-derived founder strains have reduced genome-wide contributions in extant CC strains (18), and selection against the PWK/PhJ haplotype may have occurred at this locus to maintain reproductive compatibility over the course of inbreeding (31). Because Oipq1 surpassed the genome-wide significance threshold and had more clearly defined haplotype effects, we prioritized this locus for further analysis.
To further rank candidate genes and variants within the QTL region, we performed merge analysis (32) with modifications to accommodate multi-allelic variants, as described previously (33). In brief, this approach moves from association at the level of haplotypes to association at the level of individual variants (both single nucleotides and small insertions/deletions). By using sequence information from the Inbred Strain Variant Database (34) for each of the CC founders and CC strains that were used for QTL mapping, variants can be identified that are distributed among the CC strains in concordance with the haplotype effects pattern. Founder strain haplotypes are “merged” into 2–7 groups in accordance with how the variants are distributed (i.e., the strain distribution profile). Variants in the merged model that explain trait variation equally well or better than the full haplotype model (but with fewer parameters) can be considered candidate quantitative trait variants. In the Oipq1 region, 995 variants were identified using this method, with a −log10(P value) > 4 (roughly the cut-off used for QTL mapping; Figure 4A). Variants identified using merge analysis represented a variety of strain distribution patterns (SDPs), with two SDPs more common than the others: C57BL/6J alone or C57BL/6J and CAST/EiJ discordant from all other strains (Figure 4B), as expected based on the results described above. These variants were predicted to have several consequences on gene/protein function, though most were present within introns or other non-coding regions of the QTL (Figure 4C). Overall, this subset of variants was located within or near 21 genes (13 protein-coding, 8 predicted), most of which were concentrated within the region between ∼40.9 and 44.6 Mb containing 16 genes (10 protein-coding) (Figure 4D; Table E5).
Figure 4. Candidate variant (merge) analysis pinpoints variants and narrows the list of genes of interest in the Chr 15 QTL. (A) Merge analysis identified 995 variants whose strain distribution pattern (SDP) explains the variation in protein response with a more parsimonious fit than the 8-haplotype model used for QTL mapping (indicated by the red line). (B) Listed in the left margin of the lower tracks are the SDP for each variant tested by merge analysis in A-H founder strain order, where A=A/J, B=C57BL/6J, C=129S1SvImJ, D=NOD/ShiLtJ, E=NZO/H1LtJ, F=CAST/EiJ, G=PWK/PhJ, and H=WSB/EiJ. For example, the SDP of a biallelic variant is comprised by 0s and 1s. Thus, the first track is a biallelic variant where all seven founder strains share the same allele (0) and C57BL/6J has the alternate allele (1). (C) Corresponding consequences for each of the variants present in (A) and (B), where color-coding is consistent across both panels. Note that variants may have more than one consequence, and thus may appear across multiple tracks. (D) The 16 genes present within the genomic interval with the highest density of variants (∼41–45 Mb). Of these genes, 10 are protein-coding.
[More] [Minimize]We used a weight-of-evidence approach to further prioritize candidate genes within the interval, using the following four criteria (1): presence of variants that are concordant with the haplotype effects pattern, with an emphasis on C57BL/6J with or without CAST/EiJ discordant from all other founder strains (2); presence of coding variants that lead to amino acid alterations (3); evidence that a gene is expressed the lungs (from our previously unpublished and published data (23), or publicly available datasets); and (4) biological relevance, as judged by a gene’s known function and prior description in the literature (Table 1). Two genes within the interval met all four criteria: Angpt1 (angiopoietin-1) and Oxr1 (oxidation resistance 1). A third candidate (Rspo2 [R-spondin] 2) met three of the criteria used, but prior studies provided some additional biological plausibility.
