Am J Physiol Lung Cell Mol Physiol 314(5): L846-L859

PMC: PMC6008126

PMID: 29345197

Neonatal hyperoxia depletes pulmonary vein cardiomyocytes in adult mice via mitochondrial oxidation

INTRODUCTION

Approximately 10% of births occur before 37 wk gestation and are, thus, considered to be preterm. Fortunately, the therapeutic use of exogenous surfactant, milder ventilation strategies, and steroids has significantly improved the survival of these fragile infants. Despite these advances, reduced lung function, airway wheezing, and high rates of hospitalization following respiratory viral infections are seen in people born preterm at birth (13, 14, 43, 46, 51). Persistent pulmonary disease seen in people born preterm has been known for many years because it often manifests early in childhood. However, now that these individuals are reaching adulthood, there is growing concern that they are also predisposed to developing cardiovascular disease (5, 16). Systemic microvascular rarefaction and increased levels of circulating antiangiogenic factors, like endoglin, have been observed in young adults born preterm (31). Abnormal positioning of the left ventricular and reduced aortic size have also been found in those before 30 wk of gestation (15, 17, 27, 30). Preterm infants are often born extremely small, and growth restriction has been linked to poor cardiovascular health later in life (4). However, the cardiovascular disease seen in adolescents born preterm is not associated with intrauterine growth restriction (48). Although changes in the expression of susceptibility and resilience genes (1, 32, 50) or maternal health (44) may enhance the incidence of cardiovascular disease, a growing body of evidence suggests that early life exposure to excess oxygen (hyperoxia) used to treat preterm infants is a major contributor of bronchopulmonary dysplasia and, hence, has long-term consequences on health (25, 49).

Large and small animal models have elucidated how neonatal hyperoxia can drive the pathogenesis of neonatal chronic lung disease typically seen in preterm infants suffering from bronchopulmonary dysplasia (6, 7, 11, 37, 56). The cumulative evidence suggests elevated levels of oxygen increases the production of mitochondrial and nonmitochondrial sources of reactive oxygen species, causing inflammatory injury that disrupts postnatal lung development, resulting in hyperreactive airways with simplified alveoli and thickened septae (23, 33). While it is unclear whether these changes persist over time, emerging evidence in animal models suggests neonatal hyperoxia can also cause cardiovascular disease later in life. We previously showed that exposing newborn mice to 100% oxygen between postnatal days (PND) 0–4 promotes pulmonary hypertension, as defined pathologically by the dilation of large blood vessels, capillary rarefaction, right ventricular hypertrophy, and 50% mortality at 1 yr of age (55). Interestingly, capillary rarefaction and mortality was not observed at 8 wk of age, suggesting cardiovascular disease develops in these mice long after exposure to hyperoxia. Other investigators have reported similar findings, including increased blood pressure, vascular dysfunction, and microvessel rarefaction in adult (7–15 wk) rats exposed to 80% oxygen from PND 3–10 (24, 57). Increased collagen was also detected along the vascular wall of the rat aorta. Likewise, right ventricular hypertrophy has been seen in 14-day-old mice exposed to 75% oxygen from birth to PND 4 (12). Right ventricular dysfunction and mitochondrial dysregulation have also been observed in 1-yr-old rats exposed to 85% oxygen for the first 2 wk of life (22). Pulmonary hypertension and right ventricular dysfunction have also been observed in adult mice exposed to 70% oxygen for the first 2 wk of life (35). Whether contractile dysfunction is directly caused by neonatal hyperoxia or indirectly as a consequence of hypertension and increased vascular resistance is not clear.

To understand how neonatal hyperoxia drives cardiovascular disease later in life, we profiled transcriptional changes in lungs of adult mice before the onset of capillary rarefaction. Here, we show that neonatal hyperoxia suppresses expression of cardiomyocyte genes, which represent a depletion of cardiomyocytes wrapping the pulmonary vein within the lung parenchyma. Because these cells form a contractile sleeve that helps pump blood back to the heart (34), their early loss may contribute to the development of pulmonary hypertension seen as these mice age.

MATERIALS AND METHODS

Mice, oxygen exposures, and drug administration.

All mice were purchased from The Jackson Laboratories and were maintained as inbred colonies on a C57BL/6J background. Newborn mice were exposed to room air (21% oxygen) or hyperoxia (40, 60, 80, or 100% oxygen) between PND 0 and PND 4 (8). Most studies used 100% oxygen unless a different dose is specifically stated. Dams were cycled between litters exposed to room air and hyperoxia every 24 h to ensure their lungs were not injured by hyperoxia. Newborn mice exposed to hyperoxia were returned to room air on PND 4. Bitransgenic Myh6^{; R26}^{mice were injected intraperitoneally with tamoxifen (200 μg/mouse) or corn oil vehicle on the morning of birth and then exposed to room air or hyperoxia through PND 4. Some mice were injected intraperitoneally with mitoTEMPO (0.7 μg/g) or vehicle on PND 0, PND 1, and PND 2. All mice were housed in microisolator cages in a specified pathogen-free environment, according to a protocol approved by the University Committee on Animal Resources at the University of Rochester, and were provided food and water ad libitum.}

Affymetrix arrays and analysis.

Total RNA was isolated from male 8-wk-old mice exposed to room air or 100% oxygen between PND 0 and PND 4 using TRIzol (ThermoFisher Scientific, Waltham, MA). RNA integrity was validated by the quality of ribosomal RNA using an Agilent Bioanalyzer (Agilent Technologies, Santa Clara, CA). RNA was converted to cDNA, biotinylated with Ovation kits from NuGEN (San Carlos, CA), and hybridized overnight to the Affymetrix mouse genome 430 2.0 array (Affymetrix, Santa Clara, CA). Each array was probed with RNA isolated from an individual lung. Arrays were stained with streptavidin-phycoerythrin, as recommended by Affymetrix. The arrays were then scanned for phycoerythrin fluorescence, and the spot intensities were normalized across arrays with the RMA method in the “oligo” package (10). Differential expression of genes was assessed by limma (42). Genes significant at a false discovery rate (FDR) of 10% are reported in this study. For comparison to genes expressed by atrial cells, experiment E-GEOD-46496 (52) was downloaded from ArrayExpress. Wild-type atria samples were extracted, jointly RMA normalized with the lung data, and reported in this study. The complete array data from the mouse lung was deposited in ArrayExpress under accession number E-MTAB-6074.

Quantitative RT-PCR.

Total RNA was isolated using TRIzol reagent (ThermoFisher Scientific) and treated with DNase I using a TURBO DNA-free kits (Life Technologies, Carlsbad, CA) to remove potential contaminating genomic DNA. The RNA was reverse transcribed to cDNA using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). The cDNA was then amplified with SYBR Green I dye on CFX96 Touch and CFX384 Touch real-time PCR detection system (Bio-Rad, Hercules, CA). PCR products were amplified with sequence-specific primers or 18S rRNA used to normalize equal loading of the template cDNAs (Table 1).

Table 1.

RT-PCR oligonucleotide sequences

Genes	Primers	GenBank Accession Number	Product Size, bp
Tnnt2	F: 5′-`GTGTGCAGTCCCTGTTCAGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001130174","term_id":"391738224","term_text":"NM_001130174"}}NM_001130174.2	130
	R: 5′-`ACCCTCAGGCTCAGGTTCA`-3′
Mhy6	F: 5′-`CTCTGGATTGGTCTCCCAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001164171","term_id":"255918224","term_text":"NM_001164171"}}NM_001164171.1	150
	R: 5′-`GTCATTCTGTCACTCAAACTCTGG`-3′
Mhy7	F: 5′-`GCCTCAGCAGAGGAGTACAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_080728","term_id":"1371981807","term_text":"NM_080728"}}NM_080728.2	86
	R: 5′-`ATGGCTGAGCCTTGGATTCTC`-3′
Mhl3	F: 5′-`AGATCTCCATAGCCTTTGGCCT`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009701","term_id":"117940061","term_text":"NM_009701"}}NM_009701.4	124
	R: 5′-`AGCAGAGAGATCTGGTTGCCTA`-3′
Tnni3	F: 5′-`CGAAGCAGGAGATGGAACGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009406","term_id":"1134612221","term_text":"NM_009406"}}NM_009406.3	129
	R: 5′-`GCTGTCGGCATAAGTCCTGAA`-3′
Mybpc3	F: 5′-`TCCAGATGAGTGGCAGCAAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_008653","term_id":"134031946","term_text":"NM_008653"}}NM_008653.2	178
	R: 5′-`TCTTCCAGGGACCGAGTGAT`-3′
a-SMA	F: 5′-`CCAAATCATTCCTGCCCAAAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_007392","term_id":"440309867","term_text":"NM_007392"}}NM_007392	74
	R: 5′-`TAGGCCAGGGCTACAAGTTAAG`-3′
18 s	F: 5′-`CGGCTACCACATCCAAGGAA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NR_003278","term_id":"374088232","term_text":"NR_003278"}}NR_003278.1	187
	R: 5′-`GCTGGAATTACCGCGGCT`-3′

F, forward primer; R, reverse primer.

Immunohistochemstry.

Lung tissues were inflation fixed overnight in 10% neutral buffered formalin, embedded in paraffin, sectioned, and stained as described previously (53, 54). Sections were stained with antibodies against TNNT2 (ThermoFisher Scientific), enhanced green fluorescent protein (EGFP), and Ki67 (Abcam, Cambridge, MA). Sections were incubated with fluorescently labeled secondary antibody (Jackson Immune Research, West Grove, PA) and stained with 4′, 6-diamidino-2-phenylindole (DAPI) (Life Technologies, Carlsbad, CA). Sections were examined and photographed using a Nikon E800 fluorescence microscope (Microvideo Instruments, Avon, MA) and a SPOT-RT digital camera (Diagnostic Instruments, Sterling Heights, MI).

Statistical analysis.

Quantitative RT-PCR data were evaluated using JMP12 software (SAS Institute, Cary, NC) and were graphed as means ± SE of a single experiment that was repeated 3–5 times to confirm consistency. A two-way ANOVA was used to determine overall significance, followed by Tukey-Kramer honestly significant difference tests. Affymetrix expression data were processed in R 3.4, Bioconductor version 3.5.

Mice, oxygen exposures, and drug administration.

Affymetrix arrays and analysis.

Quantitative RT-PCR.

Table 1.

