Heart and Circulatory Physiology

Hyperoxia-induced hypertrophy and ion channel remodeling in left ventricle

Siva K. Panguluri, Jared Tur, Jutaro Fukumoto, Wei Deng, Kevin B. Sneed, Narasaiah Kolliputi, Eric S. Bennett, Srinivas M. Tipparaju


Ventricular arrhythmias account for high mortality in cardiopulmonary patients in intensive care units. Cardiovascular alterations and molecular-level changes in response to the commonly used oxygen treatment remains unknown. In the present study we investigated cardiac hypertrophy and cardiac complications in mice subjected to hyperoxia. Results demonstrate that there is a significant increase in average heart weight to tibia length (22%) in mice subjected to hyperoxia treatment vs. normoxia. Functional assessment was performed in mice subjected to hyperoxic treatment, and results demonstrate impaired cardiac function with decreased cardiac output and heart rate. Staining of transverse cardiac sections clearly demonstrates an increase in the cross-sectional area from hyperoxic hearts compared with control hearts. Quantitative real-time RT-PCR and Western blot analysis indicated differential mRNA and protein expression levels between hyperoxia-treated and control left ventricles for ion channels including Kv4.2 (−2 ± 0.08), Kv2.1 (2.54 ± 0.48), and Scn5a (1.4 ± 0.07); chaperone KChIP2 (−1.7 ± 0.06); transcriptional factors such as GATA4 (−1.5 ± 0.05), Irx5 (5.6 ± 1.74), NFκB1 (4.17 ± 0.43); hypertrophy markers including MHC-6 (2.17 ± 0.36) and MHC-7 (4.62 ± 0.76); gap junction protein Gja1 (4.4 ± 0.8); and microRNA processing enzyme Drosha (4.6 ± 0.58). Taken together, the data presented here clearly indicate that hyperoxia induces left ventricular remodeling and hypertrophy and alters the expression of Kv4.2 and MHC6/7 in the heart.

  • ion channel regulation
  • hyperoxia
  • heart
  • hypertrophy
  • potassium channel
  • redox

patients in critical or intensive care units (ICU) with acute lung injury or cardiac disease are often administered 100% O2 for treatment. Recent studies indicate that hyperoxia induces cardiac injury due to dysfunctional lung and compromised pulmonary functioning (37), even though the exact nature of this problem remains unknown. Here, we evaluated changes in expression of the ion channel and key transcriptional factors in the heart that occur with hyperoxia and likely play a role in cardiovascular remodeling.

Potassium channels and their auxiliary subunits such as potassium channel interacting protein-2 (KChIP2) are abundantly expressed in the heart (5, 7, 35). It is established that the potassium channels Kv4.2 and Kv1.5 are responsive to oxygen changes (29, 39). In the present study, we investigate whether hyperoxia alters expression of the transcription factors Irx5 and Mef2c, which are implicated to play a direct role in regulating Kv4.2 expression (7, 15, 22). Cardiac-specific markers used to identify hypertrophy and transcriptional changes (9, 22) were also evaluated by assessing myosin heavy chain-6, and -7 (MHC6, MHC7), zinc finger transcription factor (GATA4), histone-lysine N-methyltransferase (Ezh2), and Six-1 expression levels with hyperoxia.

Key inflammatory mediators such as TNFα and NFκB are central regulators or master switches for many pathological processes (10, 20, 30). Recent evidence indicates that NFκB regulates KChIP2, which in turn regulates Kv4.2 expression (26). Therefore we assessed the levels of Kv4.2, KChIP2, and NFκB in the mouse heart subjected to hyperoxia. We hypothesized that hyperoxia induces cardiac hypertrophy and alters Kv4.2, NFκB, and KChIP2 in the heart. Physiological and biochemical assessment was performed using echocardiography, histochemistry, as well as qRT-PCR and Western blotting for expression of key components. Molecular targets such as potassium channels (Kv4.2, Kv1.4, Kv1.5, Kv2.1), sodium channel (Scn1a, Scn5b), auxiliary subunit (KChIP2), transcription factors (Irx5, Mef2c), and hypertrophic markers (MHC-6, MHC-7, and GATA-4) were altered in hyperoxic mouse hearts.