| Gene Symbol* | Gene Name | # of Variants† | Variants Consistent with SDP of Haplotype Effects‡ | Expressed in Lungs | Relevant Biological Function | Variants Alter Amino Acid Sequence |
|---|---|---|---|---|---|---|
| Abra | Actin-binding Rho activating protein | 27 | + | + | − | − |
| Angpt1 | Angiopoietin 1 | 308 | + | + | + | + |
| Eif3e | Eukaryotic translation initiation factor 3, subunit E | 8 | + | + | − | − |
| Emc2 | ER membrane protein complex subunit 2 | 38 | + | + | − | − |
| Ext1 | Exostosin glycosyltransferase 1 | 1 | − | + | + | − |
| Oxr1 | Oxidation resistance 1 | 503 | + | + | + | + |
| Pkhd1l1 | Polycystic kidney and hepatic disease 1-like 1 | 1 | − | + | − | − |
| Rspo2 | R-spondin 2 | 7 | + | + | + | − |
| Samd12 | Sterile α motif domain containing 12 | 1 | − | + | − | − |
| Taf2 | TATA-box binding protein associated factor 2 | 1 | − | + | − | − |
| Tmem74 | Transmembrane protein 74 | 2 | + | − | − | − |
| Trhr | Thyrotropin releasing hormone receptor | 2 | + | − | − | − |
| Zfpm2 | Zinc finger protein, multitype 2 | 1 | + | + | + | − |
Using information from Ensembl, UniProt, and multiple variant consequence prediction tools, we inspected the three missense variants within Oxr1 (rs50179186, rs31574788) and Angpt1 (rs32511504) to determine whether any had putative consequences on protein structure and/or function (Tables 2 and E6). All three variants had SDPs where C57BL/6J was distinct from the other founder strains. Both variants within Oxr1 were present within a predicted disordered region, while the variant within Angpt1 was in a linker region, between coiled-coiled and fibrinogen domains. This analysis was largely inconclusive for two of the three variants (rs50179186 in Oxr1 and rs32511504 in Angpt1), as the results across the in silico prediction tools were conflicting. The second variant in Oxr1 (rs31574788) appeared to have no predicted effect on protein function across all four tools.
| Gene | Build 38 position | SDP* | rsID | AA length | AA shift | SIFT Conseq. | PolyPhen-2 Conseq. | PROVEAN Conseq. | SNAP2 Conseq. |
|---|---|---|---|---|---|---|---|---|---|
| Oxr1† | 41819897 | 01000000 | rs50179186 | 866 | Ser308Ala | Deleterious (0.04) | Benign (0.045/0.034) | Neutral (−1.542) | Neutral (−24) |
| Oxr1† | 41820278 | 01000000 | rs31574788 | 866 | Thr435Pro | Tolerated (0.49) | Benign (0.004/0.016) | Neutral (−1.05) | Neutral (−17) |
| Angpt1 | 42496251 | 01000000 | rs32511504 | 498 | Ile262Val | Tolerated (1) | Benign (0) | Neutral (−0.137) | Effect (25) |
In an attempt to further evaluate the three candidate genes, we measured gene expression using lung tissue from 19 strains sampled across all eight CC haplotypes and both exposure groups, with twelve representative strains for the six haplotypes with low protein response and seven representative strains for the high protein response group (Figure E10). In these samples, we measured highest expression levels for Angpt1, followed closely by Oxr1 with lowest levels for Rspo2. Both Oxr1 probe sets detect Oxr1 transcript variants 2 and 4 (NM_001130163, NM_001130165) and probe set 2 detects additional transcript variants 1, 3, and 5 (NM_130885, NM_001130164, NM_001130166). A previous study indicated that transcript variant 5 (referred to as Oxr1a) is only detected in the murine brain, while the remaining variants are detected throughout the body (35). To estimate main effects of haplotype grouping (C57BL/6J and CAST/EiJ versus remaining 6 haplotypes), O3 exposure, and haplotype-by-exposure interactive effects on gene expression, we used linear regression. Though none of the three candidate genes displayed significant exposure or haplotype-by-exposure effects, we detected a statistically significant, albeit modest, effect of haplotype on Oxr1 expression (probe set 1 P < 0.005, probe set 2 P < 0.05; Figures E10C and E10D; Table E7).
Finally, we examined protein expression of two of the three candidate genes (ANGPT1 and OXR1; RSPO2 was not evaluated due to antibody performance issues) in the lungs of control and O3-exposed C57BL/6J mice (Figures 5 and E11). Immunohistochemical staining for ANGPT1 expression in lungs from filtered air-exposed mice exposed localized to the cytoplasm of bronchiolar epithelial cells and less prominently in type 2 alveolar epithelial cells (Figures 5A and 5B). In the lungs of mice exposed to O3, additional staining was detected in focal cytoplasmic areas in alveolar outpockets lining proximal alveolar ducts (Figures 5C and 5D). The signal we detected in alveolar ducts appeared to derive from epithelial cells, endothelial cells lining septal capillaries, or both. We note also that ANGPT1 expression was also detected in the large airways of both control and O3-exposed mice (Figure E11). In concordance with our qPCR data, we observed lower protein levels for OXR1 compared with ANGPT1 in both the distal (Figures 5E–5H) and proximal airways (Figure E11). OXR1 expression in filtered air-control animals was most conspicuous in the nuclei of bronchiolar epithelial cells with modest staining in the cytoplasm of these cells (Figures 5E and 5F). With exposure to ozone, OXR1 appeared in the nuclei of cells of alveolar outpockets lining proximal alveolar ducts (Figures 5G and 5H). We could not determine if these nuclei were epithelial or endothelial, or both. Further, we detected OXR1 protein expression in the smooth muscle in the upper airways (Figure E11). In total, these findings confirm expression of two of the three candidate genes in cellular populations targeted by O3 exposure and/or involved in repair of alveoli after injury, providing plausibility for their involvement in response to O3-induced acute lung injury.