RT-PCR oligonucleotide sequences

Genes	Primers	GenBank Accession Number	Product Size, bp
Tnnt2	F: 5′-`GTGTGCAGTCCCTGTTCAGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001130174","term_id":"391738224","term_text":"NM_001130174"}}NM_001130174.2	130
	R: 5′-`ACCCTCAGGCTCAGGTTCA`-3′
Mhy6	F: 5′-`CTCTGGATTGGTCTCCCAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001164171","term_id":"255918224","term_text":"NM_001164171"}}NM_001164171.1	150
	R: 5′-`GTCATTCTGTCACTCAAACTCTGG`-3′
Mhy7	F: 5′-`GCCTCAGCAGAGGAGTACAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_080728","term_id":"1371981807","term_text":"NM_080728"}}NM_080728.2	86
	R: 5′-`ATGGCTGAGCCTTGGATTCTC`-3′
Mhl3	F: 5′-`AGATCTCCATAGCCTTTGGCCT`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009701","term_id":"117940061","term_text":"NM_009701"}}NM_009701.4	124
	R: 5′-`AGCAGAGAGATCTGGTTGCCTA`-3′
Tnni3	F: 5′-`CGAAGCAGGAGATGGAACGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009406","term_id":"1134612221","term_text":"NM_009406"}}NM_009406.3	129
	R: 5′-`GCTGTCGGCATAAGTCCTGAA`-3′
Mybpc3	F: 5′-`TCCAGATGAGTGGCAGCAAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_008653","term_id":"134031946","term_text":"NM_008653"}}NM_008653.2	178
	R: 5′-`TCTTCCAGGGACCGAGTGAT`-3′
a-SMA	F: 5′-`CCAAATCATTCCTGCCCAAAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_007392","term_id":"440309867","term_text":"NM_007392"}}NM_007392	74
	R: 5′-`TAGGCCAGGGCTACAAGTTAAG`-3′
18 s	F: 5′-`CGGCTACCACATCCAAGGAA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NR_003278","term_id":"374088232","term_text":"NR_003278"}}NR_003278.1	187
	R: 5′-`GCTGGAATTACCGCGGCT`-3′

F, forward primer; R, reverse primer.

Immunohistochemstry.

Statistical analysis.

Mice, oxygen exposures, and drug administration.

Affymetrix arrays and analysis.

Quantitative RT-PCR.

Table 1.

RT-PCR oligonucleotide sequences

Genes	Primers	GenBank Accession Number	Product Size, bp
Tnnt2	F: 5′-`GTGTGCAGTCCCTGTTCAGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001130174","term_id":"391738224","term_text":"NM_001130174"}}NM_001130174.2	130
	R: 5′-`ACCCTCAGGCTCAGGTTCA`-3′
Mhy6	F: 5′-`CTCTGGATTGGTCTCCCAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_001164171","term_id":"255918224","term_text":"NM_001164171"}}NM_001164171.1	150
	R: 5′-`GTCATTCTGTCACTCAAACTCTGG`-3′
Mhy7	F: 5′-`GCCTCAGCAGAGGAGTACAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_080728","term_id":"1371981807","term_text":"NM_080728"}}NM_080728.2	86
	R: 5′-`ATGGCTGAGCCTTGGATTCTC`-3′
Mhl3	F: 5′-`AGATCTCCATAGCCTTTGGCCT`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009701","term_id":"117940061","term_text":"NM_009701"}}NM_009701.4	124
	R: 5′-`AGCAGAGAGATCTGGTTGCCTA`-3′
Tnni3	F: 5′-`CGAAGCAGGAGATGGAACGA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_009406","term_id":"1134612221","term_text":"NM_009406"}}NM_009406.3	129
	R: 5′-`GCTGTCGGCATAAGTCCTGAA`-3′
Mybpc3	F: 5′-`TCCAGATGAGTGGCAGCAAG`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_008653","term_id":"134031946","term_text":"NM_008653"}}NM_008653.2	178
	R: 5′-`TCTTCCAGGGACCGAGTGAT`-3′
a-SMA	F: 5′-`CCAAATCATTCCTGCCCAAAGC`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NM_007392","term_id":"440309867","term_text":"NM_007392"}}NM_007392	74
	R: 5′-`TAGGCCAGGGCTACAAGTTAAG`-3′
18 s	F: 5′-`CGGCTACCACATCCAAGGAA`-3′	{"type":"entrez-nucleotide","attrs":{"text":"NR_003278","term_id":"374088232","term_text":"NR_003278"}}NR_003278.1	187
	R: 5′-`GCTGGAATTACCGCGGCT`-3′

F, forward primer; R, reverse primer.

Immunohistochemstry.

Statistical analysis.

RESULTS

Neonatal hyperoxia suppresses cardiomyocyte gene expression in the adult lung.

Although capillary rarefaction is seen in 1-yr-old mice exposed to neonatal hyperoxia, it was not observed at 8 wk of age (53). Therefore, we chose this age to identify transcriptional changes that might explain the loss of pulmonary endothelial cells. RNA was isolated from whole lungs of three 8-wk-old mice exposed to room air or 100% oxygen between PND 0–4 and was used to probe the mouse genome 430 2.0 array from Affymetrix (Fig. 1A). Out of 45,101 probes present on the array, 54 transcripts were differentially expressed between the room air and the 100% oxygen mice using a FDR of 10% (Fig. 1, B and C).

An external file that holds a picture, illustration, etc.
Object name is zh50031874140001.jpg

Fig. 1.

Neonatal hyperoxia alters gene expression in the adult lung. A: cartoon showing when lung transcriptomics was evaluated in adult mice exposed to room air or hyperoxia as neonates. B: heat map plots comparing genes differentially expressed in adult lung exposed to room air vs. hyperoxia (left) and relative to those detected in published data sets of mouse atria (right). Note that the top gene A230055J 12Rik present in whole lung is not represented in the comparison to atria. The relative log fold change, relative to the average expression of the room air/hyperoxia experiment, in expression of genes that increased (shades of red) and decreased (shades of blue) is shown above each map. C: volcano plot depicts the log fold change of all genes detected in adult mouse lungs exposed to room air vs. hyperoxia (x-axis) vs. the negative log P values that the differences were significantly different (y-axis). Each dot represents an individual gene with red representing those that were significantly different between room air and hyperoxia, and blue representing nonsignificant genes. Data were derived from individual mice (n = 3 per group).

Neonatal hyperoxia increased the expression of 11 genes, most of which regulate angiogenesis or endothelial function (Fig. 1 and Table 2). Gene ontology (GO) analysis revealed that five of these were enriched within the vasculature of the mouse cortex and hippocampus (58). These included the transcription factor Krüppel-like factor 4 (Klf4), a key regulator of endothelial function (45), RAS guanyl-releasing protein 3 (Rasgprp3), a VEGF target that promotes endothelial cell migration (40), ADAM metallopeptidase with thrombospondin type 1 motif 1 (Adamts1), a metalloproteinase that cleaves VEGF and promotes pathological angiogenesis (19), Apolipoprotein L domain containing 1 (Apold1), a lipid-binding protein upregulated in endothelial cells exposed to hypoxia (59), and the glutamate transporter solute carrier family 38 member 5 (Slc38a5) (26). Neonatal hyperoxia also increased the expression of regulator of calcineurin 1 (Rcan1), a VEGF target gene that regulates vascular function (41), UNC-5 netrin receptor C (Unc5c), a guidance receptor that inhibits angiogenesis (38), and proteoglycan 4 (Prg4), a cell surface protein whose expression increases in the lungs of COPD patients (29). The majority of genes that are increased in the lungs of mice exposed to neonatal hyperoxia are, thus, involved in angiogenesis and the regulation of endothelial cell function.

Table 2.

Adult genes whose expression in the lung is stimulated by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O2	log2_fc	P Value (Unadjusted)
Unc-5 netrin receptor 5C	Unc5c	1449522_at	8.6	9.2	0.5	1.9E-04
RAS guanyl-releasing protein 3	Rasgrp3	1438030_at	10.2	10.8	0.5	1.9E-04
Solute carrier family 38 member 5	Slc38a5	1454622_at	9.1	9.7	0.6	2.0E-04
DNAJ heat shock family (HSP40) member B1	Dnajb1	1416755_at	11.5	12.1	0.6	7.7E-05
Synaptotagmin-like 2	Sytl2	1421594_a_at	7.8	8.4	0.6	1.5E-04
Krüppel-like Factor 4	Klf4	1417394_at	11.0	11.6	0.6	2.1E-04
Regulator of calcineurin 1	Rcan1	1416601_a_at	5.6	6.4	0.8	1.3E-04
ADAM metallopeptidase with thrombospondin type 1Motif 13	Adamts1	1450716_at	9.4	10.2	0.8	8.1E-05
Apolipoprotein L domain containing 1	Apold1	1441228_at	8.6	9.5	0.9	6.4E-05
Heat shock 70 protein 8	Hspa8	1431182_at	8.3	9.3	1.0	4.9E-05
Proteoglycan 4	Prg4	1449824_at	5.5	6.9	1.4	6.6E-05

Total RNA was isolated from three lungs of 8-wk-old mice exposed to room air or hyperoxia between PND 0 and PND 4. The RNA was hybridized to the mouse genome 430 2.0 array from Affymetrix. The mean average signal intensities for each probe and the relative fold change of hyperoxia to room air were determined. Using a 10-fold false discovery rate cutoff, we list in the table those genes whose expression was significantly increased by neonatal hyperoxia.

Neonatal hyperoxia reduced expression of 43 genes (Fig. 1 and Table 3). Surprisingly, most of the suppressed genes were related to cardiac muscle (18). Neonatal hyperoxia decreased expression of genes related to contractile function [myosin light chains (Myl4, Myl7), myosin heavy chains (Myh6, Myh7), Titin Cap (Tcap)], calcium signaling [troponins (Tnni3, Tnnc1), sarcalumenin (Srl)], a cardiac-specific mitochondrial gene [cytochrome-c oxidase subunit VIa, polypeptide 2 (Cox6A2)] (3), and genes involved in vasodilation [atrial natriuretic peptide (Nppa) and corin (Crn)] (60). The impeded expression of these cardiac genes correlated tightly with their absolute expression levels in atrial cells (52), suggesting that we observed the loss of a subpopulation of atrial-like cells within the lung (Fig. 1B).

Table 3.