Young adult male mice (C57BL/6 strain) 8–9 wk of age obtained from Harlan Laboratories (Indianapolis, IN) were randomly assigned and subjected to room air or hyperoxia and were used as normoxia or hyperoxia groups, respectively. For hyperoxia treatment, mice were exposed to hyperoxia (100% O2) for 72 h and euthanized right after the exposure period ended. For euthanasia, the mice were injected with euthasol (50 mg/kg ip). All animal experiments conform to the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of South Florida (Tampa, FL). Mice were provided with uninterrupted access to food and water ad libitum and maintained on a 12:12-h light/dark cycle.

Oxygen exposure.

Mice, aged 8–9 wk old, were placed in cages in an airtight chamber (50 × 50 × 30 cm) with access to food and water ad libitum and exposed to 100% oxygen for 72 h or room air. The oxygen concentration in the chamber was monitored with an oxygen analyzer (Vascular Technology, Chelmsford, MA) as described previously (21). Mice were euthanized and thoracotomy was performed for collection of blood from heart. Thereafter hearts were excised and snap frozen in liquid N2, and stored at −80°C. Hearts were dissected into different regions such as right and left ventricle, apex, atrium, and septum and stored separately at −80°C until further use.


Transthoracic echocardiography was performed using Vevo770 Ultrasonograph (VisualSonics) equipped with a 30-MHz transducer. Cardiac function was measured from normoxic or hyperoxic mice, using previously published procedures for echocardiographic measurements (3). Briefly, the mice were anesthetized with 1–1.5% isoflurane and the body temperature was maintained at 37°C. Two-dimensional (2-D) mode in the parasternal long axis and parasternal short axis at the midpapillary muscle level were imaged. From this parasternal short axis view, the 2-D guided M-mode across the anterior wall and posterior wall were recorded. Left ventricular anterior wall (LVAW), left ventricular interior diameter (LVID), and left ventricular posterior wall thickness (LVPW) at systole (s) and diastole (d) were measured. Ejection fraction (EF%) was calculated as (EDV − ESV)/EDV × 100%, where EDV and ESV are end-diastolic and end-systolic volume, respectively. Fractional shortening (FS%) was calculated by (LVIDd − LVIDs)/LVIDd × 100%. Hyperoxic mice developed severe bradycardia after 72 h treatment; thus the echocardiography procedures in this group did not require use of anesthesia.

Quantitative real-time-PCR (qRT-PCR).

Total RNA was isolated from control and hyperoxia-treated hearts using the Exiqon miRCURY RNA Isolation kit (Exiqon, Woburn, MA), and all the procedures were followed according to manufacturer's protocol. The total RNA concentration was determined by NanoDrop spectrophotometer, and RNA quality was assessed by 18S/28S ribosomal peak intensity.

Ion channel genes, transcriptional factors, hypertrophic markers, and other mediators (Table 1) that are implicated in ion channel regulation were selected to study relative expressions by qRT-PCR using a MyiQ single color real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA). All the qRT-PCR procedures were performed according to our previously published protocol (28). Briefly, qRT-PCR was performed in a total of 20 μl reaction volume containing 1 μl of cDNA, 1 μl of each of forward and reverse sequence specific primers (from 10 μM primer stock), 10 μl of supermix (Bio-Rad Laboratories), and 7 μl of nuclease-free water. All the qRT-PCR reactions were performed using the following reaction conditions: initial incubation of 95°C for 3 min followed by 45 cycles of 95°C for 10 s, 60°C for 20 s, and 72°C for 30 s. A fluorescence reading determined the extent of amplification at the end of each cycle. The expression of mouse hypoxanthine phosphoribosyl transferase (HPRT) was used as an internal control. For all the candidate genes, the quantities of the mRNA expression relative to HRPT mRNA levels were obtained.

View this table:
Table 1.

Sequence of primers used for quantitative real-time RT-PCR

Western blotting.