Figure 5. Protein products of two candidate genes within Oipq1, ANGPT1 and OXR1, are expressed in bronchiolar and alveolar epithelial cells of C57BL/6J mice, with variable intensity and cell-type localization. Representative light photomicrographs of lung tissue in centriacinar regions immunohistochemically stained for ANGPT1 (A–D) or OXR1 (E–H), and counterstained with hematoxylin. (A, C, E, and G) are low magnifications and (B, D, F, and H) are at high magnifications of the lung tissue from a control mouse exposed to filtered air (A, B, E, and F) and an ozone-exposed mouse (C, D, G, and H). Immunohistochemical positive staining (reddish brown chromagen) of ANGPT1 in lung tissue of the control mouse (A and B) was restricted to cytoplasm of bronchiolar epithelial cells and type 2 alveolar epithelial cells. In the lungs of mice exposed to 2 ppm ozone (C and D), ANGPT1 positive staining was present in focal cytoplasmic areas in alveolar outpockets lining proximal alveolar ducts (solid arrows) and in alveolar epithelial type 2 cells (stippled arrow). Immunohistochemical positive staining of OXR1 in lung tissue of the control mouse (E and F) was restricted to nuclei in bronchiolar epithelial cells with modest cytoplasmic staining. In the lungs of mice exposed to 2 ppm ozone (G and H), OXR1 nuclear staining was also present in epithelial cells of alveolar outpockets lining proximal alveolar ducts (solid arrows). Results of immunohistochemistry in large airways are shown in Figure E11. a = alveolus; AD = alveolar duct; e = bronchiolar epithelium; TB = terminal bronchiole.
[More] [Minimize]Here, we present discovery of two loci associated with ozone (O3)-induced lung injury, Oipq1 and Oipq2, located on Chr 15 and 10, respectively. Notably, the region encompassed by Oipq1 has been previously implicated in two studies aiming to identify genetic variants associated with lung injury. One study utilized a C57BL/6J:129X1/SvJ F2 population to map QTL driving fatality due to hyperoxia (36). In that study, the C57BL/6J allele at the QTL was associated with higher lung injury, consistent with the results of our study. Similarly, an overlapping QTL for pulmonary hemorrhage was identified in a study utilizing an F2 intercross of two CC strains (CC003/Unc x CC053/Unc) to identify genetic determinants of pulmonary responses to severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1; 37). Though this QTL was not the focus of that study, the authors did observe a positive association between the CAST/EiJ allele and pulmonary hemorrhage and the inverse relationship with the PWK/PhJ allele, matching the haplotype effects observed in our study. Therefore, there is supporting evidence that this locus mediates responses to multiple stimuli that induce lung injury. We note, however, that previous studies have posited that C57BL/6J mice are resistant to O3-induced lung injury based on mortality rates due to prolonged O3 exposure (10) and only modest increases in lung protein following O3 exposure; however, direct comparisons between the present study and these previous reports cannot be drawn because CC strains have mosaic genomes with representation from all eight founder strains. Thus, other C57BL/6J-derived genomic regions that contribute to resistance may exist, but these were not identified in this study.
Within Oipq1, we nominated three candidate genes: Rspo2, Angpt1 (angiopoietin-1), and Oxr1 (oxidation resistance 1). Conclusions about the effects of missense variants within Angpt1 and Oxr1 are difficult to make because the results from variant prediction tools were conflicting and/or largely suggested that the missense variants had neutral effects on protein function. While we used a suite of approaches that each incorporate unique features for prediction including sequence homology, physico-chemical properties, and predicted secondary structure, even state-of-the-art methods fail to fully recapitulate functional characterization through experimental strategies (e.g., deep mutational scanning or multiplex assays of variant effect), which remain the gold standard. Thus, direct experimental classification is needed to determine whether these variants have effects on protein function. We have also demonstrated that Oxr1 expression moderately varies by haplotype group (i.e., has one or more eQTL). Thus, it is possible that variation in gene expression underlies the Oipq1 QTL. Future studies will be required to resolve these outstanding questions.