Adult genes whose expression in the lung is suppressed by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O₂	log2_fc	P value (Unadjusted)
Natriuretic peptide A	Nppa	1456062_at	10.2	7.0	−3.2	1.4E-04
Potassium voltage-gated channel subfamily J member 3	Kcnj3	1455374_at	7.2	5.0	−2.2	1.1E-04
Synaptopodin 2-like	Synpo2l	1447657_s_at	8.0	6.0	−2.0	9.3E-06
Sarcolipin	Sln	1420884_at	9.8	8.1	−1.7	6.2E-05
Troponin C1	Tnnc1	1418370_at	10.7	9.0	−1.7	7.7E-05
Sarcalumenin	Srl	1436867_at	10.5	8.8	−1.7	1.3E-04
Triadin	Trdn	1426142_a_at	8.0	6.4	−1.7	9.9E-05
Cysteine and glycine-rich protein 3	Csrp3	1460318_at	10.1	8.4	−1.6	1.7E-04
Somatostatin receptor 4	Sstr4	1457440_at	7.5	5.9	−1.6	1.2E-06
Myosin heavy chain 6/7	Myh6/Myh7	1448826_at	11.7	10.2	−1.5	6.4E-05
Myosin binding protein H-like	Mybphl	1430269_at	7.2	5.7	−1.5	1.8E-05
Troponin T2	Tnnt2	1418726_a_at	11.8	10.4	−1.4	2.1E-05
Troponin I, cardiac 3	Tnni3	1422536_at	9.8	8.4	−1.4	6.6E-05
Corin	Crn	1419017_at	6.3	5.0	−1.4	2.4E-05
Phosphoglycerate mutase 2	Pgam2	1418373_at	8.1	6.7	−1.4	8.3E-05
Titin cap	Tcap	1423145_a_at	8.9	7.6	−1.4	6.2E-05
Heat shock protein family, member 7	Hspb7	1434927_at	7.8	6.5	−1.3	7.2E-05
Myosin light chain 7	Myl7	1449071_at	11.5	10.2	−1.3	3.2E-05
Sarcoplasmic reticulum histidine-rich calcium-binding protein	Hrc	1419109_at	7.5	6.3	−1.3	1.4E-04
Actinin, α₂	Actn2	1448327_at	8.7	7.4	−1.3	4.7E-05
Myosin light chain 3	Myl3	1427769_x_at	6.5	5.3	−1.2	4.3E-06
Leiomodin 2	Lmod2	1452345_at	7.1	5.9	−1.2	3.0E-05
Myomesin 2	Myom2	1438175_x_at	6.9	5.7	−1.2	1.4E-04
T-box 20	Tbx20	1425158_at	5.9	4.9	−1.1	6.7E-05
Myosin light chain 4	Mylk4	1441111_at	8.1	7.0	−1.1	1.4E-04
Phosphodiesterase 4 interacting protein	Pde4dip	1417626_at	7.6	6.5	−1.1	6.0E-05
RNA binding motif protein 20	Rbm20	1429024_at	8.2	7.1	−1.0	1.7E-05
Phosphorylase kinase gamma 1	Phkg1	1425164_a_at	7.0	6.0	−1.0	1.6E-04
Myopalladin	Mypn	1435813_at	6.7	5.7	−1.0	8.9E-05
Kelch-like family member 29	Klhl29	1434639_at	8.4	7.4	−1.0	1.2E-05
Ankyrin 3 repeat domain 63	Ankrd63	1443287_at	9.6	8.6	−1.0	6.5E-05
Leiomodin 3	Lmod3	1439658_at	5.7	4.7	−1.0	1.9E-04
Ferm domain containing protein 5	Frmd5	1436243_at	7.1	6.2	−0.9	5.1E-05
LIM domain binding 3	Ldb3	1433783_at	9.5	8.6	−0.9	3.0E-06
RIKEN cDNA A230055J12	A230055J12Rik	1443457_at	8.0	7.2	−0.8	1.5E-04
Cytochrome-c oxidase subunit VIa, polypeptide 2	Cox6a2	1417607_at	10.3	9.5	−0.8	8.2E-05
Tetraspanin 8	Tspan8	1455709_at	7.9	7.2	−0.7	1.4E-04
Period circadian homolog 3	Per3	1442243_at	9.8	9.1	−0.6	1.4E-04
		1459947_at	10.5	9.9	−0.6	8.2E-05
Zinc finger protein 503	Zfp503	1423835_at	10.2	9.7	−0.6	1.7E-04
thymic stromal lymphopoietin	Tslp	1450004_at	9.8	9.3	−0.5	1.0E-04
D site albumin promoter binding protein	Dbp	1438211_s_at	12.9	12.5	−0.4	1.0E-04

Total RNA was isolated from three lungs of 8-wk-old mice exposed to room air or hyperoxia between PND 0 and PND 4. The RNA was hybridized to the mouse genome 430 2.0 array from Affymetrix. The mean average signal intensities for each probe and the relative fold change of hyperoxia to room air were determined. Using a 10% false discovery rate cutoff, we list in the table those genes whose expression was significantly decreased by neonatal hyperoxia.

Neonatal hyperoxia depletes pulmonary vein cardiomyocytes within the lung.

We sought to understand the meaning of these transcriptional changes by investigating the expression and localization of cardiac-specific isoforms of myosin and troponin in the lung. Quantitative RT-PCR confirmed that neonatal hyperoxia suppressed expression of myosin heavy chains 6 and 7, myosin light chain 3, myosin binding protein 3, and troponin T type 2 and troponin I type 3 in the adult lung (Fig. 2A). These changes were specific for striated and cardiac muscle genes because neonatal hyperoxia did not significantly suppress expression of α-smooth muscle actin. To identify pulmonary cells expressing cardiomyocyte genes, lung sections of adult mice exposed to room air or hyperoxia as neonates were stained for the cardiac-specific isoform of troponin T (TNNT2). In adult mice exposed to room air as neonates, TNNT2 was robustly detected in a thick band of myocardial cells wrapping the intralobular pulmonary vein (Fig. 2B). In contrast, TNNT2 was detected faintly in adult lungs exposed to neonatal hyperoxia as a patchy band of thin cells wrapping the pulmonary vein. High-power hematoxylin-eosin stains of the pulmonary vein suggested neonatal hyperoxia depleted cardiomyocytes without obviously affecting the vessel wall (Fig. 2B).

An external file that holds a picture, illustration, etc.
Object name is zh50031874140002.jpg

Fig. 2.

Neonatal hyperoxia suppresses expression of cardiomyocyte genes in the adult lung. A: quantitative RT-PCR (qRT-PCR) was used to evaluate RNA expression in lungs of adult (PND 56) mice exposed to room air or hyperoxia as neonates. Values represent means ± SE of four mice per group with *P < 0.05 and **P < 0.01 compared with room air. B: lungs were stained for cardiac-specific isoform of troponin T (TNNT2; red) and DAPI (blue). Arrows point to gaps in TNNT2 staining. C: lungs were stained with hematoxylin and eosin. Insets are enlarged in the lower images that contain arrows pointing to the vessel wall of the pulmonary vein. B and C: scale bar = 20 μm.

The intact lung with heart was sectioned and stained for TNNT2 to determine whether the loss of cardiomyocytes extended outside of the lung. While the extralobular pulmonary vein near the hilum of mice exposed to room air contained a thick layer of cardiomyocytes, a partial layer that did not fully cover the circumference of the vessel was seen in mice exposed to room air (Fig. 3, A and B). These changes were less evident where the pulmonary vein attached to the left ventricle (Fig. 3, C and D). Here, a thick band of cardiomyocytes was seen wrapping the entire circumference of the pulmonary vein of mice exposed to room air or neonatal hyperoxia. However, high-power imaging of the muscle revealed it was thinner and less organized in mice exposed to neonatal hyperoxia when compared with room air (Fig. 3, E and F). Mild effects of neonatal hyperoxia were also seen in the left atria. The left atria of mice exposed to neonatal hyperoxia appeared more dilated compared with control mice (Fig. 3, G and H). Although the overall myocardial wall was intact, the myocardium of oxygen-exposed mice was thinner and more loosely organized in some areas (Fig. 3, I and J). This loss of cardiomyocytes was not evident in the right atrium or the ventricles. Instead, the right ventricles appeared slightly hypertrophic, and that was supported by a slight increase in the Fulton index (weight of the right ventricle to the left ventricle plus septum) (20). But this did not reach statistical significance until the mice were 16 wk old (data not shown) or as previously reported in 1-yr-old mice (55). Taken together, these findings suggest neonatal hyperoxia causes a graded loss of cardiomyocytes from the lung to the left atrium. These changes occurred before overt loss of pulmonary capillaries or prominent right ventricular hypertrophy.

An external file that holds a picture, illustration, etc.
Object name is zh50031874140003.jpg

Fig. 3.

Neonatal hyperoxia causes a graded loss of cardiomyocytes wrapping the pulmonary vein and in the left atrium. The extralobular pulmonary vein near the hilum (A and B) and the left atrium (C–F) were stained for TNNT2 (red) and DAPI (blue). Arrows point to cardiomyocytes expressing TNNT2 wrapping the pulmonary vein (A and B), while arrowheads (B) point to regions devoid of TNNT2+ cardiomyocytes in mice exposed to neonatal hyperoxia. Hearts were stained for TNNT2 (red) (G and H) and DAPI (blue) (I and J). Arrows (I and J) point to thinned regions of the left atrium. PV, pulmonary vein; LA, left atrium.

Lungs were stained for TNNT2 at the end of the oxygen exposure (PND 4) and every other week during recovery in room air (PND 7–PND 56) to determine when they were depleted. In mice exposed to room air, TNNT2 was detected in a thick bundle of striated cells wrapping the pulmonary vein within the lung parenchyma (Fig. 4A). Staining always appeared uniform and wrapped the entire vein. While TNNT2 was also detected at PND 4 and PND 7 in mice exposed to neonatal hyperoxia, gaps in staining were evident by PND 14 that increased with age such that cells expressing TNNT2 were rarely detected on PND 56. Consistent with these findings, Tnnt2 mRNA was significantly reduced in PND 14, PND 28, and PND 42 mice exposed to neonatal hyperoxia when compared with younger mice exposed to hyperoxia or siblings exposed to room air (Fig. 4B). Reduced expression of Tnnt2 mRNA also associated with reduced suppression of Myh6 and Myh7 mRNA (Fig. 4, C and D); however, loss of Myh7 mRNA was temporally delayed when compared with Tnnt2 and Myh6.

An external file that holds a picture, illustration, etc.
Object name is zh50031874140004.jpg

Fig. 4.

Pulmonary vein cardiomyocytes are depleted during recovery in room air. A: newborn mice were exposed to room air or hyperoxia between PND 0 and PND 4, and then the oxygen-exposed mice were returned to room air. Lungs collected on PND 4, PND 7, PND 14, PND 28, PND 42, and PND 56 were stained for TNNT2 (red) and DAPI (blue). The pulmonary vein is outlined in older mice exposed to neonatal hyperoxia with arrows pointing to gaps in TNNT2 staining. Scale bar = 20 μm. B–D: quantitative RT-PCR was used to evaluate Tnnt2, Myh6, and Myh7 mRNA in lungs of mice exposed to room air or hyperoxia as neonates. Values are expressed as means ± SE of 4–6 mice per group with *P < 0.05 and **P < 0.01 compared with room air.

TUNEL staining was used to determine whether neonatal hyperoxia stimulated apoptosis of Tnnt2+ cells. Although TUNEL+ cells were occasionally detected in lungs exposed to hyperoxia, they were surprisingly not found in Tnnt2+ cardiomyocytes at PND 4 or during recovery in room air (data not shown). Therefore, we stained lungs for Ki67 to determine whether neonatal hyperoxia inhibited proliferation of these cells. On PND 4 and PND 7, Ki67 was detected in ~20% of Tnnt2+ cardiomyocytes and then progressively declined with age (Fig. 5, A and B). Neonatal hyperoxia significantly reduced the number of Tnnt2+ cells that expressed Ki67, and this persisted even during recovery in room air.

An external file that holds a picture, illustration, etc.
Object name is zh50031874140005.jpg

Fig. 5.

Neonatal hyperoxia suppresses proliferation of pulmonary vein cardiomyocytes (CMs). Newborn mice were exposed to room air or hyperoxia between PND 0 and PND 4, and then the oxygen-exposed mice were returned to room air. A: lungs collected on PND 4, PND 7, PND 14, and PND 28 were stained for Ki67 (green), TNNT2 (red), and DAPI (blue). Arrows point to Ki67+ / TNNT2+ cells. B: proportion of Ki67+ cells that were also TNNT2 and DAPI+ was quantified and graphed. Values are expressed as means ± SE of 4–6 mice per group with *P < 0.05 compared with room air.

