Cardioprotective Effects of Dexmedetomidine in an Oxidative-Stress In Vitro Model of Neonatal Rat Cardiomyocytes

Preterm birth is a risk factor for cardiometabolic disease. The preterm heart before terminal differentiation is in a phase that is crucial for the number and structure of cardiomyocytes in further development, with adverse effects of hypoxic and hyperoxic events. Pharmacological intervention could attenuate the negative effects of oxygen. Dexmedetomidine (DEX) is an α2-adrenoceptor agonist and has been mentioned in connection with cardio-protective benefits. In this study, H9c2 myocytes and primary fetal rat cardiomyocytes (NRCM) were cultured for 24 h under hypoxic condition (5% O2), corresponding to fetal physioxia (pO2 32–45 mmHg), ambient oxygen (21% O2, pO2 ~150 mmHg), or hyperoxic conditions (80% O2, pO2 ~300 mmHg). Subsequently, the effects of DEX preconditioning (0.1 µM, 1 µM, 10 µM) were analyzed. Modulated oxygen tension reduced both proliferating cardiomyocytes and transcripts (CycD2). High-oxygen tension induced hypertrophy in H9c2 cells. Cell-death-associated transcripts for caspase-dependent apoptosis (Casp3/8) increased, whereas caspase-independent transcripts (AIF) increased in H9c2 cells and decreased in NRCMs. Autophagy-related mediators (Atg5/12) were induced in H9c2 under both oxygen conditions, whereas they were downregulated in NRCMs. DEX preconditioning protected H9c2 and NRCMs from oxidative stress through inhibition of transcription of the oxidative stress marker GCLC, and inhibited the transcription of both the redox-sensitive transcription factors Nrf2 under hyperoxia and Hif1α under hypoxia. In addition, DEX normalized the gene expression of Hippo-pathway mediators (YAP1, Tead1, Lats2, Cul7) that exhibited abnormalities due to differential oxygen tensions compared with normoxia, suggesting that DEX modulates the activation of the Hippo pathway. This, in the context of the protective impact of redox-sensitive factors, may provide a possible rationale for the cardio-protective effects of DEX in oxygen-modulated requirements on survival-promoting transcripts of immortalized and fetal cardiomyocytes.


Introduction
Improved perinatal and neonatal care of preterm infants has yielded higher survival rates, especially for very and extremely preterm infants [1]. The first preterm infants to benefit from the care-related medical improvements are now 30 to 40 years old. The primary focus, with interventions in the first years of life to reduce neonatal and early childhood mortality and morbidity, inevitably leads to a secondary focus on the long-term health in adulthood of the prematurely born. It is already known that prematurity can lead to longer-term impairments. These long-term sequelae include motor and cognitive deficits [2], impaired psychological morbidities and socio-emotional behavioral disorders, such as attention deficit hyperactivity disorder (ADHD) and autism spectrum disorders (ASDs) [3][4][5], as well as higher levels of cardiometabolic risk factors as adults [6]. These

Cellular Model and Defining Oxygen Level
To recognize the generated results and their distinction of the used oxygen concentrations for the relevance of the cell culture models, it must be mentioned that the use of the terms "hypoxia" (5% O 2 ), "normoxia" (21% O 2 ), and "hyperoxia" (80% O 2 ) was useful here because of the simplified reference to the oxygen concentrations. The oxygen data refer to the composition of the gas space of the humidified incubator and do not correspond to the actual cellular oxygen environment of the cells. This is influenced by multiple factors, such as cell density, medium layer thickness, as well as incubation time [42]. Under these conditions, at an O 2 content of 21%, the oxygen tension or oxygen partial pressure (pO 2 ) is ∼150 mmHg and "normoxia" is accordingly used synonymously for atmospheric oxygen pressure [43]. Under physiological conditions, tissue pO 2 ranges between 3% and 8% O 2 (usually pO 2   and is often referred to as "physioxia"; defined here as "hypoxia" in experimental terminology in relation to associated Hif1α targeting under low-oxygen conditions. The hypoxic condition fits with the fetal oxygen tension. The fetal pO 2 is approximately~45 mmHg at mid-gestation and~32 mmHg closer to term [44]. Correlating with their gestational age, extremely premature infants show large variations in pO 2 measurements in the first days of life, which may well display maximum values >150 mmHg [45]. "Hyperoxia" of cardiomyocytes in the experimental setting at 80% O 2 with a pO 2~3 00 mmHg is used to simulate extreme oxygen tensions.