The first candidate gene, Angpt1, encodes ANGPT1 (angiopoietin-1), which is a secreted ligand that binds to and activates the Tie2 receptor tyrosine kinase (encoded by Tek) to promote endothelial barrier function and vascular growth (38). It has also been suggested that ANGPT1 activity can improve and reinforce leaky or otherwise poorly functioning vessels (39). Previous work has established that ANGPT1 is largely produced by vascular support cells and platelets, and its activity can be antagonized by ANGPT2 (40). Disrupted balance of ANGPT2/ANGPT1 in serum is associated with poor outcomes in a variety of disease states, including acute respiratory distress syndrome, bronchopulmonary dysplasia, and sepsis (41, 42). Genetic association studies have identified variants in ANGPT2 associated with susceptibility to acute lung injury (41). Together, these studies suggest that ANGPT2 is often a pathogenic regulator of acute lung injury and barrier dysfunction, and balance to this system can be restored by supplementation with ANGPT1. Thus, one hypothesis arising from our study is that in strains with haplotypes associated with more severe O3-induced lung injury (i.e., C57BL/6J and potentially CAST/EiJ), there may be variants in or near Angpt1 that alter its activity and/or function, thereby disrupting its ability to negatively regulate ANGPT2/TEK signaling. In vitro and in vivo functional validation studies in models of O3 exposure or other types of acute lung injury will be necessary to evaluate this hypothesis. We detected increased expression of ANGPT1 due to O3 in alveolar out-pockets lining proximal alveolar ducts, which is the site of greatest injury as a function of increased O3 deposition at this location (43), adding an additional layer of biological support for this gene/protein.
Our second candidate gene, Oxr1 (oxidation resistance 1), has less evidence tying its functions to acute lung injury and epithelial barrier integrity, though these remain to be examined. This protein, while aptly named for a role in response to oxidant gases, has largely been implicated in maintaining genome integrity and cell survival in the face of stressors that cause either oxidative stress-dependent or -independent DNA damage (e.g., reactive oxygen species, radiation, alkylating agents). This gene was first discovered in an E. coli screen for human genes involved in repairing or preventing oxidative DNA damage (44), and later discovered to function largely in the mitochondria (45). Oxr1 has since been studied largely in the context of neurological diseases including amyotrophic lateral sclerosis and other neurodegenerative conditions (46). Only a few studies have mentioned this gene in the context of lung disease, including one examining vanadium pentoxide (V2O5)-induced occupational bronchitis where OXR1 expression was induced in human lung fibroblasts after exposure to V2O5 (47). It is worth mentioning that mice express multiple isoforms of Oxr1 whose tissue specificity was recently characterized (35): many tissues expressed the shortest version (Oxr1D) and Oxr1B1-4, while the longest (Oxr1A) was restricted to the brain. The cited study went on to characterize the functions of the OXR1A isoform, demonstrating that its TLDc domain (present in all OXR1 isoforms) facilitates interactions with a variety of proteins including the PRMT5 methyltransferase, thus representing one pathway by which this gene alters cellular function in response to stress. Oxr1 was the only candidate gene whose expression differed by chromosome 15 haplotype, and thus the genetic evidence is strongest for this gene. We also detected O3-induced OXR1 expression in nuclei of cells of alveolar out-pockets lining proximal alveolar ducts, consistent with a purported role in response to oxidative stress-dependent DNA damage. However, further studies will be required to better understand this protein’s role(s) in oxidative stress responses in the lungs and assess its relationship to the protein response phenotype measured here.