We previously showed that 60% oxygen between PND 0 and PND 4 was sufficient to alter alveolar lung development, while 100% oxygen was required to also alter host response to influenza A virus infection (53). Changes in lung development or host response to infection were not seen in adult mice exposed to 40% oxygen. To define the dose of oxygen that impairs growth of pulmonary vein cardiomyocytes, newborn mice were exposed to room air (21%), 40, 60, 80, or 100% oxygen between PND 0 and PND 4. Mice exposed to oxygen recovered in room air until PND 56 when their lungs were stained for TNNT2. TNNT2 was detected around the intralobular pulmonary veins of mice exposed to room air and 40 or 60% oxygen (Fig. 6A). It was markedly diminished in lungs of mice exposed to 80 or 100% oxygen. Consistent with these findings, exposure to 80 or 100% oxygen significantly reduced pulmonary expression of Tnnt2, Myh6, and Myh7 mRNA (Fig. 6B).

An external file that holds a picture, illustration, etc.
Object name is zh50031874140006.jpg

Fig. 6.

Neonatal hyperoxia causes dose-dependent loss of pulmonary vein cardiomyocytes. A: newborn mice were exposed to room air (21%), 40, 60, 80, or 100% oxygen between PND 0 and PND 4. Mice exposed to 40–100% oxygen were then returned to room air. Lungs were collected on PND 56 and stained for TNNT2 (red) and DAPI (blue). The pulmonary vein is outlined in some images with arrows pointing to gaps in TNNT2 staining. Fat arrows point to background fluorescence produced by erythrocytes within pulmonary vein. Scale bar = 20 μm. B–D: qRT-PCR was used to evaluate Tnnt2, Myh6, and Myh7 mRNA in lungs of adult (PND 56) mice exposed to room air or different doses of hyperoxia as neonates. Values are expressed as means ± SE of four mice per group with *P < 0.05, **P < 0.01, and ***P < 0.001 compared with room air.

Genetic lineage labeling studies confirm neonatal hyperoxia depletes pulmonary vein cardiomyocytes.

Genetic lineage labeling studies using Myh6^{; R26R}^{mice (}47) were used to determine whether neonatal hyperoxia depleted pulmonary vein cardiomyocytes or suppressed expression of cardiomyocyte genes. We first confirmed that neonatal hyperoxia suppressed Myh6 expression in pulmonary vein cardiomyocytes by exposing the bitransgenic mice to room air or 100% oxygen between PND 0 and PND 4 (Fig. 7A). Around PND 56, mice were injected for three consecutive days with tamoxifen or corn oil vehicle. On the 7th day, lungs were harvested and stained for EGFP and TNNT2. In lungs of mice exposed to room air, EGFP was detected in striated cells expressing TNNT2 and not elsewhere within the lung (Fig. 7B). Although EGFP and TNNT2 were also detected in lungs of mice exposed to neonatal hyperoxia, they were coexpressed in a thin patchy pattern surrounding the pulmonary vein. EGFP was never detected in room air- or hyperoxia-exposed mice that had been administered corn oil vehicle (data not shown).

An external file that holds a picture, illustration, etc.
Object name is zh50031874140007.jpg

Fig. 7.

Genetic lineage mapping confirms neonatal hyperoxia depletes pulmonary vein CMs. A: cartoon showing Myh6^{; R26R}^{mice were exposed to room air or hyperoxia as neonates and then administered tamoxifen or corn oil vehicle as adults.}B: lungs were collected on the 7th day after the first administration of tamoxifen and stained for EGFP (green), TNNT2 (red), and DAPI (blue). Thin arrows point to gaps in EGFP and TNNT2 staining. Dashed boxes are enlarged below each image. Scale bar = 20 μm. C: cartoon showing Myh6^{; R26R}^{mice were administered tamoxifen or corn oil vehicle at birth and then exposed to room air or hyperoxia. Mice exposed to hyperoxia were then recovered in room air. Lungs were collected on PND 4, PND 14, PND 28, and PND 56, and stained for EGFP (green), TNNT2 (red), and DAPI (blue). Thin arrows point to gaps in EGFP and TNNT2 staining. Scale bar = 20 μm.}

We then conditionally labeled cardiomyocytes with EGFP by injecting mice with tamoxifen or vehicle once at birth (Fig. 7C). The mice were then exposed to room air or 100% oxygen (hyperoxia) through PND 4. Lungs were either harvested at this time or after mice were allowed to recover in room air. Lung sections were stained for EGFP and TNNT2. EGFP+; TNNT2+ cells were seen in the lungs of mice exposed to room air (Fig. 7C). While EGFP+; TNNT2+ cells were also seen in PND 4 mice exposed to hyperoxia, they were clearly less abundant by 2 wk, suggesting they were reduced in numbers by this age. Taken together, this demonstrates that neonatal hyperoxia depletes cardiomyocytes (CMs) that wrap the intralobular pulmonary vein.

Mitochondrial antioxidants inhibit the oxygen-dependent loss of pulmonary vein CMs.

It is widely accepted that neonatal hyperoxia alters postnatal lung development via the production of mitochondrial and nonmitochondrial sources of reactive oxygen species (ROS), and the ensuing inflammatory injury (23, 33). Consistent with this conclusion, transgenic Sftpc^{mice that overexpress extracellular superoxide dismutase (ECSOD) in alveolar epithelial type II cells do not develop alveolar simplification or demonstrate an altered host response to influenza A virus infection when they have been exposed to neonatal hyperoxia (}8). Because ECSOD is secreted, its ability to preserve postnatal lung development and the host response to infection may extend beyond the alveolar epithelium (2). To determine whether ECSOD could blunt the loss of pulmonary vein CMs, PND 56 lungs were collected from wild-type (WT) and Sftpc^{transgenic (TG) mice exposed to room air or neonatal hyperoxia and stained for TNNT2. TNNT2 staining was readily detected around the pulmonary vein of WT and TG mice exposed to room air (}Fig. 8A). While neonatal hyperoxia reduced TNNT2 staining in lungs of adult wild-type mice, it surprisingly also reduced TNNT2 staining in lungs of Sftpc^{mice. Quantitative RT-PCR confirmed that neonatal hyperoxia inhibited}Tnnt2, Myh6, and Myh7 mRNA in both adult wild-type and Sftpc^{mice (}Fig. 8B). These findings indicate overexpression of extracellular SOD in alveolar epithelial type II cells is not sufficient to prevent the oxygen-dependent loss of pulmonary vein CMs.

An external file that holds a picture, illustration, etc.
Object name is zh50031874140008.jpg

Fig. 8.

Overexpression of extracellular SOD in alveolar epithelial type II cells does not block the oxygen-dependent loss of pulmonary vein cardiomyocytes. A: newborn wild-type (WT) and Sftpc^{transgenic (TG) mice were exposed to room air or hyperoxia between PND 0 and PND 4 and then the oxygen-exposed mice were returned to room air. Lungs collected on PND 56 were stained for TNNT2 (red) and DAPI (blue). The pulmonary vein is outlined in mice exposed to neonatal hyperoxia, and arrows point to gaps in TNNT2 staining. Scale bar = 20 μm.}C: qRT-PCR was used to evaluate Tnnt2, Myh6, and Myh7 mRNA in lungs of adult WT and TG mice exposed to room air or hyperoxia as neonates. Values are expressed as means ± SE of four mice per group with *P < 0.05 and **P < 0.01 compared with room air.

Since targeted expression of an antioxidant to alveolar epithelial type II cells was not sufficient to prevent the oxygen-dependent loss of pulmonary vein CMs, we administered mitoTEMPO or vehicle to newborn mice exposed to room air or hyperoxia on PND 1, PND 2, and PND 3 (Fig. 9A). MitoTEMPO is a mitochondrial targeted superoxide dismutase mimetic that specifically scavenges superoxide. We first confirmed that mitoTEMPO reduced oxidative injury by staining PND 4 lungs for TNNT and 8-oxoguanine. In lungs of mice exposed to hyperoxia, increased 8-oxoguanine staining was observed in TNNT2+ cells but not when mice were administered mitoTEMPO (data not shown). We then investigated whether administration of mitoTEMPO during neonatal hyperoxia prevented the loss of pulmonary vein CMs seen in adult mice. As shown earlier, neonatal hyperoxia reduced TNNT2 staining in lungs of adult mice (Fig. 9B). In contrast, TNNT2 staining was readily detected in hyperoxia-exposed lungs that had been administered mitoTEMPO. Similarly, neonatal hyperoxia suppressed expression of Tnnt2, Myh6, and Myh7 mRNA in adult lung (Fig. 9C). Administration of mitoTEMPO significantly blunted the loss of Myh6 and Tnnt2 mRNA and fully preserved expression of Myh7 in adult mice exposed to neonatal hyperoxia. MitoTEMPO did not alter expression of these genes in mice exposed to room air.

An external file that holds a picture, illustration, etc.
Object name is zh50031874140009.jpg

Fig. 9.

mitoTEMPO inhibits the oxygen-dependent loss of pulmonary vein cardiomyocytes. A: cartoon showing newborn mice were administered mitoTEMPO (arrows) or vehicle on PND 0, PND 1, and PND 2 while being exposed to room air or hyperoxia between PND 0 and PND 4. B: lungs were collected on PND 56 and stained for TNNT2 (red) and DAPI (blue). The pulmonary vein is outlined, and thin arrows point to gaps in TNNT2 staining in mice exposed to neonatal hyperoxia. Fat arrows point to background fluorescence produced by erythrocytes within pulmonary vein. Scale bar = 50 μm. C: qRT-PCR was used to evaluate Tnnt2, Myh6, and Myh7 mRNA in lungs of adult mice exposed to room air or hyperoxia as neonates. Values are expressed as means ± SE of 4–6 mice per group with *P < 0.05, **P < 0.01, and ***P < 0.001 compared with room air. Scale bar = 50 μm.

Neonatal hyperoxia suppresses cardiomyocyte gene expression in the adult lung.

Fig. 1.

Table 2.

Adult genes whose expression in the lung is stimulated by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O2	log2_fc	P Value (Unadjusted)
Unc-5 netrin receptor 5C	Unc5c	1449522_at	8.6	9.2	0.5	1.9E-04
RAS guanyl-releasing protein 3	Rasgrp3	1438030_at	10.2	10.8	0.5	1.9E-04
Solute carrier family 38 member 5	Slc38a5	1454622_at	9.1	9.7	0.6	2.0E-04
DNAJ heat shock family (HSP40) member B1	Dnajb1	1416755_at	11.5	12.1	0.6	7.7E-05
Synaptotagmin-like 2	Sytl2	1421594_a_at	7.8	8.4	0.6	1.5E-04
Krüppel-like Factor 4	Klf4	1417394_at	11.0	11.6	0.6	2.1E-04
Regulator of calcineurin 1	Rcan1	1416601_a_at	5.6	6.4	0.8	1.3E-04
ADAM metallopeptidase with thrombospondin type 1Motif 13	Adamts1	1450716_at	9.4	10.2	0.8	8.1E-05
Apolipoprotein L domain containing 1	Apold1	1441228_at	8.6	9.5	0.9	6.4E-05
Heat shock 70 protein 8	Hspa8	1431182_at	8.3	9.3	1.0	4.9E-05
Proteoglycan 4	Prg4	1449824_at	5.5	6.9	1.4	6.6E-05

Table 3.