Immunohistochemistry
In accordance with the experimental conditions, H9c2 cells were cultured on chambered microscope glass slides (8-well chamber, removable, Ibidi GmbH, Gräfelfing, Germany, catalog #80841) in a density of 12 × 10 3 cells per well. Cells were washed three times in PBS and then fixed in 4% paraformaldehyde (PFA) for 15 min at room temperature (RT). After being washed three more times in PBS, cells were permeabilized by applying 0.2% TritonX-100 for 15 min at RT. Permeabilized cells were washed in PBS, and the following primary antibodies diluted in Antibody Diluent (Agilent Technologies, Santa Clara, CA, USA, cat S080983-2) were added: rabbit anti-α-sarcomeric actinin (1:100, Abcam, Cambridge, UK, catalog cat ab137346) and mouse anti-Ki67 (1:100, Abcam, cat ab279653). After a 24 h incubation at 4 • C, the cells were washed three times in PBS, and secondary antibodies were added: goat-anti-rabbit Alexa Fluor 594 and goat-anti-mouse Alexa Fluor 488 (1:200 in Antibody Diluent, Life Technologies, Carlsbad, CA, USA, cat A11037/A11029). After a 1 h incubation at room temperature, cells were washed three times in PBS, and slides were mounted using 4 ,6-diamidino-2-phenylindole (DAPI)-containing fluorescence-protecting mounting medium (ProLong Gold Antifade Mountant with DAPI, Thermo Fisher Science, cat P36935). Slides were viewed blinded using a Keyence BZ 9000 compact fluorescence microscope with BZ-II Viewer software and BZ-II Analyzer software (Keyence, Osaka, Japan), and four separate sections per chamber were analyzed. Nuclei and Ki67-positive cells were counted manually by using Adobe Photoshop software 22.0.0 (Adobe Systems Software Ireland Limited, Dublin, Republic of Ireland) and the cell surface area was measured by using ImageJ (National Institute of Health, Bethesda, MD, USA). The data in the ROI of the control group were used as 100% values.

LDH-Assay and CCK8
As an indicator of cell death, LDH release into the culture medium was evaluated using a Cytotoxicity Detection Kit (Roche Diagnostics, Mannheim, Germany, cat 11644793001). As instructed by the manufacturer's protocol, supernatant was taken, centrifuged at 100× g for 5 min, and subsequently applied to a 96-well plate in triplicate with a volume of 100 µL per well. After adding 100 µL of a reagent, including a catalyst and a dye solution, to each well, the plates were incubated for 30 min under protection from light. Measurement was conducted at 490 and 655 nm.
Cell Counting Kit-8 (CCK-8; Abcam, cat ab228554) was used to evaluate cell viability. H9c2 cardiomyocytes were seeded at 5 × 10 3 cells/well in 96-well plates. After the various group treatments and conditions, 10 µL of CCK-8 solution was added to each well followed by 1 h incubation at 37 • C, according to the manufacturer's instructions, and then read for optical density at 460 nm using a microplate spectrophotometer.

Statistical Analyses
Analysis of the data was carried out with GraphPad Prism 8.0 software (GraphPad Software, La Jolla, CA, USA). Data were analyzed using a multivariate repeated measures analysis of variance (ANOVA). Depending on which ANOVA test was used, multiple comparisons of means were carried out using Bonferroni's post hoc test. GraphPad Prism Software was used for the generation of graphs. Data are presented as box and whisker plots, with the line representing the median while whiskers show the data variability outside the upper and lower quartiles. Differences were considered statistically significant at a p-value of <0.05.

Proliferation Capacity and Hypertrophy
To determine the proliferative capacity relative to cell number and hypertrophic effects of oxygen-stressed H9c2 cardiomyocytes, we stained H9c2 cells with anti-Ki 67, anti-sarcomeric alpha actinin, and DAPI. Submitting H9c2 cells to both hyperoxia and hypoxia for 24 h had negative effects on their proliferative capacity; additionally, hyperoxic exposure resulted in an increase in cell area (hypertrophy) (Figure 1).
Compared to the normoxic control group, the ratio of Ki67-positive cells diminished by half in either case; however, the administration of DEX (1 µM) led to a modest increase in Ki67-positive cells in H9c2 cells submitted to 5% O 2 ( Figure 2A). The hypertrophic phenotype of H9c2 cells was induced by hyperoxia without a protective DEX effect ( Figure 2B). Consistent with reduced mitotic activities, both H9c2 cells and NRCMs showed a significant decrease in CycD2 gene expression under 5% O 2 and 80% O 2 that was restored to normoxic levels after DEX treatment (10 µM) (Figure 2A,B). Lower DEX concentrations demonstrated effects in NRCMs and H9c2 cells submitted to hypoxia (1 µM) and in hyperoxia exposed H9c2 cells (0.1 µM, 1 µM). The administration of DEX did not alter the expression of CycD2 in cells cultivated at 21% O 2 (Figure 2A,B). The complete data are presented in Supplementary Table S1. sarcomeric alpha actinin, and DAPI. Submitting H9c2 cells to both hyperoxia and hypoxia for 24 h had negative effects on their proliferative capacity; additionally, hyperoxic exposure resulted in an increase in cell area (hypertrophy) (Figure 1).
Compared to the normoxic control group, the ratio of Ki67-positive cells diminished by half in either case; however, the administration of DEX (1 µM) led to a modest increase in Ki67-positive cells in H9c2 cells submitted to 5% O2 ( Figure 2A). The hypertrophic phenotype of H9c2 cells was induced by hyperoxia without a protective DEX effect ( Figure  2B). Consistent with reduced mitotic activities, both H9c2 cells and NRCMs showed a significant decrease in CycD2 gene expression under 5% O2 and 80% O2 that was restored to normoxic levels after DEX treatment (10 µM) (Figure 2A,B). Lower DEX concentrations demonstrated effects in NRCMs and H9c2 cells submitted to hypoxia (1 µM) and in hyperoxia exposed H9c2 cells (0.1 µM, 1 µM). The administration of DEX did not alter the expression of CycD2 in cells cultivated at 21% O2 (Figure 2A,B). The complete data are presented in Supplementary Table S1. Figure 1. Reduced proliferation capacity caused by hypoxia and hyperoxia, as well as induction of hypertrophy by hyperoxia of H9c2 cardiomyocytes. Dexmedetomidine (DEX) was found to enhance hypoxia-induced decreased proliferation. Representative images of co-labeled H9c2 cells with Ki67 (green), α-sarcomeric actinin (α-actinin, red) and DAPI (blue) exposed to normoxia (21% O 2 ), hypoxia (5% O 2 ) or hyperoxia (80% O 2 ) compared to H9c2 cells treated with dexmedetomidine (DEX; 0.1 µM, 1 µM, 10 µM). Analyses were conducted after 24 h oxygen exposure. Magnification of 20×. Quantitative analysis was conducted for the DEX-treated cells with the concentrations 0.1 µM (light gray bars), 1 µM (gray bars) and 10 µM (dark gray bars) in comparison to the untreated control group (black bars). Data are normalized to the level of cells exposed to normoxia (21% O 2 ) at each time point (control 100%, black bars). n = 6 individual experiments/group. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 (ANOVA, Bonferroni's post hoc test).