One of the original candidate genes at Oipq1, Rspo2, is a member of the R-spondin protein family, a class of secreted ligands known to regulate Wnt signaling, tissue regeneration and organization in various regions of the body. Specifically, RSPO2 is required for lung, limb, and craniofacial development (48), and Rspo2-deficient mice are born with various skeletal defects and die immediately upon birth due to respiratory failure (49). Recessive mutations in RSPO2 cause tetra-amelia syndrome-2, a human syndrome characterized by partial or complete absence of limbs along with incomplete lung development (50). Jackson and colleagues recently examined the consequences of Rspo2 conditional deletion in adult mice. They reported that Rspo2 loss caused lung neutrophilia via a disrupted lung endothelial barrier (51). Forward genetic studies also have identified Rspo2 as a key gene that helps orchestrate epithelial differentiation and organization in the colon after Citrobacter rodentium infection (52, 53). Additionally, human RSPO2 was found to be upregulated in epithelial cells and fibroblasts of patients with idiopathic pulmonary fibrosis (54). While there are no known exonic variants within this gene among the CC strains we studied, we noted multiple noncoding variants within the in/near the gene, and its function is plausibly related to features of the pathology caused by O3 exposure. We could not detect protein expression of RSPO2 owing to antibody limitations. Others have found that Rspo2 gene expression is largely restricted to the lung mesenchyme, and at virtually undetectable levels throughout all life stages beyond embryonic development. This is in concordance with our qPCR data, where we observed very low expression in comparison to the other two candidate genes. Thus, while our gene expression data do not provide support for Rspo2 as the causal gene for Oipq1, the phenotypes observed after its conditional deletion in adult mice are compelling and it is upregulated in fibrosis, indicating that further investigation in the context of O3-induced lung injury is warranted.
We did not detect any loci associated with variation in airway neutrophilia, another hallmark O3-induced phenotype, even though others have previously identified QTL for this trait including one that encompasses Tlr4 (11). We did not expect to identify this locus because none of the CC founder strains have the mutant Tlr4Lps-d allele present in C3H/HeJ (which was used to confirm Tlr4’s role in O3 responses) or any similar defective allele. We note, however, that we observed clear strain-dependent neutrophil recruitment in our study, leading to high estimates of heritability (H2 = 0.47 and 0.52 for percentage and total number of neutrophils, respectively). Thus, one potential explanation for the lack of QTL is that the genetic architecture of airway neutrophilia may be more complex than lung injury and involve contributions from many loci with individually small effects. One option to address this would be to incorporate information about phenotypes that lie along the causal chain from O3 exposure to airway neutrophilia (e.g., cytokines), as we expect these intermediate phenotypes to have higher heritabilities and simpler genetic architectures. Additionally, we could more explicitly examine the dynamics of neutrophil recruitment and clearance, two processes that are not immediately separable when examining BAL neutrophil count at a single time point. Alternatively, to achieve greater statistical power for QTL mapping, one could make use of the Diversity Outbred mouse population since a larger number of genetically distinct mice can be examined in that population (55). However, we note here that the protein response QTL we identified here was detected using a “delta” framework (in which the phenotype was the difference between O3- versus FA-exposed mice), which required inbred strains so that baseline FA effects could be accounted for.
In conclusion, we have identified a significant QTL on mouse Chr 15 associated with O3-induced lung injury. Through additional genetic and bioinformatic analyses, we delimited the QTL region and identified three high-priority candidate genes worthy of additional investigation. Our study also demonstrates the utility of the CC genetic reference population for identifying interactions between genetic variants and environmental exposures.
The authors would like to acknowledge the assistance of Daniel Vargas and Jessica Bustamante (logistical support); Courtney Nesline and the UNC Division of Comparative Medicine; Darla Miller, Ginger Shaw, and Dr. Rachel Lynch of the UNC Systems Genetics Core Facility (Collaborative Cross mice); Drs. Greg Keele and Wes Crouse (maintenance of and guidance with using the miqtl and TIMBR R packages, respectively); Drs. Will Valdar and Yanwei Cai (QTL mapping); Amy Porter of the Michigan State University Investigative Histopathology Laboratory for assistance with histology and immunohistochemistry; and two anonymous referees for their insightful suggestions.