Adult genes whose expression in the lung is suppressed by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O₂	log2_fc	P value (Unadjusted)
Natriuretic peptide A	Nppa	1456062_at	10.2	7.0	−3.2	1.4E-04
Potassium voltage-gated channel subfamily J member 3	Kcnj3	1455374_at	7.2	5.0	−2.2	1.1E-04
Synaptopodin 2-like	Synpo2l	1447657_s_at	8.0	6.0	−2.0	9.3E-06
Sarcolipin	Sln	1420884_at	9.8	8.1	−1.7	6.2E-05
Troponin C1	Tnnc1	1418370_at	10.7	9.0	−1.7	7.7E-05
Sarcalumenin	Srl	1436867_at	10.5	8.8	−1.7	1.3E-04
Triadin	Trdn	1426142_a_at	8.0	6.4	−1.7	9.9E-05
Cysteine and glycine-rich protein 3	Csrp3	1460318_at	10.1	8.4	−1.6	1.7E-04
Somatostatin receptor 4	Sstr4	1457440_at	7.5	5.9	−1.6	1.2E-06
Myosin heavy chain 6/7	Myh6/Myh7	1448826_at	11.7	10.2	−1.5	6.4E-05
Myosin binding protein H-like	Mybphl	1430269_at	7.2	5.7	−1.5	1.8E-05
Troponin T2	Tnnt2	1418726_a_at	11.8	10.4	−1.4	2.1E-05
Troponin I, cardiac 3	Tnni3	1422536_at	9.8	8.4	−1.4	6.6E-05
Corin	Crn	1419017_at	6.3	5.0	−1.4	2.4E-05
Phosphoglycerate mutase 2	Pgam2	1418373_at	8.1	6.7	−1.4	8.3E-05
Titin cap	Tcap	1423145_a_at	8.9	7.6	−1.4	6.2E-05
Heat shock protein family, member 7	Hspb7	1434927_at	7.8	6.5	−1.3	7.2E-05
Myosin light chain 7	Myl7	1449071_at	11.5	10.2	−1.3	3.2E-05
Sarcoplasmic reticulum histidine-rich calcium-binding protein	Hrc	1419109_at	7.5	6.3	−1.3	1.4E-04
Actinin, α₂	Actn2	1448327_at	8.7	7.4	−1.3	4.7E-05
Myosin light chain 3	Myl3	1427769_x_at	6.5	5.3	−1.2	4.3E-06
Leiomodin 2	Lmod2	1452345_at	7.1	5.9	−1.2	3.0E-05
Myomesin 2	Myom2	1438175_x_at	6.9	5.7	−1.2	1.4E-04
T-box 20	Tbx20	1425158_at	5.9	4.9	−1.1	6.7E-05
Myosin light chain 4	Mylk4	1441111_at	8.1	7.0	−1.1	1.4E-04
Phosphodiesterase 4 interacting protein	Pde4dip	1417626_at	7.6	6.5	−1.1	6.0E-05
RNA binding motif protein 20	Rbm20	1429024_at	8.2	7.1	−1.0	1.7E-05
Phosphorylase kinase gamma 1	Phkg1	1425164_a_at	7.0	6.0	−1.0	1.6E-04
Myopalladin	Mypn	1435813_at	6.7	5.7	−1.0	8.9E-05
Kelch-like family member 29	Klhl29	1434639_at	8.4	7.4	−1.0	1.2E-05
Ankyrin 3 repeat domain 63	Ankrd63	1443287_at	9.6	8.6	−1.0	6.5E-05
Leiomodin 3	Lmod3	1439658_at	5.7	4.7	−1.0	1.9E-04
Ferm domain containing protein 5	Frmd5	1436243_at	7.1	6.2	−0.9	5.1E-05
LIM domain binding 3	Ldb3	1433783_at	9.5	8.6	−0.9	3.0E-06
RIKEN cDNA A230055J12	A230055J12Rik	1443457_at	8.0	7.2	−0.8	1.5E-04
Cytochrome-c oxidase subunit VIa, polypeptide 2	Cox6a2	1417607_at	10.3	9.5	−0.8	8.2E-05
Tetraspanin 8	Tspan8	1455709_at	7.9	7.2	−0.7	1.4E-04
Period circadian homolog 3	Per3	1442243_at	9.8	9.1	−0.6	1.4E-04
		1459947_at	10.5	9.9	−0.6	8.2E-05
Zinc finger protein 503	Zfp503	1423835_at	10.2	9.7	−0.6	1.7E-04
thymic stromal lymphopoietin	Tslp	1450004_at	9.8	9.3	−0.5	1.0E-04
D site albumin promoter binding protein	Dbp	1438211_s_at	12.9	12.5	−0.4	1.0E-04

Total RNA was isolated from three lungs of 8-wk-old mice exposed to room air or hyperoxia between PND 0 and PND 4. The RNA was hybridized to the mouse genome 430 2.0 array from Affymetrix. The mean average signal intensities for each probe and the relative fold change of hyperoxia to room air were determined. Using a 10% false discovery rate cutoff, we list in the table those genes whose expression was significantly decreased by neonatal hyperoxia.

Neonatal hyperoxia depletes pulmonary vein cardiomyocytes within the lung.

Genetic lineage labeling studies confirm neonatal hyperoxia depletes pulmonary vein cardiomyocytes.

Fig. 7.

Mitochondrial antioxidants inhibit the oxygen-dependent loss of pulmonary vein CMs.

Fig. 8.

Fig. 9.

Neonatal hyperoxia suppresses cardiomyocyte gene expression in the adult lung.

Fig. 1.

Table 2.

Adult genes whose expression in the lung is stimulated by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O2	log2_fc	P Value (Unadjusted)
Unc-5 netrin receptor 5C	Unc5c	1449522_at	8.6	9.2	0.5	1.9E-04
RAS guanyl-releasing protein 3	Rasgrp3	1438030_at	10.2	10.8	0.5	1.9E-04
Solute carrier family 38 member 5	Slc38a5	1454622_at	9.1	9.7	0.6	2.0E-04
DNAJ heat shock family (HSP40) member B1	Dnajb1	1416755_at	11.5	12.1	0.6	7.7E-05
Synaptotagmin-like 2	Sytl2	1421594_a_at	7.8	8.4	0.6	1.5E-04
Krüppel-like Factor 4	Klf4	1417394_at	11.0	11.6	0.6	2.1E-04
Regulator of calcineurin 1	Rcan1	1416601_a_at	5.6	6.4	0.8	1.3E-04
ADAM metallopeptidase with thrombospondin type 1Motif 13	Adamts1	1450716_at	9.4	10.2	0.8	8.1E-05
Apolipoprotein L domain containing 1	Apold1	1441228_at	8.6	9.5	0.9	6.4E-05
Heat shock 70 protein 8	Hspa8	1431182_at	8.3	9.3	1.0	4.9E-05
Proteoglycan 4	Prg4	1449824_at	5.5	6.9	1.4	6.6E-05

Table 3.

Adult genes whose expression in the lung is suppressed by neonatal hyperoxia

Gene Name	Symbol	PROBEID	log2(exp) RA	log2(exp) O₂	log2_fc	P value (Unadjusted)
Natriuretic peptide A	Nppa	1456062_at	10.2	7.0	−3.2	1.4E-04
Potassium voltage-gated channel subfamily J member 3	Kcnj3	1455374_at	7.2	5.0	−2.2	1.1E-04
Synaptopodin 2-like	Synpo2l	1447657_s_at	8.0	6.0	−2.0	9.3E-06
Sarcolipin	Sln	1420884_at	9.8	8.1	−1.7	6.2E-05
Troponin C1	Tnnc1	1418370_at	10.7	9.0	−1.7	7.7E-05
Sarcalumenin	Srl	1436867_at	10.5	8.8	−1.7	1.3E-04
Triadin	Trdn	1426142_a_at	8.0	6.4	−1.7	9.9E-05
Cysteine and glycine-rich protein 3	Csrp3	1460318_at	10.1	8.4	−1.6	1.7E-04
Somatostatin receptor 4	Sstr4	1457440_at	7.5	5.9	−1.6	1.2E-06
Myosin heavy chain 6/7	Myh6/Myh7	1448826_at	11.7	10.2	−1.5	6.4E-05
Myosin binding protein H-like	Mybphl	1430269_at	7.2	5.7	−1.5	1.8E-05
Troponin T2	Tnnt2	1418726_a_at	11.8	10.4	−1.4	2.1E-05
Troponin I, cardiac 3	Tnni3	1422536_at	9.8	8.4	−1.4	6.6E-05
Corin	Crn	1419017_at	6.3	5.0	−1.4	2.4E-05
Phosphoglycerate mutase 2	Pgam2	1418373_at	8.1	6.7	−1.4	8.3E-05
Titin cap	Tcap	1423145_a_at	8.9	7.6	−1.4	6.2E-05
Heat shock protein family, member 7	Hspb7	1434927_at	7.8	6.5	−1.3	7.2E-05
Myosin light chain 7	Myl7	1449071_at	11.5	10.2	−1.3	3.2E-05
Sarcoplasmic reticulum histidine-rich calcium-binding protein	Hrc	1419109_at	7.5	6.3	−1.3	1.4E-04
Actinin, α₂	Actn2	1448327_at	8.7	7.4	−1.3	4.7E-05
Myosin light chain 3	Myl3	1427769_x_at	6.5	5.3	−1.2	4.3E-06
Leiomodin 2	Lmod2	1452345_at	7.1	5.9	−1.2	3.0E-05
Myomesin 2	Myom2	1438175_x_at	6.9	5.7	−1.2	1.4E-04
T-box 20	Tbx20	1425158_at	5.9	4.9	−1.1	6.7E-05
Myosin light chain 4	Mylk4	1441111_at	8.1	7.0	−1.1	1.4E-04
Phosphodiesterase 4 interacting protein	Pde4dip	1417626_at	7.6	6.5	−1.1	6.0E-05
RNA binding motif protein 20	Rbm20	1429024_at	8.2	7.1	−1.0	1.7E-05
Phosphorylase kinase gamma 1	Phkg1	1425164_a_at	7.0	6.0	−1.0	1.6E-04
Myopalladin	Mypn	1435813_at	6.7	5.7	−1.0	8.9E-05
Kelch-like family member 29	Klhl29	1434639_at	8.4	7.4	−1.0	1.2E-05
Ankyrin 3 repeat domain 63	Ankrd63	1443287_at	9.6	8.6	−1.0	6.5E-05
Leiomodin 3	Lmod3	1439658_at	5.7	4.7	−1.0	1.9E-04
Ferm domain containing protein 5	Frmd5	1436243_at	7.1	6.2	−0.9	5.1E-05
LIM domain binding 3	Ldb3	1433783_at	9.5	8.6	−0.9	3.0E-06
RIKEN cDNA A230055J12	A230055J12Rik	1443457_at	8.0	7.2	−0.8	1.5E-04
Cytochrome-c oxidase subunit VIa, polypeptide 2	Cox6a2	1417607_at	10.3	9.5	−0.8	8.2E-05
Tetraspanin 8	Tspan8	1455709_at	7.9	7.2	−0.7	1.4E-04
Period circadian homolog 3	Per3	1442243_at	9.8	9.1	−0.6	1.4E-04
		1459947_at	10.5	9.9	−0.6	8.2E-05
Zinc finger protein 503	Zfp503	1423835_at	10.2	9.7	−0.6	1.7E-04
thymic stromal lymphopoietin	Tslp	1450004_at	9.8	9.3	−0.5	1.0E-04
D site albumin promoter binding protein	Dbp	1438211_s_at	12.9	12.5	−0.4	1.0E-04

Total RNA was isolated from three lungs of 8-wk-old mice exposed to room air or hyperoxia between PND 0 and PND 4. The RNA was hybridized to the mouse genome 430 2.0 array from Affymetrix. The mean average signal intensities for each probe and the relative fold change of hyperoxia to room air were determined. Using a 10% false discovery rate cutoff, we list in the table those genes whose expression was significantly decreased by neonatal hyperoxia.