LDH Release and Cell Viability
Cardiomyocyte cells were cultured under normoxic, hypoxic, and hyperoxic conditions, and after 24 h, the release of lactate dehydrogenase (LDH) into the culture medium was measured for quantitative determination of when the plasma membrane was damaged ( Figure 3A,B). In H9c2 cells, no statistically significant alterations in LDH were detected in either hypoxia or hyperoxia ( Figure 3A). Interestingly, in NRCMs, submission to 80% O 2 for 24 h led to a two-fold increase in the LDH release, while hypoxia did not have any effects ( Figure 3B). The administration of DEX did not alter the LDH release in any case ( Figure 3A,B).

Apoptosis and Autophagy
The extrinsic apoptotic pathway is mainly mediated via the dimerization and activation of initiator caspase 8 (Casp8) secondary to extracellular apoptotic stimuli. It subsequently activates executioner caspases, such as caspase 3 (Casp3), to initiate cellular degradation [48]. Under hyperoxia and hypoxia, Casp3 was upregulated in both H9c2 cells ( Figure 4A) and NRCMs ( Figure 4B). DEX reduced the expression of Casp3 at all concen- To assess viability, H9c2 cells were incubated in 5%, 21%, or 80% oxygen for 24 h and then subjected to CCK 8 analysis ( Figure 3C). Based on the detection of dehydrogenase activity in viable cells, cell viability was decreased in H9c2 cells submitted to 80% O 2 . The administration of DEX did not produce significant alterations in cell viability. Hypoxia did not influence cell viability either ( Figure 3C). The complete data are presented in Supplementary Table S2.

Apoptosis and Autophagy
The extrinsic apoptotic pathway is mainly mediated via the dimerization and activation of initiator caspase 8 (Casp8) secondary to extracellular apoptotic stimuli. It subsequently activates executioner caspases, such as caspase 3 (Casp3), to initiate cellular degradation [48]. Under hyperoxia and hypoxia, Casp3 was upregulated in both H9c2 cells ( Figure 4A) and NRCMs ( Figure 4B). DEX reduced the expression of Casp3 at all concentrations and under either O 2 condition, effectively reaching values located below the normoxic control group (e.g., in NRCMs under 5% O 2 with 0.1 µM DEX). Casp8 was upregulated in cells submitted to hyperoxia regardless of the cell type ( Figure  The apoptosis-inducing factor (AIF) is known as an effector of intrinsic-caspaseindependent apoptotic cell death [49]. In H9c2 cells, hypoxia and hyperoxia led to an upregulation of AIF expression ( Figure 4E). DEX restored normoxic levels at all concentrations. Interestingly, both hyperoxia and hypoxia exerted contrary effects on NRCMs ( Figure 4D). Here, we observed a marked downregulation of AIF in either case. In NRCMs, DEX (1 µM and 10 µM) induced a modest increase in AIF expression under hypoxia and showed no effects under hyperoxia. Autophagy is of pivotal importance to cardiomyocytes for the maintenance of cellular homeostasis and in the generation of cardiocirculatory diseases. Autophagy-related 5 (Atg5) and autophagy-related 12 (Atg12) play a crucial role in the formation of the autophagosome [50]. We found that in H9c2 cells, the expression of Atg5 and Atg12 was upregulated in cells cultivated at both 5% O 2 and 80% O 2 for 24 h ( Figure 5A,C). DEX induced a decrease in the expression of both transcripts, reaching control levels at all DEX and oxygen concentrations except for Atg5 in hypoxia-exposed H9c2 cells treated with 1 µM DEX ( Figure 5A). Intriguingly, diametrically opposed impacts on NRCMs were determined ( Figure 5B,D). Both hypoxia and hyperoxia reduced the expression of Atg5 and Atg12. No statistically significant alterations in the expression of Atg5 were found in NRCMs after the administration of DEX ( Figure 5B). The downregulation of Atg12 in NRCMs under 80% O 2 was compensated by DEX at a concentration of 1 µM ( Figure 5D). DEX did not cause any alterations in the expression of Atg5 or Atg12 in cells submitted to normoxia. The complete data are presented in Supplementary Tables S3 and S4. The apoptosis-inducing factor (AIF) is known as an effector of intrinsic-caspase-independent apoptotic cell death [49]. In H9c2 cells, hypoxia and hyperoxia led to an upregulation of AIF expression ( Figure 4E). DEX restored normoxic levels at all concentrations. Interestingly, both hyperoxia and hypoxia exerted contrary effects on NRCMs (Figure 4D). Here, we observed a marked downregulation of AIF in either case. In NRCMs, DEX (1 µM and 10 µM) induced a modest increase in AIF expression under hypoxia and showed no effects under hyperoxia. Autophagy is of pivotal importance to cardiomyo- pacts on NRCMs were determined ( Figure 5B,D). Both hypoxia and hyperoxia reduced the expression of Atg5 and Atg12. No statistically significant alterations in the expression of Atg5 were found in NRCMs after the administration of DEX ( Figure 5B). The downregulation of Atg12 in NRCMs under 80% O2 was compensated by DEX at a concentration of 1 µM ( Figure 5D). DEX did not cause any alterations in the expression of Atg5 or Atg12 in cells submitted to normoxia. The complete data are presented in Supplementary Tables S3 and S4.