Protein extracts for Western blotting were from both normoxic and hyperoxic mouse hearts. Left and right ventricles were homogenized in tissue protein extraction reagent (Thermo Fisher Scientific, Waltham, MA) along with 10 mM DDT, 10 mM protease inhibitor, and phosphatase inhibitor cocktail II (Sigma-Aldrich, St. Louis, MO). The mixture was centrifuged at 14,000 rpm for 5 min at 4°C and the supernatant collected and stored at −80°C. The proteins were denatured with SDS-loading buffer (Bio Rad Laboratories) containing β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO) at 90°C for 10 min prior to loading on SDS-PAGE. Equivalent amounts of protein were loaded and resolved as determined by Pierce 660 assay (Thermo Fisher Scientific, Waltham, MA) on 4–20% gradient SDS polyacrylamide gels (Bio-Rad Laboratories). Proteins were detected with a dilution of primary antibody as follows: 1:200 (Kv1.4, MHC α), 1:10 (MHC β), 1:250 (IRX-5), 1:500 (GAPDH), and 1:1,000 (Kv4.2, KChIP2, NFκB) mono- and polyclonal primary antibodies. Primary antibodies such as Kv1.4 (AB5926), Kv4.2 (07–491), MHC β (MAB1548), GAPDH (MAB-374), NFkB total (04–234), and phospho-NF-k-B p65 (04–1000) were obtained from Millipore (Billerica, MA); IRX-5 (AB56681), MHC α (AB50967) and KChIP2 (AB99041) are from Abcam (Cambridge, MA). For all the blots blocking was performed with 5% milk done for 1 h followed by overnight incubation with primary antibody in 5% nonfat milk at 4°C. All the procedures were followed according to the standard techniques in our lab (27). Immunoblots were quantified using ImageJ software, and band densities were normalized with the corresponding GAPDH band densities.

Histological analysis.

Normoxic and hyperoxic hearts were cryosectioned into 30-μm slices using a Microm HM 505 N cryostat (Southeast Pathology instrument service, Charleston, SC) and stained with hematoxylin-eosin (H and E) according to the manufacturer's protocol (BBC Biochemicals, Mount Vernon, WA). Slides were stained with a gradient of 100–70% alcohol followed by hematoxylin and acid-wash. Slides were then stained with Blueing solution and BBC Special Eosin and BBC S2 Histo and final washing with xylene (Thermo Fisher Scientific, Waltham, MA). Finally, coverslips were mounted using DPX mounting solution (Sigma-Aldrich, St. Louis, MO). Sections were visualized utilizing a MIRAX Scanner (Zeiss, Germany) and volumetric quantification of the entire section was measured in microns squared using MIRAX Scan software (Zeiss, Germany).

TNF-α measurements.

A total of 20 μg of protein was taken from each sample and TNF-α was quantified by using an ELISA kit (Invitrogen, Camarillo, CA). All the procedures were followed according to manufacturer's protocol. Briefly, standards and unknown samples were loaded (50 μl) in triplicates and incubated with mouse TNF-α biotin conjugate solution (50 μl) for 90 min at room temperature. Wells were aspirated and washed four times, then incubated with streptavidin-HRP (100 μl) for 30 min at room temperature. Wells were aspirated and washed four times again, then incubated with stabilized chromogen (100 μl) for 20–25 min at room temperature in the dark. Finally stop solution was added (100 μl) and TNF-α was quantified at 450 nm in a microplate reader (Biotek, Winooski, VT).

Network analysis.

Differentially expressed genes with P ≤ 0.05 from qRT-PCR data were selected for network analysis using GeneGo software (Thomson Reuters). Based on the existing literature, MetaCore pathways analysis identified the pathways from its library of canonical pathways that were most significant to the data set. The significance of the association between the data set and the pathway network was measured by a ratio of the number of genes from the data set that map to the pathway divided by the total number of genes that map to the canonical pathway. False discovery rate (FDR) (P ≤ 0.05) was considered as significant, and top cellular networks, toxicology pathways and molecular function networks were generated using the database.

Data analysis.

Data were analyzed using Student's t-tests. Analysis was performed with investigator blinded to treatment groups, and results were expressed as means ± SE. The Student's t-test was used to compare quantitative data populations with normal distributions and equal variance. A value of P < 0.05 was considered statistically significant unless otherwise specified.


Physical parameters.

Exposure of mice to hyperoxia conditions (100% O2) for 72 h showed significant decrease in the body weight (10.8%, P ≤ 0.005) (Fig. 1A), while whole heart weights significantly increased (22%, P ≤ 0.007) (Fig. 1, B and C). Representative whole heart pictures from hyperoxia and normoxia groups show that hyperoxia causes hypertrophy compared with normoxia (Fig. 1B), consistent with the heart weight measurements and suggest remodeling of the heart induced by hyperoxia exposure. We also performed H and E staining of 30-μm heart sections (TS), and overall cross-sectional area was measured using MIRAX scan software (Gottingen, Germany). Our data clearly show that the area of hyperoxic heart sections was significantly increased (16%) compared with the normoxia group (Fig. 1D).