Neonatal hyperoxia depletes pulmonary vein cardiomyocytes within the lung.

Genetic lineage labeling studies confirm neonatal hyperoxia depletes pulmonary vein cardiomyocytes.

Fig. 7.

Mitochondrial antioxidants inhibit the oxygen-dependent loss of pulmonary vein CMs.

Fig. 8.

Fig. 9.

DISCUSSION

Persistent lung disease seen in children born preterm is mediated, in part, by their early life exposure to oxygen. However, now that these individuals are reaching adulthood, there is growing concern that they are also at risk for elevated blood pressure, reduced circulating levels of antiangiogenic factors, systemic and pulmonary capillary rarefaction, and increased right ventricular mass (15, 17, 27, 30, 31). These findings suggest the long-term cardiovascular health of these individuals will be poor and may even result in a shortened lifespan. Since this pathology takes years to manifest, there is an urgent need to identify early antecedents that initiate or drive cardiovascular disease. Here, RNA transcriptomics was used to identify molecular changes that preceded overt cardiovascular disease seen in adult mice exposed briefly to hyperoxia at birth. We discovered neonatal hyperoxia suppressed expression of cardiac-specific genes in the lung and that this reflected loss in the number of cardiomyocytes wrapping the pulmonary vein. While the greatest loss of pulmonary vein cardiomyocytes was seen within the lung, a graded loss of cells was seen beginning at the hilum and extending into the left atrium. Therefore, the early loss of these cardiac muscle cells may contribute to the pathogenesis of cardiovascular disease seen in these mice and perhaps in people born preterm.

Cardiomyocytes wrapping the pulmonary vein extend from the left atrium to the hilum in all species except rodents where they also enter lobes of the lung (28, 36). These cells are thought to help pump oxygen-rich blood out of the lung because partial removal of pulmonary vein from humans with atrial fibrillation causes deterioration of left atrial contraction and filling (34). These hemodynamic changes were associated with reduced pulmonary vein contraction measured by CT scans before and after surgery. The higher respiration rate of rodents may necessitate deeper migration of these pulmonary vein sleeve cells into the lung parenchyma. Regardless of how deep these cells extend into the lungs of different species, their loss within and outside the lung probably impairs blood flow and, thus, helps explain why early life oxygen causes dilation of pulmonary arterioles and veins (55). Right ventricular hypertrophy observed as these mice age may reflect an adaptive response as the heart seeks to maintain proper blood flow against an increasing intrapulmonary pressure. This right-sided cardiac pathology has been seen in adolescents born preterm (30) and is considered a hallmark of pulmonary hypertension and overall heart disease. Whether loss of pulmonary vein cardiomyocytes contributes to other cardiovascular pathologies (24, 57), such as capillary rarefaction and aortic constriction, is not clear and merits further investigation. Nonetheless, it is very likely that the early loss of pulmonary vein sleeve cardiomyocytes shown in the current study contributes to or mediates many changes in cardiovascular health seen in these mice and perhaps in people born preterm.

Early life oxygen exposure is one of many known mediators of chronic lung disease in survivors of preterm birth. Hence, there has been great focus in human clinical trials and in experimental animal models to define the optimal dose of oxygen that does the least harm. In the current study, newborn mice were exposed to 100% oxygen between PND 0 and PND 4, which represents the saccular phase of lung development in mice and parallels the same period of lung development taking place when many infants are born preterm. Mice are then returned to room air during alveolar development, which also reflects how human alveolar development takes place in preterm infants who leave the neonatal intensive care unit breathing room air. Using this model, we previously identified two thresholds of oxygen that cause modest (60 and 80% oxygen) or severe (100% oxygen) alveolar simplification and reduced lung compliance (53). However, only the severe 100% oxygen exposure was sufficient to impair how the adult lung responds to a subsequent influenza A virus infection (9). Unlike this two-threshold dose-response to lung development, we found that exposure to 80 or 100% oxygen was required to deplete pulmonary vein cardiomyocytes. This suggests that the amount of oxygen required to alter lung function, promote cardiovascular disease, and alter the host response to viral infection is different. Because the dose of oxygen is different, the mechanism driving disease may also be different. Consistent with this idea, lung development and some aspects of the altered host response to influence A virus infection were preserved in transgenic Sftpc^{mice that overexpressed extracellular superoxide dismutase in alveolar type II cells (}8), but as shown in the current study, Sftpc^{mice still lost pulmonary vein cardiomyocytes when exposed to hyperoxia, indicating that ECSOD expressed by epithelial cells is not able to protect cardiac muscle. Presumably, this reflects failure of ECSOD to reach cardiac muscle cells. However, it is also conceivable that cardiac muscle is responding to ROS that is not easily detoxified by ECSOD.}

Although overexpression of ECSOD in alveolar type II cells did not prevent the oxygen-dependent loss of pulmonary vein cardiomyocytes, administration of mitoTEMPO did. This is consistent with another study showing how mitoTEMPO preserved alveolar septation and prevented right ventricular hypertrophy seen in newborn mice exposed to 75% oxygen (12). As a specific scavenger of mitochondrial superoxide, mitoTEMPO is likely detoxifying ROS that damage pulmonary vein cardiomyocytes. Although high levels of oxygen are toxic to most cells, we were unable to detect death of pulmonary vein cardiomyocytes by TUNEL staining or histology. Instead, neonatal hyperoxia decreased Ki67 staining in these cells, which remained low even after the mice were returned to room air. The loss of the proliferation marker Ki67 in pulmonary vein cardiomyocytes, therefore, suggests that these cells have exited the cell cycle during hyperoxia and failed to return during recovery in room air. The oxygen-dependent loss of these cells was confirmed by genetic lineage-mapping studies using Myh6^{; R26R}^{mice. On the basis of these findings, we conclude that neonatal hyperoxia depletes pulmonary vein cardiomocytes via growth suppression and not overt cell death. Why these cells remain growth inhibited, even when mice returned to room air, is not clear. Hyperoxia inhibits proliferation via a DNA damage-mediated activation of p53 (}21) and can inhibit proliferation of cardiomyocytes via production of toxic ROS (39). Persistent growth inhibition may, therefore, reflect a failure to effectively repair damaged DNA. Regardless of the cause in growth delay, failure to resume proliferation of pulmonary vein cardiomyocytes while the rest of the lung and heart continue to grow postnatally helps explain why these cells appear to be depleted later in life.

Transcriptional profiling of 8-wk-old mouse lungs also showed that neonatal hyperoxia increased expression of genes that regulate angiogenesis and endothelial cell phenotype. The biological relevance of these genes remains to be investigated. However, it is well known that angiogenesis is inhibited in newborn animals and humans exposed to high oxygen and that this contributes to the pathogenesis of bronchopulmonary dysplasia. Such changes are not seen in the current model wherein newborn mice are exposed briefly to hyperoxia and then recovered in room air. While some dilation of arterioles and veins is seen at 8 wk of age, microCT scans and expression of vascular-specific genes (PECAM, Tie1, Tie2, Flt1, Flt4, and thrombin receptor) suggest the pulmonary capillary network was intact at this age (54, 55). Capillary rarefaction and loss of BMP II signaling was, however, seen by 1 yr of age. In the current study, transcriptional profiling of adult lung exposed to neonatal hyperoxia revealed increased expression of Adamts1 that activates VEGF through proteolysis and the VEGF target genes Rcan1 or Rasgprp3. These changes may reflect an attempt to activate VEGF signaling needed to maintain or stimulate vascular growth in a lung that is slowly losing capillary endothelial cells. Increased expression of Apold1, a lipid-binding protein whose expression in endothelial cells is stimulated by hypoxia, suggests vascular endothelial cells may be under localized hypoxic conditions that cause them to die. Why increased expression of these angiogenic genes fails to prevent capillary rarefaction remain to be investigated.

In summary, transcriptional profiling helped us discover that neonatal hyperoxia depletes cardiomyocyte wrapping of the pulmonary vein and increases the expression of genes that regulate endothelial cell function. A graded loss of cells was seen, with the most profound change taking place within the lung, followed by the hilum, and then extending to the left atrium. Although intrapulmonary cardiomyocytes are only found in rodents, they contribute to the cardiomyocyte sleeve that extends from the left atrium to the hilum of all air-breathing species. Their loss in this mouse model may, therefore, reflect a similar change in striated muscle cells between the lung and heart of other mammals, including humans born preterm who are at risk of developing cardiovascular disease.

GRANTS

This work was funded, in part, by National Institutes of Health (NIH) Grants R01 HL-091968 (to M. A. O’Reilly), American Heart Grant-in-Aid 15GRNT23020024 (to E. D. Cohen), and a Wine Auction Pilot from the Cardiovascular Research Institute (to M. A. O’Reilly and G. Porter). NIH Center Grant P30 ES-001247 supported the animal inhalation facility and the tissue-processing core. The University of Rochester’s Department of Pediatrics provided financial support through the Perinatal and Pediatric Origins of Disease Program.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

M.Y., E.D.C., W.D., G.A.P., and M.A.O. conceived and designed research; M.Y., E.D.C., and W.D. performed experiments; M.Y., E.D.C., G.A.P., A.N.M., and M.A.O. analyzed data; M.Y., E.D.C., G.A.P., A.N.M., and M.A.O. interpreted results of experiments; M.Y. and A.N.M. prepared figures; M.Y., E.D.C., W.D., G.A.P., A.N.M., and M.A.O. edited and revised manuscript; M.Y., E.D.C., W.D., G.A.P., A.N.M., and M.A.O. approved final version of manuscript; M.A.O. drafted manuscript.

^{Department of Pediatrics, School of Medicine and Dentistry, The University of Rochester, Rochester, New York}

^{Department of Medicine, School of Medicine and Dentistry, The University of Rochester, Rochester, New York}

^{Biostatistics and Computational Biology, School of Medicine and Dentistry, The University of Rochester, Rochester, New York}

^{Corresponding author.}

Address for reprint requests and other correspondence: M. A. O’Reilly, Dept. of Pediatrics, Box 850, The Univ. of Rochester, School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642 (e-mail: ude.retsehcor.cmru@ylliero_leahcim).