Oxidative Stress Response
The protein GCLC acts as the catalytic subunit of the glutamate cysteine ligase (GCL), an enzyme that catalyzes the first and rate-limiting step of glutathione synthesis. It is therefore involved in cellular antioxidative mechanisms [51]. Hyperoxia led to a threefold increase in the expression of GCLC in NRCMs ( Figure 6B) and to a marked upregulation in H9c2 cells ( Figure 6A). The administration of DEX induced a decrease in GCLC

Oxidative Stress Response
The protein GCLC acts as the catalytic subunit of the glutamate cysteine ligase (GCL), an enzyme that catalyzes the first and rate-limiting step of glutathione synthesis. It is therefore involved in cellular antioxidative mechanisms [51]. Hyperoxia led to a three-fold increase in the expression of GCLC in NRCMs ( Figure 6B) and to a marked upregulation in H9c2 cells ( Figure 6A). The administration of DEX induced a decrease in GCLC expression at all concentrations in cells submitted to hyperoxia. A downregulation of GCLC under hypoxic conditions was only detected in NRCMs ( Figure 6B). At atmospheric oxygen levels, DEX did not alter the expression of GCLC in H9c2 cells ( Figure 6A), whereas in NRCMs, the administration of 1 µM DEX led to a modest downregulation of GCLC expression ( Figure 6B).
Oxidative stress puts cellular homeostasis at risk. Nuclear factor-erythroid 2-related factor 2 (Nrf2) is a transcription factor that mediates a plethora of cytoprotective antioxidative mechanisms [52]. Hyperoxia induced a marked increase in Nrf2 expression in H9c2 cells ( Figure 6C), reaching peak values of approximately 200% compared to the control group. DEX restored normoxic levels of Nrf2 at all concentrations (0.1 µM, 1 µM, 10 µM). In H9c2 cells submitted to normoxia, all DEX concentrations led to a decrease in the expression of Nrf2 ( Figure 6C). The upregulation of Nrf2 in cells cultivated under 80% O 2 was less intense in NRCMs, and the administration of DEX had no effect on the expression of Nrf2 detected ( Figure 6D). expression at all concentrations in cells submitted to hyperoxia. A downregulation of GCLC under hypoxic conditions was only detected in NRCMs ( Figure 6B). At atmospheric oxygen levels, DEX did not alter the expression of GCLC in H9c2 cells ( Figure 6A), whereas in NRCMs, the administration of 1 µM DEX led to a modest downregulation of GCLC expression ( Figure 6B).  The cellular adaptation to conditions of low oxygen is orchestrated by hypoxiainducible factor 1α (Hif1α) [53]. As expected, we detected an upregulation of Hif1α in NRCMs and H9c2 cells submitted to hypoxia ( Figure 6E,F). DEX led to a decrease, reaching levels measured in the control group at all concentrations (0.1 µM, 1 µM, 10 µM) regardless of the cell type. The expression of Hif1α was not affected by hyperoxia. In H9c2 cells, DEX (1 µM) induced a statistically significant, but modest decrease in the expression of Hif1α ( Figure 6E). The complete data are presented in Supplementary Table S5.

Cardiomyocyte Tissue Structure
The cardiomyocyte-specific marker cardiac muscle troponin T (Tnnt2) is a sensitive marker of cardiac injury and plays an important role in the regulation of cardiac muscle contraction [54]. The expression of Tnnt2 was upregulated in H9c2 cells when submitted to 80% O 2 ( Figure 7A). DEX showed lowering effects at all doses. In contrast, in NRCMs, cells cultivated under hyperoxic conditions showed a decrease in the expression of Tnnt2 by half compared to the control group ( Figure 7B). Hypoxia had no effects on Tnnt2 of H9c2 cells and NRCMs. In both cases, the administration of DEX did not lead to any alterations ( Figure 7A,B).

Hippo Pathway
The Hippo signaling pathway is known to play a critical role in the cardiomy cell cycle [56]. Briefly, a kinase cascade containing Mst1 and Lats2 leads to the phos ylation of Yap, preventing its nuclear translocation. Consequently, Yap can neither a ate with its cofactors, such as Tead1, nor bind to the DNA to act as a transcription f Tissue inhibitor of metalloproteinases (TIMPs) maintains the homeostatic balance of myocardial ECM by inhibiting activated matrix metalloproteinases (MMPs) that contribute to myocardial fibrosis through different mechanisms [55]. When cultivated under 80% O 2 , both H9c2 cells and NRCMs showed an upregulation of Timp1and Timp2-expression ( Figure 7C-F). In all cases, administration of DEX effectively lowered the increased expression of Timp1 and Timp2 at 0.1 µM. Higher doses (1 µM and 10 µM) showed statistically significant effects on every group except for Timp2 in NRCMs. Hypoxia did not have any effect on the expression of Timp1, regardless of the cell type ( Figure 7C,D). In hypoxiaexposed H9c2 cells, no alterations of the expression of Timp2 were detected ( Figure 7E). In NRCMs, however, Timp2 was expressed to a lesser extent in NRCMs submitted to hypoxia. Under these circumstances, DEX (10 µM) led to an increase in Timp2 expression ( Figure 7F). The complete data are presented in Supplementary Table S6.