Fig. 1.

Physical parameters of heart. A: the normalized body weights of normoxia and hyperoxia mice were plotted as means ± SE (n = 14). *P ≤ 0.05. B: whole heart. Representative pictorial depiction is provided for the difference in size of the whole heart between hyperoxia and normoxia groups. C: normalized heart weight of hyperoxia- and normoxia-treated mice expressed in g/cm on y-axis. All the values plotted are means ± SE (n = 13 for normoxic and n = 9 for hyperoxic). *P ≤ 0.05. D: hematoxylin and eosin (H and E) staining of normoxia- and hyperoxia-treated hearts. Data represented are means ± SE (n = 3). *P ≤ 0.05.

Functional assessment by echocardiography.

We investigated if the changes in cardiac physical parameters measured above can be corroborated using echocardiography. The echocardiographic data show significant reduction of heart rate (37.3%) and cardiac output (49.5%) in hyperoxia-treated mice compared with normoxia. Additionally, we also observed marginal reduction (5%) of ejection fraction (EF%) and fractional shortening (8%) in hyperoxia-treated mice (not significant) (Table 2).

View this table:
Table 2.

Echo analysis of hyperoxic hearts

Molecular markers and hypertrophy detection.

Because of the increased heart size suggesting hypertrophy of the hyperoxia-treated mice, we investigated the mRNA expression levels of hypertrophic markers myosin heavy chain-α (MHC-α/MHC-6) and myosin heavy chain-β (MHC-β/MHC-7). The quantitative real-time RT-PCR data showed that there is a significant increase in expression of MHC-6 (2.17 ± 0.36) and MHC-7 (4.62 ± 0.76) in the left ventricle of the hyperoxia group compared with the normoxia group (Fig. 2A). These data reveal that the classical markers of hypertrophy are upregulated, further confirming hyperoxia-induced left ventricular hypertrophy.

Fig. 2.

Transcriptional changes in hypertrophic and ion channel gene expression. Left ventricle of normoxic and hyperoxic groups were used for real-time qRT-PCR with hypertrophic markers MHC-6 or MHC-α and MHC-7 or MHC-β (A) and ion channel genes including Kv4.2, Kv2.1, Kv4.3, Kcnj2, Scn5a and their related proteins such as KChIP2, Irx5 and Gja1 (B). Values shown on y-axis are means ± SE (n = 3) and are normalized with housekeeping gene HPRT expression level. The numbers above the bar represent the fold values compared with normoxia group and all the genes are with P ≤ 0.05.

Ion channel expression profile.

Ion channel remodeling occurs during hypertrophy and myopathies. Therefore, we investigated the expression profiles of potassium and sodium channel genes. QRT-PCR data showed that there is significant decrease in the expression of Kv4.2 (−2 ± 0.08), whereas genes including Kv2.1 (2.54 ± 0.48), Kv4.3 (1.98 ± 0.23), Kcnj2 (4 ± 1.49), and Scn5a (1.4 ± 0.07) were significantly increased in hearts from mice exposed to hyperoxia (Fig. 2B). However, no significant changes in the expression levels of ion channels such as Kv1.4, Kv1.5, and Scn1b were observed (Table 3). The qRT-PCR data also showed that KChIP2, which is a key potassium channel interacting protein for Kv4.2, is significantly downregulated (−1.7 ± 0.06) in left ventricle from hyperoxia group (Fig. 2B). The role of Iroquois family protein, Irx5, was also assessed and our measurements indicated that Irx5 was significantly upregulated (5.6 ± 1.74) suggesting that an increase in Irx5 expression is likely to repress Kv4.2 transcription (Fig. 2B). Gap junction protein Gja1 plays an important role in the conduction and propagation of electrical impulses in the heart; we examined its expression and found that the transcript levels of Gja1 (4.4 ± 0.8) were significantly increased in left ventricle from hyperoxia group (Fig. 2B).

View this table:
Table 3.

Quantitative real-time RT-PCR data of hyperoxia-treated left ventricle compared with normoxia group

Transcriptional factors profile.