Received 2017 Sep 7; Revised 2018 Jan 10; Accepted 2018 Jan 12.

Abstract

Supplemental oxygen given to preterm infants has been associated with permanently altering postnatal lung development. Now that these individuals are reaching adulthood, there is growing concern that early life oxygen exposure may also promote cardiovascular disease through poorly understood mechanisms. We previously reported that adult mice exposed to 100% oxygen between postnatal days 0 and 4 develop pulmonary hypertension, defined pathologically by capillary rarefaction, dilation of arterioles and veins, cardiac failure, and a reduced lifespan. Here, Affymetrix Gene Arrays are used to identify early transcriptional changes that take place in the lung before pulmonary capillary rarefaction. We discovered neonatal hyperoxia reduced expression of cardiac muscle genes, including those involved in contraction, calcium signaling, mitochondrial respiration, and vasodilation. Quantitative RT-PCR, immunohistochemistry, and genetic lineage mapping using Myh6^{; Rosa26}^{mice revealed this reflected loss of pulmonary vein cardiomyocytes. The greatest loss of cadiomyocytes was seen within the lung followed by a graded loss beginning at the hilum and extending into the left atrium. Loss of these cells was seen by 2 wk of age in mice exposed to ≥80% oxygen and was attributed, in part, to reduced proliferation. Administering mitoTEMPO, a scavenger of mitochondrial superoxide during neonatal hyperoxia prevented loss of these cells. Since pulmonary vein cardiomyocytes help pump oxygen-rich blood out of the lung, their early loss following neonatal hyperoxia may contribute to cardiovascular disease seen in these mice, and perhaps in people who were born preterm.}

Keywords: cardiomyocytes, hyperoxia, preterm, pulmonary hypertension

Abstract

F, forward primer; R, reverse primer.

Total RNA was isolated from three lungs of 8-wk-old mice exposed to room air or hyperoxia between PND 0 and PND 4. The RNA was hybridized to the mouse genome 430 2.0 array from Affymetrix. The mean average signal intensities for each probe and the relative fold change of hyperoxia to room air were determined. Using a 10% false discovery rate cutoff, we list in the table those genes whose expression was significantly decreased by neonatal hyperoxia.

ACKNOWLEDGMENTS

We thank Robert Gelein for maintaining the oxygen exposure facility, Daria Krenitsky for tissue processing and sectioning, and the University of Rochester’s Genomic Research Center for processing the Affymetrix Arrays.