Hippo Pathway
The Hippo signaling pathway is known to play a critical role in the cardiomyocyte cell cycle [56]. Briefly, a kinase cascade containing Mst1 and Lats2 leads to the phosphorylation of Yap, preventing its nuclear translocation. Consequently, Yap can neither associate with its cofactors, such as Tead1, nor bind to the DNA to act as a transcription factor with positive effects on cell cycle progression [56].
We found that Cul7, a ubiquitin ligase that promotes the degradation of Mst1, is expressed at lower levels under 5% O 2 and 80% O 2 compared to the control group in both H9c2 cells and NRCMs ( Figure 8A In H9c2 cells, both hyperoxia and hypoxia led to an induction of Lats2 expression. Treatment with DEX at low doses (0.1 µM) induced a decrease in either case. Higher doses (1 µM and 10 µM) showed statistically significant effects in cells submitted to hyperoxia for 24 h ( Figure 8C). In NRCMs, the expression of Lats2 was increased after incubating the cells at 80% O 2 , whereas, despite a recognizable tendency towards higher expression of Lats2, no statistical significance could be detected under hypoxic conditions. After the application of DEX, Lats2 levels decreased at all DEX concentrations for the hypoxic group and at 0.1 µM and 1 µM for the hyperoxic group ( Figure 8D). DEX did not change the expression of Lats2 at 80% O 2 .
Interestingly, the expression of YAP1 was markedly upregulated in both H9c2 cells and NRCMs submitted to hyperoxia for 24 h (Figure 8G,H). At all concentrations, DEX led to a decrease in YAP1 and reached normoxic control levels. Hypoxia did not alter the expression of YAP1. The complete data are presented in Supplementary Table S7. the expression up to the normoxic level in H9c2 cells (80% O2 and 5% O2, Figure 8A whereas a significant increase with DEX (1 µM and 10 µM) in NRCMs could be detecte in cells cultivated under 5% O2. DEX did not cause any alterations in cells submitted t 21% O2.