Our data clearly identify a decrease in Kv4.2 expression. We next investigated the plausible roles of transcription factors and other key mediators that might be regulating expression of this major ion channel. We found that the key transcriptional factors such as myocyte enhancing factor 2 (Mef2c) (−1.8 ± 0.09), GATA4 (−1.5 ± 0.05), PPARγ (−1.5 ± 0.09), and Hif-α (−1.4 ± 0.06) were significantly downregulated in left ventricle of hyperoxia-treated mice (Fig. 3). We next evaluated the inflammatory marker NFκB1 (4.17 ± 0.43), and serum response factor or Srf (10 ± 0.98) in hyperoxic group and observed that the levels of these mediators were upregulated significantly in hyperoxia group (Fig. 3). No significant changes in the expression of GATA6, Pitx2c, NFκB2, Ezh-2, Six-1, and MMP9 were observed in the hyperoxia group compared with normoxia (Table 3).

Fig. 3.

Gene expression profile of transcriptional factors. qRT-PCR expression profile of transcriptional factors such as Mef2c, GATA4, NFκB1, PPARγ, Srf, and Hif1α. Values shown on y-axis are means ± SE (n = 3) and are normalized with housekeeping gene HPRT expression level. The numbers above the bar represents the fold values compared with normoxia group, and all the genes are with P ≤ 0.05.

MicroRNA processing and gene profile.

Recent advances in microRNA (miRNA) support the idea that they can differentially regulate the expression of ion channel genes and various transcriptional factors (26, 41). This intrigued us to examine the expression profile of miRNA processing genes such as Dicer, Drosha. and Exportin-5. The mRNA expression level of Drosha is found to be elevated 4.6-fold and is altered significantly (P < 0.05) in hyperoxia group compared with normoxia, whereas no significant differences in Dicer and Exportin-5 expression levels were detected (Table 3).

Protein abundances.

More than 50% of the mRNAs expressed are predicted to be regulated posttranscriptionally by microRNAs (6) and upregulation of miRNA processing enzyme Drosha in our data suggest the possible degradation of some of the genes differentially expressed in the present study. Western blot analysis showed that protein expression levels of Kv4.2 and KChIP2 were downregulated, whereas Irx5 was upregulated in the hyperoxic hearts (Fig. 4). We also observed a significant increase in protein expression of myosin heavy chain-β (MHC-7) which is a marker for hypertrophy (Fig. 4). These data clearly indicate that Kv4.2 expression is inversely correlated to Irx5 expression. In addition, a significant increase in expression of transcriptional factor NFκB in hyperoxic left ventricle was also observed (Fig. 4). Many important transcriptional factors including NFκB can be activated by phosphorylation. Therefore, we investigated whether the changes in activity of these transcriptional factors occurs also at the posttranslational level via phosphorylation events. Thus we tested the protein targets for phosphorylation using their phosphospecific antibodies. Our data show an elevation of phospho-NFκB (p65) in left ventricle of hyperoxic mice (Fig. 4), suggesting that phospho-NFκB plays a vital role as key mediator for downstream events. Based on previous information that in the normal heart there is heterogeneous expression of ion channels and other transcriptional regulators, we investigated whether the changes in expression of these proteins are specific to left ventricle. For this we analyzed protein expression of the genes in right ventricle and our data show that similar to left ventricle, there is significant reduction of Kv4.2 and KChIP2 expression, and significant elevation of Irx5 in right ventricle of hyperoxic mice (Fig. 5). In contrast, we did not find any significant difference in expression of other proteins including Kv1.4, MHC-7, and NFκB (Fig. 5). We also observed no difference in expression levels of phospho-NFκB in right ventricle of hyperoxic mice indicating ventricular heterogeneity in the observed protein expression/activation levels. Comparison of left and right ventricular regions clearly suggests that the heart is remodeled and the mechanisms are differentially regulated in a region-specific manner.

Fig. 4.

Translational changes in ion channels, transcriptional factors and hypertrophic markers in left ventricle (LV). Western blot analysis from hyperoxic or normoxic group developed with anti-Kv4.2, anti-Irx5, anti-Gapdh, anti-Kv1.4, anti-MHC7, anti-NFκB, anti-KChIP2 and anti-phosphorylated NFκB antibodies (A); lane M represents prestained protein ladder. The band intensities from all the blots were measured using ImageJ software and were normalized with their corresponding Gapdh band intensities (B). Values shown on y-axis are means ± SE (n = 3) of normalized band intensities. The numbers above the bar represent the fold values compared with normoxia group. *P ≤ 0.05.