| 1. | Last JA, Pinkerton KE, Schelegle ES. 15.20. Ozone and oxidant toxicity. In: McQueen CA, editor. Comprehensive toxicology, 3rd ed. Boston, Amsterdam: Elsevier; 2018. pp. 389–402. |
| 2. | Darrow LA, Klein M, Flanders WD, Mulholland JA, Tolbert PE, Strickland MJ. Air pollution and acute respiratory infections among children 0-4 years of age: an 18-year time-series study. Am J Epidemiol 2014;180:968–977. |
| 3. | Ciencewicki J, Trivedi S, Kleeberger SR. Oxidants and the pathogenesis of lung diseases. J Allergy Clin Immunol 2008;122:456–470. |
| 4. | Thurston GD, Balmes JR, Garcia E, Gilliland FD, Rice MB, Schikowski T, et al. Outdoor air pollution and new-onset airway disease. An Official American Thoracic Society Workshop Report. Ann Am Thorac Soc 2020;17:387–398. |
| 5. | Yang IA, Holz O, Jörres RA, Magnussen H, Barton SJ, Rodríguez S, et al. Association of tumor necrosis factor-alpha polymorphisms and ozone-induced change in lung function. Am J Respir Crit Care Med 2005;171:171–176. |
| 6. | Alexeeff SE, Litonjua AA, Wright RO, Baccarelli A, Suh H, Sparrow D, et al. Ozone exposure, antioxidant genes, and lung function in an elderly cohort: VA normative aging study. Occup Environ Med 2008;65:736–742. |
| 7. | Holz O, Jörres RA, Timm P, Mücke M, Richter K, Koschyk S, et al. Ozone-induced airway inflammatory changes differ between individuals and are reproducible. Am J Respir Crit Care Med 1999;159:776–784. |
| 8. | McDonnell WF III, Horstman DH, Abdul-Salaam S, House DE. Reproducibility of individual responses to ozone exposure. Am Rev Respir Dis 1985;131:36–40. |
| 9. | Alexis NE, Zhou H, Lay JC, Harris B, Hernandez ML, Lu T-S, et al. The glutathione-S-transferase Mu 1 null genotype modulates ozone-induced airway inflammation in human subjects. J Allergy Clin Immunol 2009;124:1222–1228.e5. |
| 10. | Prows DR, Shertzer HG, Daly MJ, Sidman CL, Leikauf GD. Genetic analysis of ozone-induced acute lung injury in sensitive and resistant strains of mice. Nat Genet 1997;17:471–474. |
| 11. | Kleeberger SR, Levitt RC, Zhang LY, Longphre M, Harkema J, Jedlicka A, et al. Linkage analysis of susceptibility to ozone-induced lung inflammation in inbred mice. Nat Genet 1997;17:475–478. |
| 12. | Kleeberger SR, Reddy S, Zhang LY, Jedlicka AE. Genetic susceptibility to ozone-induced lung hyperpermeability: role of toll-like receptor 4. Am J Respir Cell Mol Biol 2000;22:620–627. |
| 13. | Bauer AK, Rondini EA, Hummel KA, Degraff LM, Walker C, Jedlicka AE, et al. Identification of candidate genes downstream of TLR4 signaling after ozone exposure in mice: a role for heat-shock protein 70. Environ Health Perspect 2011;119:1091–1097. |
| 14. | Fakhrzadeh L, Laskin JD, Laskin DL. Deficiency in inducible nitric oxide synthase protects mice from ozone-induced lung inflammation and tissue injury. Am J Respir Cell Mol Biol 2002;26:413–419. |
| 15. | Connor AJ, Laskin JD, Laskin DL. Ozone-induced lung injury and sterile inflammation. Role of toll-like receptor 4. Exp Mol Pathol 2012;92:229–235. |
| 16. | Cho HY, Zhang LY, Kleeberger SR. Ozone-induced lung inflammation and hyperreactivity are mediated via tumor necrosis factor-alpha receptors. Am J Physiol Lung Cell Mol Physiol 2001;280:L537–L546. |
| 17. | Shore SA, Schwartzman IN, Le Blanc B, Murthy GG, Doerschuk CM. Tumor necrosis factor receptor 2 contributes to ozone-induced airway hyperresponsiveness in mice. Am J Respir Crit Care Med 2001;164:602–607. |
| 18. | Srivastava A, Morgan AP, Najarian ML, Sarsani VK, Sigmon JS, Shorter JR, et al. Genomes of the Mouse Collaborative Cross. Genetics 2017;206:537–556. |
| 19. | Kelada SNP, Aylor DL, Peck BCE, Ryan JF, Tavarez U, Buus RJ, et al. Genetic analysis of hematological parameters in incipient lines of the collaborative cross. G3 (Bethesda) 2012;2:157–165. |
| 20. | Kelada SN, Carpenter DE, Aylor DL, Chines P, Rutledge H, Chesler EJ, et al. Integrative genetic analysis of allergic inflammation in the murine lung. Am J Respir Cell Mol Biol 2014;51:436–445. |