Discussion
In this study, we demonstrated that compared with ambient air, lower oxygen concentration (hypoxia or physioxia) and higher oxygen tension (hyperoxia) decreased the proliferative capacity and associated transcripts of immortalized and primary rat cardiomyocytes, and hyperoxia-induced cellular hypertrophy. Apoptosis-related transcripts were upregulated under both hypoxic and hyperoxic conditions. Autophagy-related mediators were induced for immortalized H9c2 cells, whereas these were downregulated for primary cardiomyocytes (NRCM). This was consistent with the induction of oxidativestress-response-related mediators in response to hypoxia as well as hyperoxia. Cardiac stress and tissue remodulation factors appeared to be more influenced by hyperoxia. DEX was able to counteract these oxygen-modulated effects at the transcript level in both immortalized and primary rat cardiomyocytes. Beside redox-sensitive signaling mediators, the Hippo signaling pathway affected the cardio-protective effects.
Oxidative stress is an imbalance between pro-oxidant trigger factors and corresponding counter-regulatory mechanisms. Free radicals and ROS are formed under physiological conditions during oxidative metabolism, but in concentrations that are not toxic to cells. Insufficient antioxidant capacities of cells or tissues cause an imbalance, so that ROS accumulate. Increased ROS causes damage to macromolecules, as a result of which cell death, proliferation, or cell differentiation factors and mediators are regulated and involved in the pathogenesis of disease in the premature setting [22,23,57]. Abnormal production of free radicals causes alterations in molecular signaling pathways that underlie the development of cardiovascular diseases [58,59]. In the postnatal situation, a rapid adaptation of the heart from lower in utero oxygen to a higher ex utero oxygen supply occurs and causes a proliferation arrest of cardiomyocytes. The change in oxygen partial pressure with increased ROS is considered causative for myocardial cell cycle arrest [25]. Hypoxia is essential for the developing heart, whereas decreased oxygenation in the adult heart leads to cell death, cardiomyocyte hypertrophy, fibrosis, and impaired cardiac function. Hypoxia critically modulates the expression of genes responsible for cell death, metabolism, proliferation, and differentiation. An important player in the regulation of these processes is the transcription factor Hif1α [26]. Another redox-sensitive transcription factor is Nrf2, which is involved in regulating the expression of genes for antioxidants and detoxification [60].
In parallel with cardiomyocytes isolated from fetal hearts, embryonic rat ventricular myocyte H9c2 cells were used to study proliferation and regulation of various transcripts [61]. Our in vitro study clearly showed the impairment of proliferation by low and high oxygen concentrations, as well as the induction of a hypertrophic phenotype after hyperoxic exposure of the immortalized cells. The direct hypoxia/hyperoxia proliferation inhibitory effect on H9c2 myocytes was confirmed by immunofluorescence staining of the nuclear protein Ki-67, which is expressed in all cell cycle phases except G0 [62]. Quantitative analysis of α-sarcomeric actinin and Ki67-positive cells suggested that hypoxia and hyperoxia reduced proliferative capacity, although complete cell cycle arrest could not be inferred. An induction of expression of relevant transcription factors was demonstrated for Nrf2 for hyperoxia and for Hif1α for hypoxia. This is consistent with findings from an in vitro study of neonatal cardiomyocytes, in which 1% oxygen resulted in the induction of Hif1α but inhibited mitotic activities and induced cell death [63]. Furthermore, already maternal hypoxia (10.5% O 2 ) in pregnant mice caused cell cycle arrest in fetal cardiomyocytes [64].
In line with this, 24 h of hypoxia conditioned an increase in Casp3 and Casp8 gene expression for both cell lines and primary cells, which is significant for caspase-dependent apoptosis. Hypoxia serves the Hif1α signaling pathway for progression of proliferation as well as regulation of cardiomyocyte apoptosis, according to these results. In the complex orchestration of the regulation of cellular processes in cardiomyocytes, other mediators must be involved. The apoptosis-related caspase-independent factor AIF and the autophagy-involved members of the ATG family, Atg5 and Atg12, were induced by hypoxia in immortalized H9c2 cells, suggesting that differential pathways may trigger cell death. Considering the results obtained for primary cardiomyocytes, hypoxia induced the caspase-dependent apoptosis factors Casp3 and Casp8 as well but reduced the expression of AIF and the autophagy mediators Atg5/12. Numerous studies document differential types of cell death, such as apoptosis, autophagy, and ferroptosis, in myocardial ischemia [65][66][67]. The underlying mechanisms for myocyte cell death are not completely understood.
Lysosome-mediated autophagy has certainly been linked to cardiac disease, with both a degenerative and protective effect seeming possible [68]. Impaired progression of autophagy appears to be a factor in myocardial injury [68,69] and is orchestrated by an overlapping of a variety of signaling pathways (for an overview, see [70]). Hif1α also appears to be autophagy-regulatory via the Hif1α/BNIP3 pathway. Shown in H9c2 cells under OGD/R injury as well as in CoCl 2 -induced injury to mimic hypoxia [71], both Hif1α and autophagy-related mediators are induced [72]. Compared to previous studies, DEX demonstrated a protective effect on cardiomyocytes by inhibiting ROS-induced apoptosis [41], an antioxidant effect on the developing rodent brain [73], as well as known neuroprotective mechanisms in general [74,75]. Li et al. [76] demonstrated the protection of H9c2 cells against CoCl 2 -induced damage using the antioxidant naringin, a natural bioflavonoid, by enhancing autophagic flux via activation of the Hif1α/BNIP3 pathway. In our data, transcripts for Hif1α as well as Atg5/12 are decreased by DEX under hypoxia. Hif1α induction also occurred by hypoxia in primary NRCMs; however, in this primary cell type DEX tended to induce the expression of autophagicAtg5/12.
Secondly, Hifs regulate other cellular processes, such as energy substrate metabolism. In this context, Hif1α, triggered by hyperoxia, induces the expression of lactate dehydrogenase (LDH) [77]. We detected an LDH increase only for primary NRCMs under hyperoxia, but without Hif1α regulation. It should be noted here that the primary NRCMs had to withstand very high-stress levels under hyperoxia, recognizable by a high number of intracellular vesicles. A tendency may be discussed for LDH induction with increased Hif1α for the H9c2 cells, which might support this statement. In a low oxygen environment, in vivo associated with the in utero situation of the fetal maturing heart, and Hif1α is again a game changer when it comes to the transition of oxygen concentration. Hif1α controlling cardiac energy metabolism may have far-reaching consequences for the newborn heart in a premature situation due to the multiple effects of the targeted genes. [78]. Postnatally, the proliferation rate of cardiomyocytes is drastically reduced as hypoxia-sensitive transcription factors decrease; without growth stimuli, cardiomyocytes hypertrophy. Compared with physiological hypertrophy, progressive pathophysiological hypertrophy is associated with cellular changes, including marked changes in gene expression [78].
Nrf2 is another important transcription factor involved in redox homeostasis and oxidative stress response. The Nrf2 signaling pathway is equally implicated in the interactions between autophagy and cardiovascular disease. By inhibiting oxidative stress responses, Nrf2 signaling may be beneficial to the regulation of stability, with clear evidence that stimulation of Nrf2 signaling pathways may prove cytoprotective [79,80]. Our in vitro analyses revealed an induction of Nrf2 transcription under hyperoxic insult for both the cell line and primary cardiomyocytes. This could be downregulated by DEX in the H9c2 myocytes and displayed a decreasing trend in the primary NRCMs. The transcription factor Nrf2 not only regulates the oxidative stress response but is also an anti-inflammatory protein and exhibits pro-inflammatory properties under different physiological circumstances. Upregulated Nrf2 often shows up in the early damage phase and can enhance myocyte injury and apoptosis [81]. Whether Nrf2 up-regulation is the initial event or represents a secondary response is difficult to assess. The hypothesis of constitutively activated Nrf2 as well as Nrf2 induced under oxidative stress supports the contention that Nrf2 represents an important regulator by activating nearby antioxidant genes, such as heme oxygenase (HO1) and superoxide dismutase (SOD) [79].