ACKNOWLEDGMENTS

REFERENCES

References

1. Ambalavanan N, Cotten CM, Page GP, Carlo WA, Murray JC, Bhattacharya S, Mariani TJ, Cuna AC, Faye-Petersen OM, Kelly D, Higgins RD; Genomics and Cytokine Subcommittees of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Integrated genomic analyses in bronchopulmonary dysplasia. J Pediatr166: 531–7.e13, 2015. doi:10.1016/j.jpeds.2014.09.052. ] [
2. Auten RL, O’Reilly MA, Oury TD, Nozik-Grayck E, Whorton MH. Transgenic extracellular superoxide dismutase protects postnatal alveolar epithelial proliferation and development during hyperoxia. Am J Physiol Lung Cell Mol Physiol290: L32–L40, 2006. doi:10.1152/ajplung.00133.2005. ] [
3. Bachman NJ, Riggs PK, Siddiqui N, Makris GJ, Womack JE, Lomax MI. Structure of the human gene (COX6A2) for the heart/muscle isoform of cytochrome-c oxidase subunit VIa and its chromosomal location in humans, mice, and cattle. Genomics42: 146–151, 1997. doi:10.1006/geno.1997.4687. [] [[PubMed]
4. Barker DJ, Osmond C. Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet1: 1077–1081, 1986. doi:10.1016/S0140-6736(86)91340-1. [] [[PubMed]
5. Bensley JG, De Matteo R, Harding R, Black MJ. The effects of preterm birth and its antecedents on the cardiovascular system. Acta Obstet Gynecol Scand95: 652–663, 2016. doi:10.1111/aogs.12880. [] [[PubMed]
6. Berger J, Bhandari V. Animal models of bronchopulmonary dysplasia. The term mouse models. Am J Physiol Lung Cell Mol Physiol307: L936–L947, 2014. doi:10.1152/ajplung.00159.2014. ] [
7. Buczynski BW, Maduekwe ET, O’Reilly MA. The role of hyperoxia in the pathogenesis of experimental BPD. Semin Perinatol37: 69–78, 2013. doi:10.1053/j.semperi.2013.01.002. ] [
8. Buczynski BW, Yee M, Martin KC, Lawrence BP, O’Reilly MA. Neonatal hyperoxia alters the host response to influenza A virus infection in adult mice through multiple pathways. Am J Physiol Lung Cell Mol Physiol305: L282–L290, 2013. doi:10.1152/ajplung.00112.2013. ] [
9. Buczynski BW, Yee M, Paige Lawrence B, O’Reilly MA. Lung development and the host response to influenza A virus are altered by different doses of neonatal oxygen in mice. Am J Physiol Lung Cell Mol Physiol302: L1078–L1087, 2012. doi:10.1152/ajplung.00026.2012. ] [
10. Carvalho BS, Irizarry RA. A framework for oligonucleotide microarray preprocessing. Bioinformatics26: 2363–2367, 2010. doi:10.1093/bioinformatics/btq431. ] [
11. D’Angio CT, Ryan RM. Animal models of bronchopulmonary dysplasia. The preterm and term rabbit models. Am J Physiol Lung Cell Mol Physiol307: L959–L969, 2014. doi:10.1152/ajplung.00228.2014. [] [[PubMed]
12. Datta A, Kim GA, Taylor JM, Gugino SF, Farrow KN, Schumacker PT, Berkelhamer SK. Mouse lung development and NOX1 induction during hyperoxia are developmentally regulated and mitochondrial ROS dependent. Am J Physiol Lung Cell Mol Physiol309: L369–L377, 2015. doi:10.1152/ajplung.00176.2014. ] [
13. Doyle LW, Anderson P, Callanan C, Carse E, Casalaz D, Charlton MP, Davis N, Duff J, Ford G, Fraser S, Freezer N, Hayes M, Kaimakamis M, Kelly E, Opie G, Watkins A, Woods H, Yu V; Victorian Infant Collaborative Study Group . Respiratory function at age 8-9 years in extremely low birthweight/very preterm children born in Victoria in 1991-1992. Pediatr Pulmonol41: 570–576, 2006. doi:10.1002/ppul.20412. [] [[PubMed]
14. Doyle LW, Faber B, Callanan C, Freezer N, Ford GW, Davis NM. Bronchopulmonary dysplasia in very low birth weight subjects and lung function in late adolescence. Pediatrics118: 108–113, 2006. doi:10.1542/peds.2005-2522. [] [[PubMed]
15. Doyle LW, Faber B, Callanan C, Morley R. Blood pressure in late adolescence and very low birth weight. Pediatrics111: 252–257, 2003. doi:10.1542/peds.111.2.252. [] [[PubMed]
16. Eber E, Zach MS. Long-term sequelae of bronchopulmonary dysplasia (chronic lung disease of infancy). Thorax56: 317–323, 2001. doi:10.1136/thorax.56.4.317. ] [
17. Edstedt Bonamy AK, Bengtsson J, Nagy Z, De Keyzer H, Norman M. Preterm birth and maternal smoking in pregnancy are strong risk factors for aortic narrowing in adolescence. Acta Paediatr97: 1080–1085, 2008. doi:10.1111/j.1651-2227.2008.00890.x. [] [[PubMed]
18. Frey N, Luedde M, Katus HA. Mechanisms of disease: hypertrophic cardiomyopathy. Nat Rev Cardiol9: 91–100, 2012. doi:10.1038/nrcardio.2011.159. [] [[PubMed]
19. Fu Y, Nagy JA, Brown LF, Shih SC, Johnson PY, Chan CK, Dvorak HF, Wight TN. Proteolytic cleavage of versican and involvement of ADAMTS-1 in VEGF-A/VPF-induced pathological angiogenesis. J Histochem Cytochem59: 463–473, 2011. doi:10.1369/0022155411401748. ] [
20. Fulton RM, Hutchinson EC, Jones AM. Ventricular weight in cardiac hypertrophy. Br Heart J14: 413–420, 1952. doi:10.1136/hrt.14.3.413. ] [
21. Gehen SC, Staversky RJ, Bambara RA, Keng PC, O’Reilly MA. hSMG-1 and ATM sequentially and independently regulate the G1 checkpoint during oxidative stress. Oncogene27: 4065–4074, 2008. doi:10.1038/onc.2008.48. ] [
22. Goss KN, Kumari S, Tetri LH, Barton G, Braun RK, Hacker TA, Eldridge MW. Postnatal hyperoxia exposure durably impairs right ventricular function and mitochondrial biogenesis. Am J Respir Cell Mol Biol56: 609–619, 2017. doi:10.1165/rcmb.2016-0256OC. ] [
23. Hilgendorff A, O’Reilly MA. Bronchopulmonary dysplasia early changes leading to long-term consequences. Front Med (Lausanne)2: 2, 2015.
24. Huyard F, Yzydorczyk C, Castro MM, Cloutier A, Bertagnolli M, Sartelet H, Germain N, Comte B, Schulz R, DeBlois D, Nuyt AM. Remodeling of aorta extracellular matrix as a result of transient high oxygen exposure in newborn rats: implication for arterial rigidity and hypertension risk. PLoS One9: e92287, 2014. doi:10.1371/journal.pone.0092287. ] [
25. Jobe AH, Kallapur SG. Long-term consequences of oxygen therapy in the neonatal period. Semin Fetal Neonatal Med15: 230–235, 2010. doi:10.1016/j.siny.2010.03.007. ] [
26. Kim J, Okamoto H, Huang Z, Anguiano G, Chen S, Liu Q, Cavino K, Xin Y, Na E, Hamid R, Lee J, Zambrowicz B, Unger R, Murphy AJ, Xu Y, Yancopoulos GD, Li WH, Gromada J. Amino acid transporter Slc38a5 controls glucagon receptor inhibition-induced pancreatic α-cell hyperplasia in mice. Cell Metab25: 1348–1361.e8, 2017. doi:10.1016/j.cmet.2017.05.006. [] [[PubMed]
27. Kowalski RR, Beare R, Doyle LW, Smolich JJ, Cheung MM, Callanan C, Davis N, De Luca CR, Duff J, Hutchinson E, Kelly E, McDonald M, Molloy C, Wilson-Ching M, Anderson P, Burnett A, Carse E, Charlton MP, Hayes M, Kelly E, Opie G, Watkins A, Williamson A, Woods H, Roberts G, Robertson C, Wood S; Victorian Infant Collaborative Study Group . Elevated blood pressure with reduced left ventricular and aortic dimensions in adolescents born extremely preterm. J Pediatr172: 75–80.e2, 2016. doi:10.1016/j.jpeds.2016.01.020. [] [[PubMed]
28. Kramer AW Jr, Marks LS. The occurrence of cardiac muscle in the pulmonary veins of Rodenita. J Morphol117: 135–149, 1965. doi:10.1002/jmor.1051170202. [] [[PubMed]
29. Lee KY, Chuang HC, Chen TT, Liu WT, Su CL, Feng PH, Chiang LL, Bien MY, Ho SC. Proteoglycan 4 is a diagnostic biomarker for COPD. Int J Chron Obstruct Pulmon Dis10: 1999–2007, 2015. doi:10.2147/COPD.S90926. ] [
30. Lewandowski AJ, Augustine D, Lamata P, Davis EF, Lazdam M, Francis J, McCormick K, Wilkinson AR, Singhal A, Lucas A, Smith NP, Neubauer S, Leeson P. Preterm heart in adult life: cardiovascular magnetic resonance reveals distinct differences in left ventricular mass, geometry, and function. Circulation127: 197–206, 2013. doi:10.1161/CIRCULATIONAHA.112.126920. [] [[PubMed]
31. Lewandowski AJ, Davis EF, Yu G, Digby JE, Boardman H, Whitworth P, Singhal A, Lucas A, McCormick K, Shore AC, Leeson P. Elevated blood pressure in preterm-born offspring associates with a distinct antiangiogenic state and microvascular abnormalities in adult life. Hypertension65: 607–614, 2015. doi:10.1161/HYPERTENSIONAHA.114.04662. [] [[PubMed]
32. Li J, Yu KH, Oehlert J, Jeliffe-Pawlowski LL, Gould JB, Stevenson DK, Snyder M, Shaw GM, O’Brodovich HM. Exome sequencing of neonatal blood spots and the identification of genes implicated in bronchopulmonary dysplasia. Am J Respir Crit Care Med192: 589–596, 2015. doi:10.1164/rccm.201501-0168OC. ] [
33. Madurga A, Mizíková I, Ruiz-Camp J, Morty RE. Recent advances in late lung development and the pathogenesis of bronchopulmonary dysplasia. Am J Physiol Lung Cell Mol Physiol305: L893–L905, 2013. doi:10.1152/ajplung.00267.2013. [] [[PubMed]
34. Mamchur S, Mamchur I, Khomenko E, Kokov A, Bokhan N, Sherbinina D. Mechanical function of left atrium and pulmonary vein sleeves before and after their antrum isolation. Medicina (Kaunas)50: 353–359, 2014. doi:10.1016/j.medici.2014.11.008. [] [[PubMed]
35. Menon RT, Shrestha AK, Reynolds CL, Barrios R, Shivanna B. Long-term pulmonary and cardiovascular morbidities of neonatal hyperoxia exposure in mice. Int J Biochem Cell Biol94: 119–124, 2018. doi:10.1016/j.biocel.2017.12.001. ] [
36. Mueller-Hoecker J, Beitinger F, Fernandez B, Bahlmann O, Assmann G, Troidl C, Dimomeletis I, Kääb S, Deindl E. Of rodents and humans: a light microscopic and ultrastructural study on cardiomyocytes in pulmonary veins. Int J Med Sci5: 152–158, 2008. doi:10.7150/ijms.5.152. ] [
37. O’Reilly M, Thébaud B. Animal models of bronchopulmonary dysplasia. The term rat models. Am J Physiol Lung Cell Mol Physiol307: L948–L958, 2014. doi:10.1152/ajplung.00160.2014. [] [[PubMed]
38. Pircher A, Wellbrock J, Fiedler W, Heidegger I, Gunsilius E, Hilbe W. New antiangiogenic strategies beyond inhibition of vascular endothelial growth factor with special focus on axon guidance molecules. Oncology86: 46–52, 2014. doi:10.1159/000356871. [] [[PubMed]
39. Puente BN, Kimura W, Muralidhar SA, Moon J, Amatruda JF, Phelps KL, Grinsfelder D, Rothermel BA, Chen R, Garcia JA, Santos CX, Thet S, Mori E, Kinter MT, Rindler PM, Zacchigna S, Mukherjee S, Chen DJ, Mahmoud AI, Giacca M, Rabinovitch PS, Aroumougame A, Shah AM, Szweda LI, Sadek HA. The oxygen-rich postnatal environment induces cardiomyocyte cell-cycle arrest through DNA damage response. Cell157: 565–579, 2014. doi:10.1016/j.cell.2014.03.032. ] [
40. Randhawa PK, Rylova S, Heinz JY, Kiser S, Fried JH, Dunworth WP, Anderson AL, Barber AT, Chappell JC, Roberts DM, Bautch VL. The Ras activator RasGRP3 mediates diabetes-induced embryonic defects and affects endothelial cell migration. Circ Res108: 1199–1208, 2011. doi:10.1161/CIRCRESAHA.110.230888. ] [
41. Riper DV, Jayakumar L, Latchana N, Bhoiwala D, Mitchell AN, Valenti JW, Crawford DR. Regulation of vascular function by RCAN1 (ADAPT78). Arch Biochem Biophys472: 43–50, 2008. doi:10.1016/j.abb.2008.01.029. [] [[PubMed]
42. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res43: e47, 2015. doi:10.1093/nar/gkv007. ] [
43. Robin B, Kim YJ, Huth J, Klocksieben J, Torres M, Tepper RS, Castile RG, Solway J, Hershenson MB, Goldstein-Filbrun A. Pulmonary function in bronchopulmonary dysplasia. Pediatr Pulmonol37: 236–242, 2004. doi:10.1002/ppul.10424. [] [[PubMed]
44. Rogers LK, Velten M. Maternal inflammation, growth retardation, and preterm birth: insights into adult cardiovascular disease. Life Sci89: 417–421, 2011. doi:10.1016/j.lfs.2011.07.017. [] [[PubMed]
45. Sangwung P, Zhou G, Nayak L, Chan ER, Kumar S, Kang DW, Zhang R, Liao X, Lu Y, Sugi K, Fujioka H, Shi H, Lapping SD, Ghosh CC, Higgins SJ, Parikh SM, Jo H, Jain MK. KLF2 and KLF4 control endothelial identity and vascular integrity. JCI Insight2: e91700, 2017. doi:10.1172/jci.insight.91700. ] [
46. Smith VC, Zupancic JA, McCormick MC, Croen LA, Greene J, Escobar GJ, Richardson DK. Rehospitalization in the first year of life among infants with bronchopulmonary dysplasia. J Pediatr144: 799–803, 2004. doi:10.1016/j.jpeds.2004.03.026. [] [[PubMed]
47. Sohal DS, Nghiem M, Crackower MA, Witt SA, Kimball TR, Tymitz KM, Penninger JM, Molkentin JD. Temporally regulated and tissue-specific gene manipulations in the adult and embryonic heart using a tamoxifen-inducible Cre protein. Circ Res89: 20–25, 2001. doi:10.1161/hh1301.092687. [] [[PubMed]
48. Steen E, Bonamy AK, Norman M, Hellström-Westas L. Preterm birth may be a larger risk factor for increased blood pressure than intrauterine growth restriction. Acta Paediatr104: 1098–1103, 2015. doi:10.1111/apa.13095. [] [[PubMed]
49. Surate Solaligue DE, Rodríguez-Castillo JA, Ahlbrecht K, Morty RE. Recent advances in our understanding of the mechanisms of late lung development and bronchopulmonary dysplasia. Am J Physiol Lung Cell Mol Physiol313: L1101–L1153, 2017. doi:10.1152/ajplung.00343.2017. [] [[PubMed]
50. Wang H, St Julien KR, Stevenson DK, Hoffmann TJ, Witte JS, Lazzeroni LC, Krasnow MA, Quaintance CC, Oehlert JW, Jelliffe-Pawlowski LL, Gould JB, Shaw GM, O’Brodovich HM. A genome-wide association study (GWAS) for bronchopulmonary dysplasia. Pediatrics132: 290–297, 2013. doi:10.1542/peds.2013-0533. ] [
51. Weisman L. Populations at risk for developing respiratory syncytial virus and risk factors for respiratory syncytial virus severity: infants with predisposing conditions. Pediatr Infect Dis J22, Suppl: S33–S39, 2003. doi:10.1097/01.inf.0000053883.08663.e5. [] [[PubMed]
52. Wu SP, Cheng CM, Lanz RB, Wang T, Respress JL, Ather S, Chen W, Tsai SJ, Wehrens XH, Tsai MJ, Tsai SY. Atrial identity is determined by a COUP-TFII regulatory network. Dev Cell25: 417–426, 2013. doi:10.1016/j.devcel.2013.04.017. ] [
53. Yee M, Chess PR, McGrath-Morrow SA, Wang Z, Gelein R, Zhou R, Dean DA, Notter RH, O’Reilly MA. Neonatal oxygen adversely affects lung function in adult mice without altering surfactant composition or activity. Am J Physiol Lung Cell Mol Physiol297: L641–L649, 2009. doi:10.1152/ajplung.00023.2009. ] [
54. Yee M, Vitiello PF, Roper JM, Staversky RJ, Wright TW, McGrath-Morrow SA, Maniscalco WM, Finkelstein JN, O’Reilly MA. Type II epithelial cells are critical target for hyperoxia-mediated impairment of postnatal lung development. Am J Physiol Lung Cell Mol Physiol291: L1101–L1111, 2006. doi:10.1152/ajplung.00126.2006. [] [[PubMed]
55. Yee M, White RJ, Awad HA, Bates WA, McGrath-Morrow SA, O’Reilly MA. Neonatal hyperoxia causes pulmonary vascular disease and shortens life span in aging mice. Am J Pathol178: 2601–2610, 2011. doi:10.1016/j.ajpath.2011.02.010. ] [
56. Yoder BA, Coalson JJ. Animal models of bronchopulmonary dysplasia. The preterm baboon models. Am J Physiol Lung Cell Mol Physiol307: L970–L977, 2014. doi:10.1152/ajplung.00171.2014. ] [
57. Yzydorczyk C, Comte B, Cambonie G, Lavoie JC, Germain N, Ting Shun Y, Wolff J, Deschepper C, Touyz RM, Lelièvre-Pegorier M, Nuyt AM. Neonatal oxygen exposure in rats leads to cardiovascular and renal alterations in adulthood. Hypertension52: 889–895, 2008. doi:10.1161/HYPERTENSIONAHA.108.116251. [] [[PubMed]
58. Zeisel A, Muñoz-Manchado AB, Codeluppi S, Lönnerberg P, La Manno G, Juréus A, Marques S, Munguba H, He L, Betsholtz C, Rolny C, Castelo-Branco G, Hjerling-Leffler J, Linnarsson S. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science347: 1138–1142, 2015. doi:10.1126/science.aaa1934. [] [[PubMed]
59. Zhou L, Wang LM, Song HM, Shen YQ, Xu WJ, Xu JH, Liu Y, Yan WW, Jiang JF. Expression profiling analysis of hypoxic pulmonary disease. Genet Mol Res12: 4162–4170, 2013. doi:10.4238/2013.October.7.2. [] [[PubMed]
60. Zhou Y, Wu Q. Corin in natriuretic peptide processing and hypertension. Curr Hypertens Rep16: 415, 2014. doi:10.1007/s11906-013-0415-7. ] [