Fig. 5.

Western analysis of ion channels, transcriptional factors, and hypertrophic markers in right ventricle (RV). Western blot analysis from hyperoxic or normoxic group with anti-Kv4.2, anti-Irx5, anti-Gapdh, anti-Kv1.4, anti-MHC7, anti-NFκB, anti-KChIP2 and anti-phosphorylated NFκB antibodies (A); lane M represents prestained protein ladder. The band intensities from all the blots were measured using ImageJ software and were normalized with their corresponding Gapdh band intensities (B). The values shown on y-axis are means ± SE (n = 3) of normalized band intensities. The numbers above the bar represents the fold values compared with normoxia group. *P ≤ 0.05.

NFκB expression is regulated by elevated levels of proinflammatory cytokine TNF-α (4) and TNF-α is reported to cause ion channel remodeling via downregulation of Kv4.2 and Kchip2 (10, 30). Therefore, we investigated if NFκB changes are due to elevated levels of TNF-α. Our data showed a significant decrease in intracellular TNF-α expression in left ventricle of hyperoxic mice compared with normoxia controls (Fig. 6). However, no significant change in intracellular TNF-α expression in right ventricle was observed.

Fig. 6.

Intracellular TNF-α expression in hyperoxic mice. The intracellular TNF-α expression was measured using ELISA, and values were expressed in picograms of TNF-α per microgram of total protein. The values on y-axis are means ± SE (n = 3). The numbers above the bar represent the fold values compared with normoxia group. *P ≤ 0.05.


In this study we demonstrate that mice subjected to hyperoxia treatment for 72 h undergo hypertrophy. The increase in heart weight is caused primarily by left ventricular hypertrophy confirmed by elevation in hypertrophic markers (MHC-6 and -7) and cross-sectional area. We also found dysregulation of oxygen-sensitive potassium channel Kv4.2 along with its repressor Irx5 in the ventricle. Transcriptional factors such as Mef2c, GATA4, SRF, Hif1α, and PPARγ were differentially altered in left ventricle. We observed a major change in both expression as well as activity levels of the inflammatory master regulator NFκB in left ventricle, which is independent of TNFα activation. Thus in this study, we show that hyperoxia induces cardiac hypertrophy in which key genes are dysregulated in left ventricle.

Functional assessment in the hyperoxic heart.

Hemodynamic changes such as decrease in heart rate, cardiac output and stroke volume were reported in hyperoxic breathing (2, 12, 16, 24, 34). The lowering of heart rate in hyperoxic condition may be triggered by increased parasympathetic activity (24). Here, we report significant reduction in heart rate and cardiac output in hyperoxia-treated mice in our study (Table 2), which suggests acute hemodynamic effects with hyperoxic injury. Although we observed a slight reduction of ejection fraction and fractional shortening in hyperoxia-treated mice, they still fall under normal physiological range; a similar pattern was noticed with other echocardiographic parameters, suggesting normal functioning of these hearts post hyperoxia treatment (at least acute treatment). However, the physical parameters of these hearts showed a significant increase in size, cross sectional area, and weight after hyperoxia treatment (Fig. 1), suggesting that the hearts may be undergoing functional compensation while still becoming hypertrophic and, as discussed later, experiencing significant remodeling. Alternatively, given the observed severe bradycardia in the hyperoxic mice suggests cardiac dysfunction, perhaps the echocardiography procedure and/or the use of anesthesia is limited in its sensitivity to detect the changes in cardiac function the occur with acute hyperoxia. Also, previous reports evaluate the involvement of pulmonary hypertension playing an important role in right ventricular hypertrophy with 14- to 15-day O2 exposure (14, 17); however, in the present study we did not evaluate the pulmonary function and blood pressure changes related to O2 exposure for 3 days. We found hypertrophy in the heart as noted by changes in the overall weight of the heart after hyperoxia exposure; however, regional specific changes (RV, LV, and septum) were not measured in the present study.

Hyperoxia induces left ventricular hypertrophy.