In this context, it may also be mentioned that Nrf2 expression is upregulated in the early stages of myocardial remodeling and is connected to apoptosis, fibrosis, myocardial hypertrophy, and cardiac dysfunction [82]. What we found is that under hyperoxic conditions, Timp1 and Timp2 were generated at increased transcript levels, both for H9c2myocytes and primary cardiomyocytes. Remodeling of cardiac tissue and resulting dysfunction are causative for hypertrophic progression of cardiac tissue. It is quite promising that abnormal autophagy also appears to be involved. If hypertrophy is mimicked in primary, neonatal mouse cardiomyocytes and autophagy is subsequently reduced by Atg5-siRNA, the connection can be reconstructed via impaired contractility capacity [83]. As mentioned previously, maternal hypoxia inhibits cardiomyocyte proliferation in fetal and neonatal rat pups [64]. Furthermore, hypoxia has been shown to directly inhibit proliferative capacity via initiation of the gene expression of TIMPs. This was shown in our in vitro study for exposure to high oxygen concentrations; i.e., upregulated Timps1/2 is in line with decreased proliferation capacity and downregulation of CycD2 [63,64].
The cardiomyocyte marker Tnnt2, relevant to cardiomyocyte functions, such as contractility, was not modulated by hypoxia in our in vitro experiments, although it was differentially regulated under hyperoxia. H9c2 myocytes upregulated Tnnt2, whereas in primary NRCM, Tnnt2 transcription was inhibited. DEX exhibited counter-regulatory effects only for the immortalized cells. Cardiomyocytes deleted for the heterozygous and homozygous Tnnt2 gene exhibited decreased contractility compared with wild-type cardiomyocytes [84]. Cardiac diseases, such as dilated cardiomyopathy or hypertrophic cardiomyopathy, are progressive, and it is suspected that genes involved in the earliest disease development and divergence may be involved.
Hyperoxic insult as an oxidative-stress triggering factor has not been adequately studied for effects on cellular cardiac impairments. Affected cardiac functions can be demonstrated in experimental studies in mice [85]. It has been clearly shown that neonatal hyperoxia also resulted in cardiac dysfunction in adult mice [86], similar to clinical studies of premature infants with cardiac dysfunction in adults [14]. Hyperoxia-modulated redox homeostasis and its downstream signaling pathways, such as Nrf2 signaling pathways [79,80], as previously elaborated, may be causative [87,88]. Hyperoxia affects postnatal proliferation of cardiomyocytes similarly to hypoxia. Neonatal mice exposed to 100% O 2 during the first four days of life inhibited cardiomyocyte proliferation, resulting in hypertrophy [89]. Adult-human-transplant-derived cardiomyocytes responded to 48 h of hyperoxia at 95% O 2 with increased cell death as well as increased LDH levels [90], similar to our primary NRCMs.
The effects of oxidative stress, in terms of hyperoxia as well as hypoxia, condition individual interactions with redox-sensitive signaling pathways and the complex interconnectedness among these pathways [21]. Cardiomyocytes grow during embryonic development and are regulated by different signaling pathways. Nevertheless, it is worth looking at a signaling pathway that plays an important function during cardiac development, such as the Hippo/Yap signaling pathway (for review see [91][92][93]). The Hippo-YAP pathway plays a critical role in morphogenesis by regulating cardiomyocyte proliferation and differentiation [91,92,94].
The Hippo signaling pathway is affected by ROS-induced oxidative stress. This changes the transcription of mediators, such as YAP1, which is the major Hippo downstream target. In primary mouse cardiomyocytes, overexpression of YAP caused antiapoptotic effects on H 2 O 2 -induced oxidative stress [95]. In cardiomyocytes themselves, YAP1 may act as a transcriptional co-activator, contributing to the upregulation of antioxidant genes [96]. YAP1 transcription was induced to a considerable extent only by 80% oxygen in our in vitro experiments, but was also regularly reduced with DEX, whereas no changes were observed under hypoxia. Mammalian sterile 20-like kinase 1 (Mst1) and Lats2, players in the Hippo signaling pathway, regulate phosphorylation of YAP in the heart and its nuclear translocation with subsequent transactivation of cell cycle genes [97]. Mst1 has been proven to influence myocardial autophagy and strongly suggests the influence of the Hippo signaling pathway on the regulatory processes of autophagy and apoptosis [98,99]. Concomitant with the upregulation of YAP1, induction of Lats2 expression occurred in both H9c2 cells and NRCMs, but here also increased expression under low oxygen (5%). Cul7 appears to adopt pro-survival and proliferation-promoting properties in the heart [100,101]; Cul7 depletion caused hypertrophy in cardiomyocytes [102]. Moreover, Cul7 promotes the Mst1 degradation in cardiomyocytes [97]. Cul7 expression was in line with decreased cardiomyocyte proliferation in H9c2 and primary cells. DEX was able to enhance downregulated Cul7 transcripts under hypoxia for H9c2 and NRCM, as well as for H9c2 under hyperoxia. After translocation of YAP1 to the nucleus, it acts with various transcription factors, such as TEAD1, to regulate the targets involved in cardiac proliferation, survival, and differentiation [91,92,94]. The expression of YAP1 is positively correlated with the change in cardiomyocyte proliferation in response to hypoxic conditions and seems to be related to Tnnt2 expression [103]. TEAD1 mRNA expression showed a reduced level for both oxygen concentrations and could be counter-regulated by DEX. Tnnt2 acts differently and is shown to be affected only by hyperoxia. H9c2 cells induced Tnnt2, whereas primary cardiomyocytes decreased in Tnnt2 expression. The sensitive relationships between Hippo signaling and cardiomyocytes were evident with YAP1 inactivation. This decreased cardiomyocyte proliferation in vivo and in vitro, whereas fetal activation of YAP1 promoted it, although physiological hypertrophy was not affected [104].
We were able to demonstrate a great vulnerability of immortalized as well as primary fetal cardiomyocytes to oxidative stress triggered by both low and high oxygen concentrations compared with normoxic controls. Direct comparison to the influence of oxygen concentrations is limited in some aspects for the cells used. In several experimental approaches, we have illustrated that the immortalized rat H9c2 cardiac cell lines differ markedly from the corresponding neonatal primary cardiac cells in some aspects and resemble each other in other aspects [105,106]. Thus, both comparable effects of oxidative stress on apoptosis-associated mediators were identified, but concomitantly, the effects of aberrant oxygen concentrations on autophagy-associated mediators differed. This was well correlated with effects on Hippo pathway transcripts and impaired proliferation.
Exciting in this context is the influence of DEX, which could congruently counteract the oxidative stress responses almost everywhere. The results were very impressive for the transcript expression of the cell cycle regulator cyclin D2 and cell-death-associated mediators with direct reference to the Hippo pathway. A possible link with accentuated evidence for autophagy could be discussed in light of recent findings on mechanisms for DEX (reviewed in [93]) always in terms of the antioxidant properties of DEX, per se. In addition, some other antioxidants, such as N-acetylcysteine, were able to promote proliferation of neonatal cardiomyocytes in vitro and in vivo by activating the YAP signaling pathway. The dependence of high oxidative metabolism for proliferation of neonatal cardiomyocytes may account for the vulnerability to varying oxygen concentrations [93,107]. It should also be mentioned that cardiometabolic disease, with a higher prevalence in preterm infants, is a multifactorial process. Factors leading to overproduction of free radicals and thus oxidative stress include total parenteral nutrition, blood transfusions, as well as painful procedures (reviewed by [108]).
Limiting for our study in immortalized as well as primary cardiomyocytes are the oxygen concentrations because they cannot fully mimic the individual tissue pO 2 , named physioxia, since, on the one hand, concrete values for preterm infants are not available and, on the other hand, the ambient conditions of physioxia for standard isolation from rat hearts as well as the standard cell line could not be realized. Moreover, no confirmatory proteomic analyses were performed in consideration of the reduction in animal numbers according to the 3R principle, as gene expression profiling has sufficient potential to reveal the effects of exogenous factors by analyzing secondary responses to conditional regulatory events. Known cellular process-or pathway-dependent biomarkers can fundamentally describe the cellular effects that subsequently drive degenerative or protective cell fates. Importantly, we experimented with undifferentiated H9c2 cells, which may differ markedly from primary cardiac cells in morphology and may also differ in gene expression patterns. Even differentiation of H9c2 cells could not have abolished this difference. Because there is an essential need for new insights into molecular mechanisms for multifaceted cardio-associated diseases, cell lines can be used to clarify fundamental aspects. Of course, this is also done to identify alternative methods in line with the 3Rs principle. To be able to use valid data for in vitro studies and to reveal differences between tissues, primary cells and target lines that are widely used in experimental research, it is of high importance to study detailed knowledge of the strengths of cardiac cell lines. There is a significant need for experimental models to understand why high oxygen exposure at birth is a risk factor for adult cardiovascular disease.