| 21. | Donoghue LJ, Livraghi-Butrico A, McFadden KM, Thomas JM, Chen G, Grubb BR, et al. Identification of trans protein QTL for secreted airway mucins in mice and a causal role for Bpifb1. Genetics 2017;207: 801–812. |
| 22. | Ferris MT, Aylor DL, Bottomly D, Whitmore AC, Aicher LD, Bell TA, et al. Modeling host genetic regulation of influenza pathogenesis in the collaborative cross. PLoS Pathog 2013;9:e1003196. |
| 23. | Tovar A, Smith GJ, Thomas JM, Crouse WL, Harkema JR, Kelada SNP. Transcriptional profiling of the murine airway response to acute ozone exposure. Toxicol Sci 2020;173:114–130. |
| 24. | Holand AM, Steinsland I, Martino S, Jensen H. Animal models and integrated nested Laplace approximations. G3 (Bethesda) 2013;3:1241–1251. |
| 25. | Keele GR, Quach BC, Israel JW, Chappell GA, Lewis L, Safi A, et al. Integrative QTL analysis of gene expression and chromatin accessibility identifies multi-tissue patterns of genetic regulation. PLoS Genet 2020;16:e1008537. |
| 26. | Zhang Z, Wang W, Valdar W. Bayesian modeling of haplotype effects in multiparent populations. Genetics 2014;198:139–156. |
| 27. | Crouse WL, Kelada SNP, Valdar W. Inferring the allelic series at QTL in multiparental populations. Genetics 2020;216:957–983. |
| 28. | Koren HS, Devlin RB, Graham DE, Mann R, McGee MP, Horstman DH, et al. Ozone-induced inflammation in the lower airways of human subjects. Am Rev Respir Dis 1989;139:407–415. |
| 29. | Tovar A, Crouse WL, Smith GJ, Thomas JM, Keith BP, McFadden KM, et al. Integrative analysis reveals mouse strain-dependent responses to acute ozone exposure associated with airway macrophage transcriptional activity. Am J Physiol Lung Cell Mol Physiol 2022;322:L33–L49. |
| 30. | Yang H, Wang JR, Didion JP, Buus RJ, Bell TA, Welsh CE, et al. Subspecific origin and haplotype diversity in the laboratory mouse. Nat Genet 2011;43:648–655. |
| 31. | Shorter JR, Odet F, Aylor DL, Pan W, Kao C-Y, Fu C-P, et al. Male infertility is responsible for nearly half of the extinction observed in the mouse collaborative cross. Genetics 2017;206:557–572. |
| 32. | Yalcin B, Flint J, Mott R. Using progenitor strain information to identify quantitative trait nucleotides in outbred mice. Genetics 2005;171:673–681. |
| 33. | Mosedale M, Cai Y, Eaddy JS, Corty RW, Nautiyal M, Watkins PB, et al. Identification of candidate risk factor genes for human Idelalisib toxicity using a Collaborative Cross approach. Toxicol Sci 2019;172:265–278. |
| 34. | Oreper D, Cai Y, Tarantino LM, de Villena FP-M, Valdar W. Inbred strain variant database (ISVdb): a repository for probabilistically informed sequence differences among the collaborative cross strains and their founders. G3 (Bethesda) 2017;7:1623–1630. |
| 35. | Yang M, Lin X, Segers F, Suganthan R, Hildrestrand GA, Rinholm JE, et al. OXR1A, a coactivator of prmt5 regulating histone arginine methylation. Cell Rep 2020;30:4165–4178.e7. |
| 36. | Prows DR, Hafertepen AP, Winterberg AV, Gibbons WJ Jr, Liu C, Nick TG. Genetic analysis of hyperoxic acute lung injury survival in reciprocal intercross mice. Physiol Genomics 2007;30:271–281. |
| 37. | Gralinski LE, Menachery VD, Morgan AP, Totura AL, Beall A, Kocher J, et al. Allelic variation in the Toll-like receptor adaptor protein Ticam2 contributes to SARS-Coronavirus pathogenesis in mice. G3 (Bethesda) 2017;7:1653–1663. |
| 38. | Suri C, Jones PF, Patan S, Bartunkova S, Maisonpierre PC, Davis S, et al. Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 1996;87:1171–1180. |
| 39. | Thurston G, Rudge JS, Ioffe E, Zhou H, Ross L, Croll SD, et al. Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat Med 2000;6:460–463. |
| 40. | Maisonpierre PC, Suri C, Jones PF, Bartunkova S, Wiegand SJ, Radziejewski C, et al. Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 1997;277:55–60. |
