Neonatal hyperoxia inhibits proliferation and survival of atrial cardiomyocytes by suppressing fatty acid synthesis

Figures

• 

  Figure 1. Neonatal hyperoxia enlarges the left atria of mice but reduces the density of cardiomyocyte (CM) nuclei in the left atrial myocardia. (A) H&E-stained sections of hearts from P4 mice that were exposed to room air (left) or hyperoxia (right). Dotted lines outline the left atria. (B) Mean area of the left atria in sections of room air– and hyperoxia-exposed mice on P4. n = 4 mice per condition. (C) Numbers of DAPI-stained nuclei per mm2 of TNNT2+ myocardium. Room air, n= 4; hyperoxia, n = 5 mice per condition. (D and G) Sectioned hearts of P56 (D) and P365 (G) mice exposed to room air (right) or hyperoxia (right) from P0–P4 stained with Masson’s trichrome. (E and H) Graphs show mean area of the left atria in sections of room air– and hyperoxia-exposed mice on P56 (E) and P365 (F). (E) n = 4 per condition. (H) Room air, n = 3; hyperoxia, n = 4. (F and I) Graphs show numbers of DAPI-stained nuclei per mm2 of TNNT2+ myocardium in room air– and hyperoxia-exposed mice on P56 (F) and P365 (I). (F) n = 3 mice per condition. (I) n = 4 mice per condition. Bars in graphs show mean values, circles show individual data points, and error bars show SDs. The P values between data sets are from…

• 

  Figure 2. Hyperoxia reduces postnatal proliferation and survival of left atrial cardiomyocytes in neonatal mice. (A and B) Sections of left atrial appendages from P4 control (A) and hyperoxia-exposed (B) mice stained with phosphorylated Histone H3 (pHH3, green), cardiac troponin T (TNNT2, red), and DAPI (blue). Arrows point to pHH3+ cardiomyocytes. (C) Percentage of pHH3+ cardiomyocytes in P4 mice exposed to room air or hyperoxia. n = 4 mice per condition. (D and E) Sectioned left atrial appendages of P4 control (D) and hyperoxia-exposed (E) mice stained for the active, cleaved form of Caspase 3 (cl-Casp3), TNNT2, and DAPI. Arrows point to cl-Casp3+ cardiomyocytes. (F) Percentage of cl-Casp3–labeled CMs in the left atria of P4 hyperoxia-exposed and control mice. Room air, n = 4; hyperoxia, n = 5 mice per condition. Box plots show median, second quartiles, and third quartiles; whiskers indicate range, and circles represent individual data points. P values are results of unpaired 2-tailed t tests. Scale bars: 200 μm.

• 

  Figure 3. Neonatal hyperoxia causes mice to develop dilated heart failure later in life. (A–C) Graphs show mean E/A ratios (A), A-peak velocities (B), and E-peak velocities (C) of blood flow across the mitral valves of control and hyperoxia-exposed mice on P365. E/A ratios are plotted on a log10 scale. (D and E) Graphs show mean ejection fraction (D) and fractional shortening (E) of control and hyperoxia-exposed mice on P365. (F) Graph shows mean volume of the left ventricle during systole for PD365 control and hyperoxia-exposed mice. Arrows point to 2 hyperoxia-exposed mice with left ventricles that were drastically enlarged relative to controls and other hyperoxia-exposed mice. These 2 mice had the highest E/A ratios of all mice. (G–I) Box plots show the mean diastolic volume (G), stroke volume (H), and cardiac output (I) of the left ventricles in P365 hyperoxia-exposed and control mice after the 2 mice with enlarged left ventricular systolic volumes were excluded. Gray circles are values for excluded mice. Box plots show median, second quartiles, and third quartiles; whiskers show range, and circles show individual data points. F tests were used to determine if values for control and…

• 

  Figure 4. Neonatal hyperoxia suppresses genes needed for fatty acid synthesis in atria of mice. (A) Volcano plot of log2 fold changes versus log10 P values. Genes with FDR < 0.1 or annotation in the fatty acid metabolism are highlighted. Genes with FDR < 0.1 are marked by blue dots; genes involved in fatty acid metabolism are marked by green dots. (B) Gene ontology (GO) analysis of differentially expressed genes. (C–F) Results of qPCR for Slc25a1, Acly, Acaca, Acacb, Fasn, Scd1, Thrsp, and Elovl6 in atria (C and E) and ventricles (D and E) of hyperoxia-exposed and control mice on P4 (C and D) and P56 (E and F). (A and B) Room air, n = 3; hyperoxia, n = 4. (C–E) Room air, n = 5; hyperoxia, n = 6. (F) Slc25a1, Acly, Acaca, AcacB: room air, n = 5; hyperoxia, n = 5; Scd1, Fasn, Thrsp, Elovl6: room air, n = 4; hyperoxia, n = 4. (C–F) Box plots show median, second quartiles, and third quartiles; whiskers show range. Circles show values for individual control and hyperoxia-exposed mice, respectively. (C–F) *P < 0.05, **P < 0.01, ***P < 0.005, using unpaired 2-tailed t tests.