Conclusions
Cardiovascular impairment of preterm infants in adulthood may be caused by changes in cellular functionality during prematurity. An early disturbance of cardiomyocyte proliferation in favor of hypertrophic cells and a subsequent reduction of the cell pool due to an early hypoxic-hyperoxic transition would be a debatable aspect. The reduction in oxidative stress by antioxidant drugs as well as their protective effects on cellular degeneration could be a therapeutic target. The balance of the necessary or detrimental effects of oxygen is the target of diverse experimental approaches in neonatology research. The formation of ROS is not avoidable in premature birth, but it does have long-term effects on cerebral, pulmonary, and cardiovascular structural and functional organ development.
Our experimental procedures with different oxygen concentrations support this controversial debate on the role of oxygen in the effects on the preterm heart. In fact, high oxygen concentration is not found to be beneficial for cardiac development. Interestingly, in this study, we present results showing that cell cycle regulators of neonatal cardiomyocytes in vitro can be increased by pretreatment with dexmedetomidine under low and high oxygen, but the cellular "inhibitory" role of high (80% O 2 ) as well as low oxygen (5% O 2 ) concentration on cardiomyocyte proliferation was not clearly comprehensible; however, what is considered the normal oxygen concentration is relative, and the discussion could also start from the possibly normal 5% oxygen relative to moderate (21% O 2 ) and absolute (80% O 2 ) hyperoxia.
Therefore, it is crucial to determine the optimal oxygen concentration for each clinical situation to identify oxidative-stress-related late effects well and to establish new therapeutic strategies.
Supplementary Materials: The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/antiox12061206/s1. All basic data of the created diagrams are available in the following supplementary tables (Tables S1-S7).  Institutional Review Board Statement: All animal procedures were approved by the local animal welfare authorities (Landesamt für Gesundheit und Soziales (LAGeSo) Berlin, Germany, approval number T CH 0019/20) and followed institutional guidelines.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data used to support the findings of this study are available from the corresponding author upon request. The analyzed data used to create the graphs and statistical evaluation are attached in the supplemented material of this work.