The qRT-PCR data showed a significant increase in mRNA expression of hypertrophic markers myosin heavy chain-6 and -7 with hyperoxia (Fig. 2A). (22, 25). Earlier reports show elevated expression of MHC-6 and reduction of MHC-7 in O2-treated mice (36). In contrasts to this report, our study showed a significant increase in both MHC-6 and -7 expression at mRNA as well as protein expression level (Figs. 2A and 4), suggesting that the neonatal mice respond differently to O2 exposure than the adult mice. We also investigated if the hypertrophic response is confined to the left ventricle or can also be found in right ventricle; by using Western blotting we probed at key mediators. Our data showed no significant increase in MHC-7 protein levels in right ventricle of hyperoxic mice compared with normoxic group (Fig. 5). These data suggest that the thicker left ventricular wall likely caused by the hypertrophic response is primarily responsible for the observed increased heart weight, and such a response may be useful to consider and integrate into a risk assessment strategy for hyperoxia-induced cardiomyopathy.

Hyperoxia dysregulates ion channel expression.

The majority of cardiomyopathies are characterized by electrical remodeling (7, 25, 30), leading to left ventricular hypertrophy and heart failure. Here, the data suggest some ion channel remodeling occurs in the hyperoxic hearts. Mechanical and electrophysiological dysfunctions in heart failure are often observed with reduction of Kv4.2 expression and increased Kv1.4 expression (25, 31). Our data showed a significant reduction in Kv4.2 transcripts as well as protein expression (Figs. 2B and 4), but no significant change in Kv1.4 at both transcriptional as well as translational levels (Table 3 and Fig. 4). In contrast to previous studies, we found a significant increase in transcripts of Kv2.1 and 4.3 (Fig. 2B). These data indicate that the decrease in Kv4.2 levels may be compensated by the increase in Kv4.3 mRNA expression to a certain extent as they are the molecular correlates regulating transient outward currents (Itofast). However, 97% of murine Itofast is generated by Kv4.2 and only about 3% is contributed by Kv4.3.

Hyperoxia induce differential regulation of key transcriptional factors.

Cardiac hypertrophy and failure are associated with decreased Kv4.2 and 4.3 expression levels (11, 19), and Kv channel-interacting proteins (KChIP) are known to interact with these Kv channels (1, 38). KChIP2 is most abundant and highly expressed KChiP in heart tissue and decreased levels of KChiP2 have been previously reported in hypertrophy and heart failure (23, 32). Recent studies by Jin et al. (18) showed that gene transfer of KChIP2 in neonatal cardiomyocytes increased Kv4.2 and 4.3 protein expression and in vivo transfer of this gene in adult rats significantly reduced left ventricular hypertrophy. As observed in our study, Kv4.2 was significantly reduced in hyperoxia-induced mice both at transcriptional as well as translational levels; therefore we hypothesized that KChIP2 regulates Kv4.2 in this model and expected a decrease in the KChIP2 expression. Our experimental results are consistent with the hypothesis: qRT-PCR data showed a significant reduction of KChIP2 transcripts in left ventricle of hyperoxia-treated mice (Fig. 2B). We also observed a significant decrease in KChIP2 at protein expression levels in both left and right ventricle (Figs. 4 and 5).

Recent investigations also showed that homeobox transcriptional factor Iroquois protein 5 (Irx5) differentially expressed in a gradient across the left ventricle of heart (8, 33), and deletion of this gene caused arrhythmias. Therefore we investigated whether expression of Kv4.2 transcription repression was increased in the hyperoxic heart. The qRT-PCR data showed a significant increase in Irx5 transcript and protein levels in hyperoxia-treated mice (Figs. 2B and 4) in left ventricle. We also examined protein levels of Irx5 in the right ventricle, which showed a significant increase in its expression (Fig. 5) in hyperoxia-treated mice. Therefore, our data suggest that in the hyperoxic heart Irx5 may be functioning as a transcriptional repressor of Kv4.2.

Elevated expression and activation of the ubiquitous transcriptional factor nuclear factor kappa B (NFκB) is reported in cardiac hypertrophy and heart diseases (13, 40). In a recent study, activation of NFκB decreased Ito,f by decreasing KChIP2 expression and inhibition of its activity increased Ito,f and KChiP2 expression (26). Consistent with the observed decreased KChiP2 expression, we observed an increased in NFκB with hyperoxia. Furthermore we investigated if the increased NFκB is phosphorylated (p65); results clearly demonstrated elevated levels of activated NFκB in left ventricle of hyperoxia-treated mice. We also investigated if this elevated expression and activity is common for both right and left ventricle by Western blotting and our data suggest no change in expression or activity of NFκB in right ventricle of hyperoxia mice (Fig. 5). This suggests that the left ventricle is significantly affected by hyperoxia.