• 

  Figure 5. Neonatal hyperoxia represses fatty acid synthesis enzymes in murine atrial cardiomyocytes. (A, B, D, and E). Sections of left atrial appendages from P4 neonates exposed to room air (A and B) or hyperoxia (D and E) costained for FASN (red, A and B) or SCD1 (red, D and E) and TNNT2 (green, A, B, D, and E). Sections were also stained with DAPI to label nuclei (blue, A, B, D, and E). Arrows and arrowheads show TNNT2+ cardiomyocytes in left atrial appendage and LV, respectively. Scale bars: 200 μm. (C and F) Graphs show relative staining intensities for Fasn (C) and Scd1 (F) measured using NIH ImageJ 2.0/Fiji. (C and F) Room air, n = 5; hyperoxia, n = 5. Box plots show median values and inner quartiles. Whiskers show the range of values. Circles show values for individual room air– and hyperoxia-exposed mice, respectively. F tests were used to determine if samples had equal or unequal variances. P values are the results of unpaired 2-tailed t tests.

• 

  Figure 6. Suppression of fatty acid synthesis contributes to reduced HL-1 cell proliferation in hyperoxia. (A) qPCR for Fasn (left) and Scd1 (right) in HL-1 cells grown in room air or hyperoxia for 24 hours (white and blue, respectively) or 48 hours (yellow and red, respectively). (B) Numbers of HL-1 cells grown in room air (white circles) and hyperoxia (gray squares) for 48 hours relative to their density at 0 hours. (C and E) qPCR for Fasn (C) and Scd1 (E) in HL-1 cells transfected with Fasn and Scd1 siRNAs. Controls were transfected with nontargeting (NT) siRNA. (D and F) Numbers of HL-1 cells transfected with control siRNA (white circles in D and F) and Fasn (Gray squares in D) or Scd1 (Gray squares in F) siRNAs and grown for 48 hours in room air relative to 0 hrs. (G) HL-1 cells transfected with NT, Fasn, and Scd1 siRNAs, grown in room air for 22 hours and treated with EdU for 2 hours before staining. Graph shows percentages of EdU+ cells. (H) HL-1 cells transfected with NT, Fasn, and Scd1 siRNAs were grown in room air for 48 hours, stained with PI, and imaged. Graph shows numbers of PI+ NT, Fasn, and Scd1 siRNA transfected cells. (A, C, and E) n = 4 transfections per condition. Each…

• 

  Figure 7. Fasn and Scd1 overexpression increases HL-1 cell proliferation in hyperoxia. (A and C) Fasn (A) and Scd1 (C) mRNA in HL-1 cells 48 hours after transfection with empty vector or Fasn (A) and Scd1 (C) expression vectors. n = 4 transfections per condition. (B and D) Expansion of Fasn (B), Scd1 (D), and control transfected HL-1 cells over 24 hours. Markers are mean fold change in cell number for control (circles), Fasn (squares with dashed line, B), and Scd1 (squares with dashed line, D) grown in room air (white) or hyperoxia (gray). Lines and P values are linear regressions. n = 10 wells per time/condition. (E) Percentages of EdU+ control (white), Fasn (blue), and Scd1 (yellow) transfected HL-1 cells after 2 hours incubation in room air (left) and hyperoxia (right). n = 12 wells per condition. (F) Percentages of control (white), Fasn (blue), and Scd1 (yellow) transfected cells labeled after PI staining in room (left) and hyperoxia (right). n = 28 wells per condition. (G and H) HL-1 cell expansion in media with 0, 3.2, 16, and 80 μM palmitate-BSA in room air (G) or hyperoxia (H) for 36 hours. (I and J) Expansion of HL-1 cells in media containing 0, 25, 50, and 100 mM oleate-BSA and…

• 

  Figure 8. Hyperoxia suppresses fatty acid synthesis genes in left atrial tissue explanted from human infants. (A) Results of qPCR for Fasn and Scd1 in left atrial tissue explanted from human infants who died at birth due to anencephaly and exposed to room air or hyperoxia for 24 hours. n = 5 donors. (B) Sectioned explants exposed to room air (top) or hyperoxia (bottom) were stained for SCD1 (green), TNNT2 (red), and DAPI (blue). (C) Graph shows staining intensities for SCD1 in sections of control and hyperoxia-exposed mice determined using NIH ImageJ 2.0/Fiji. n = 4 donors. (D) Sections of explants exposed to room air (top) and hyperoxia (bottom) stained for the proliferation marker Ki67 (green), TNNT2 (red), and DAPI (blue). (E) Graph shows percentages of TNNT2-expressing cells with Ki67+ nuclei in explants exposed to room air or hyperoxia. n = 7 donors. (A, C, and E) Circles indicate individual values for explants of each donor. Boxes show medians and inner quartiles; whiskers represent the range. P values are the results of either single-sample 1-tailed t test and Wilcox test (A) or unpaired 2-tailed t tests (C and E). (B and D) Scale bars = 100 μm.