| 41. | Meyer NJ, Li M, Feng R, Bradfield J, Gallop R, Bellamy S, et al. ANGPT2 genetic variant is associated with trauma-associated acute lung injury and altered plasma angiopoietin-2 isoform ratio. Am J Respir Crit Care Med 2011;183:1344–1353. |
| 42. | Parikh SM, Mammoto T, Schultz A, Yuan H-T, Christiani D, Karumanchi SA, et al. Excess circulating angiopoietin-2 may contribute to pulmonary vascular leak in sepsis in humans. PLoS Med 2006;3:e46. |
| 43. | Overton JH, Graham RC, Miller FJ. A model of the regional uptake of gaseous pollutants in the lung. II. The sensitivity of ozone uptake in laboratory animal lungs to anatomical and ventilatory parameters. Toxicol Appl Pharmacol 1987;88:418–432. |
| 44. | Volkert MR, Elliott NA, Housman DE. Functional genomics reveals a family of eukaryotic oxidation protection genes. Proc Natl Acad Sci USA 2000;97:14530–14535. |
| 45. | Elliott NA, Volkert MR. Stress induction and mitochondrial localization of Oxr1 proteins in yeast and humans. Mol Cell Biol 2004;24:3180–3187. |
| 46. | Oliver PL, Finelli MJ, Edwards B, Bitoun E, Butts DL, Becker EBE, et al. Oxr1 is essential for protection against oxidative stress-induced neurodegeneration. PLoS Genet 2011;7:e1002338. |
| 47. | Ingram JL, Antao-Menezes A, Turpin EA, Wallace DG, Mangum JB, Pluta LJ, et al. Genomic analysis of human lung fibroblasts exposed to vanadium pentoxide to identify candidate genes for occupational bronchitis. Respir Res 2007;8:34. |
| 48. | Bell SM, Schreiner CM, Wert SE, Mucenski ML, Scott WJ, Whitsett JA. R-spondin 2 is required for normal laryngeal-tracheal, lung and limb morphogenesis. Development 2008;135:1049–1058. |
| 49. | Yamada W, Nagao K, Horikoshi K, Fujikura A, Ikeda E, Inagaki Y, et al. Craniofacial malformation in R-spondin2 knockout mice. Biochem Biophys Res Commun 2009;381:453–458. |
| 50. | Szenker-Ravi E, Altunoglu U, Leushacke M, Bosso-Lefèvre C, Khatoo M, Thi Tran H, et al. RSPO2 inhibition of RNF43 and ZNRF3 governs limb development independently of LGR4/5/6. Nature 2018;557:564–569. |
| 51. | Jackson SR, Costa MFDM, Pastore CF, Zhao G, Weiner AI, Adams S, et al. R-spondin 2 mediates neutrophil egress into the alveolar space through increased lung permeability. BMC Res Notes 2020;13:54. |
| 52. | Diez E, Zhu L, Teatero SA, Paquet M, Roy M-F, Loredo-Osti JC, et al. Identification and characterization of Cri1, a locus controlling mortality during Citrobacter rodentium infection in mice. Genes Immun 2011;12:280–290. |
| 53. | Papapietro O, Teatero S, Thanabalasuriar A, Yuki KE, Diez E, Zhu L, et al. R-spondin 2 signalling mediates susceptibility to fatal infectious diarrhoea. Nat Commun 2013;4:1898. |
| 54. | Munguía-Reyes A, Balderas-Martínez YI, Becerril C, Checa M, Ramírez R, Ortiz B, et al. R-Spondin-2 is upregulated in idiopathic pulmonary fibrosis and affects fibroblast behavior. Am J Respir Cell Mol Biol 2018;59:65–76. |
| 55. | Svenson KL, Gatti DM, Valdar W, Welsh CE, Cheng R, Chesler EJ, et al. High-resolution genetic mapping using the Mouse Diversity outbred population. Genetics 2012;190:437–447. |
Supported by National Institutes of Health Grants ES024965 and ES024965-S1 (S.N.P.K.), a National Institute of Environmental Health Sciences T32 training grant (ES007126-35) (G.J.S and M.B.N.), a Leon and Bertha Golberg Postdoctoral Fellowship from the UNC Curriculum in Toxicology and Environmental Medicine (G.J.S.), and a UNC Dissertation Completion Fellowship (A.T.).
Author contributions: A.T. and S.N.P.K. conceived and designed the experiments; A.T., G.J.S., M.B.N., J.M.T., K.M.M., and J.R.H. generated data and performed quality control analysis; A.T. and J.R.H. analyzed data; A.T., J.R.H., and S.N.P.K. drafted and revised the manuscript.
This article has a data supplement, which is accessible from this issue’s table of contents online at www.atsjournals.org.
Originally Published in Press as DOI: 10.1165/rcmb.2021-0326OC on July 11, 2022
Author disclosures are available with the text of this article at www.atsjournals.org.