Increasing levels of NFκB and its activation suggests the possible induction of a cytokine tumor necrosis factor-α (TNF-α). Our previous studies showed that induction of myoblasts (C2C12) with TNF-α increased both expression and activity of NFκB (4) and previous reports suggest the possibility that TNF-α can regulate Kv4.2 and KChIP2 expression (20). We found significant reduction of intracellular TNF-α concentration in left ventricle of hyperoxia-treated mice, suggesting that there may be other mechanisms, which are independent of TNF-α, in regulating NFκB activity and Kv4.2 expression in hyperoxia-induced hypertrophy.

Overall, we report differential expression of key ion channel gene, hypertrophic markers and transcriptional factors in this study. Interestingly, we found that expression of the microRNA, Drosha, was increased in left ventricle of the mice treated with hyperoxia, suggesting the possible miRNA processing of many genes and pathways posttranscriptionally with hyperoxic injury.

In order to identify the potential pathways and networks regulated by the differentially expressed genes in left ventricle from the present study, we utilized in silico analysis of these selected genes using GeneGo. The in silico analysis showed that the possible regulation of Kv4.2 by IRX5, KCNIP (KChIP2) and GATA4, whereas Drosha is regulated by NFκB (Fig. 7). Data also showed that cardiac hypertrophy, angiotensin II signaling mediated hypertrophy, oxytocin receptor signaling, Ca2+ dependent signaling, immune response to IL-6, and Mef2 mediated immune response are among the top 15 regulated cellular pathways with P ≤ 0.05 (Fig. 8). Ion channel activity, transcriptional regulation, cardiac development, proliferation signaling, and immune response are among the top 15 molecular functions (Fig. 8) This analysis is consistent with the experimental results discussed in this study.

Fig. 7.

Interacting gene network of differentially expressed genes in left ventricle of hyperoxic mice. Genes and their expression levels from qRT-PCR data with P ≤ 0.05 were considered for network analysis using GeneGo software. The software uses the curated publications to build the relations between the genes. Upregulated genes are marked with red circles; downregulated with blue circles. The checkerboard color indicates mixed expression for the gene between files or between multiple tags for the same gene.

Fig. 8.

In silico analysis of the reported genes. Top 5 cellular pathways affected by hyperoxia treatment. A: top ion channels and functional target functions affected by hyperoxia treatment, B: signaling and pathological targets affected by hyperoxia, C: toxicology networks affected by hyperoxia treatment. The values on y-axis show the top predicted hits depicted with scale of negative log (−log) of false discovery rate (FDR) corrected P values.

In conclusion, we report that hyperoxia leads to left ventricular hypertrophy and alters Kv4.2 ion channel expression along with key transcriptional mediators. Hyperoxia resulted in impaired cardiac function as noted by echocardiography, increased heart weights and left ventricular hypertrophy. Overall, the structural, functional and molecular level changes in the hearts from mice exposed to 100% oxygen treatment reveal novel mechanistic insights into hyperoxia-induced left ventricular hypertrophy.


The work was supported by National Institutes of Health (NIH) Grant R01-HL-102171 and University of South Florida College of Pharmacy startup fund and seed fund (to S. M. Tipparaju); American Heart Association National Scientist Development Grant 09SDG2260957 and NIH Grant R01-HL-105932 (to N. Kolliputi); and the Joy McCann Culverhouse Endowment to the Division of Allergy and Immunology and National Science Foundation Grant IOS-1146882 (to E. S. Bennett).


No conflicts of interest, financial or otherwise, are declared by the author(s).


Author contributions: S.K.P., J.T., J.F., and W.D. performed experiments; S.K.P. and W.D. analyzed data; S.K.P. prepared figures; S.K.P. and S.M.T. drafted manuscript; K.B.S., N.K., E.S.B., and S.M.T. edited and revised manuscript; E.S.B. and S.M.T. interpreted results of experiments; S.M.T. conception and design of research; S.M.T. approved final version of manuscript.


We acknowledge the assistance provided by Yanbin Dong. We thank Dr. Daniel Lee and Jerry Hunt for their help in utilizing the MIRAX acquisition and software. We also thank Kalyan C. Chapalamadugu for reading the paper.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
View Abstract