Leflunomide Attenuates Oxidative Stress in Fetal Human Lung Endothelial Cells via Superoxide Dismutase 2 and Catalase

Abstract

Hyperoxia-induced oxidative stress contributes to the pathogenesis of bronchopulmonary dysplasia (BPD), the most common respiratory morbidity of preterm infants. Importantly, the disease lack specific therapies and is associated with long-term cardiopulmonary and neurodevelopmental morbidities, signifying the need to discover novel therapies and decrease the disease burden. We and others have demonstrated that leflunomide, a food and drug administration approved drug to treat humans with rheumatoid arthritis, increases the expression of the anti-oxidant enzymes, NAD(P)H quinone dehydrogenase 1 (NQO1), catalase, and superoxide dismutase (SOD). However, whether this drug can decrease oxidative stress in fetal human pulmonary arterial endothelial cells (HPAECs) is unknown. Therefore, we tested the hypothesis that leflunomide will decrease hyperoxia-induced oxidative stress by upregulating these anti-oxidant enzymes in HPAECs. Leflunomide decreased hydrogen peroxide (H₂O₂) levels and increased the mRNA and protein levels of catalase, NQO1, and SOD2 in HPAECs at basal conditions. Further, leflunomide-treated cells continued to have decreased H₂O₂ and increased SOD2 levels upon hyperoxia exposure. Leflunomide did not affect the expression of other anti-oxidant enzymes, including hemoxygenase-1 and SOD1. AhR-knockdown experiments suggested that leflunomide regulated NQO1 levels via AhR-dependent mechanisms and H₂O₂, catalase, and SOD2 levels via AhR-independent mechanisms. Collectively, the results support the hypothesis that leflunomide decreases oxidative stress in HPAECs via SOD2- and catalase-dependent, but AhR- and NQO1-independent mechanisms. Our findings indicate that leflunomide is a potential drug for the management of BPD in preterm infants.

Introduction

Bronchopulmonary dysplasia (BPD) is the most common respiratory morbidity of preterm infants that is characterized by interrupted lung development [1, 2]. Hyperoxia-mediated increase in lung oxidative stress contributes to the pathogenesis of BPD and its sequelae [3]. Infants with BPD have long-term cardiopulmonary and neurodevelopmental morbidities causing a significant impact on the economy and quality of life [1, 4, 5]. Importantly, the disease lacks specific therapies. Hence, there is an urgent need for improved therapies to prevent and treat BPD.

Oxidative stress, an imbalance favoring a pro-oxidant over an anti-oxidant state, is one of the major risk factors implicated in the development of BPD. Supplemental oxygen is frequently used as a life-saving therapy in preterm infants with hypoxic respiratory failure; however, excessive or prolonged oxygen exposure results in oxidative stress characterized by increased reactive oxygen species (ROS) generation. Preterm infants are at high risk of developing oxidative stress due to their immature antioxidant defenses, increased susceptibility to infection and inflammation, and exposure to free iron [6]. Oxidative stress leads to interrupted lung development by mechanisms entailing disruption of growth factor signaling, extracellular matrix assembly, cell proliferation, apoptosis, and vasculogenesis [7].

Leflunomide or N- (4-trifluoromethylphenyl) – 5 – methylisoxazol - 4 – carboxamide) is an US Food and Drug Administration approved immunomodulatory drug that is used to treat rheumatoid arthritis in humans [8]. Several studies, including ours indicate that leflunomide increases the expression of anti-oxidant enzymes such as NAD(P)H quinone dehydrogenase 1 (NQO1), catalase, and superoxide dismutase (SOD) [9–12]. These observations suggest that leflunomide treatment may compensate for the immature antioxidant defenses of preterm infants and increase their resilience against hyperoxia-mediated lung injury. Thus, the goal of this study was to investigate if leflunomide modulates hyperoxia-induced oxidative stress in fetal human pulmonary arterial endothelial cells (HPAECs). We chose HPAECs for our study because lung angiogenesis actively contributes to lung development in infants and oxidative-stress mediated endothelial cell dysfunction is a major risk factor for the development of BPD [13, 14]. Using these cells, we tested the hypothesis that leflunomide decreases oxidative stress by upregulating the anti-oxidant enzymes, catalase, NQO1, and SOD2 in HPAECs.

Materials and Methods

Cell culture, treatment, and hyperoxia exposure

The fetal human pulmonary artery endothelial cells (HPAECs) were obtained from ScienCell research laboratories (San Diego, CA; 3100) and grown according to the manufacturer’s protocol. Cells were treated with either 0.01% v/v dimethyl sulfoxide (DMSO) (Sigma Aldrich, St. Louis, MO; 276855) or up to 10 μM leflunomide (Sigma Aldrich; L5025). Hyperoxia experiments were conducted in a plexiglass sealed chamber into which a mixture of 95% O₂ and 5% CO₂ was circulated continuously [15].

Cell viability assay

Cell viability was determined by a colorimetric assay based on the ability of viable cells to reduce the tetrazolium salt, MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide), to formazan [16].

Measurement of hydrogen peroxide generation

Hydrogen peroxide (H₂O₂) levels were quantified by the ROS-Glo™ H₂O₂ Assay (Promega, Madison, WI; G8820). Briefly, cells grown to 60–70% confluence on 96-well plates were treated with DMSO or leflunomide, and exposed to normoxia (21% O₂ and 5% CO₂) or hyperoxia (95% O₂ and 5% CO₂) for up to 24 h, following which the H₂O₂ levels were determined [17].

Real-time RT-PCR assays

Total mRNA was isolated and reverse transcribed to cDNA [16]. Real-time RT-PCR analysis was performed using iTaq Universal SYBR Green Supermix (Biorad, Hercules, CA; 1725121). The sequences of the primer pairs were aryl hydrocarbon receptor (AhR): 5'-CACCGATGGGAAATGATACTATCC-3' and 5'-GGTGACCTCCAGCAAATGAGTT -3'; catalase: 5'-TGT GCA TGC TAA AGG AGC AG -3' and 5'-CGA TGG GAG TCT TCT TTC CA -3'; NQO1: 5'-ACGCCC-GAATTCAAATCCT-3’ and 5'-CCTGCCTGGAAGTTTAGGTCAA-3'; heme oxygenase-1 (HO-1): 5'-AGGCCAAGACTGCGTTCC-3’ and 5'-GCAGAATCTTGCACTTTGTTGCT-3’; superoxide dismutase (SOD) 1: 5'-TGG CCG ATG TGT CTA TTG AA -3' and 5'-ACC TTT GCC CAA GTC ATC TG -3'; SOD2: 5'-TTC AAT AAG GAA CGG GGA CA -3' and 5'-ACA CAT CAA TCC CCA GCA GT -3'; and glyceraldehyde 3-phosphate dehydrogenase (GAPDH): 5'-TGG AAG GAC TCA TGA CCA CA -3' and 5'-GAG GCA GGG ATG ATG TTC TG -3'.

Western blotting

Whole-cell protein extracts obtained by using radio immunoprecipitation assay lysis buffer system (Santa Cruz Biotechnologies, Santa Cruz, CA; sc-24948) were subjected to western blotting using antibodies against: β-actin (Santa Cruz Biotechnologies; sc-47778, dilution 1:4000), catalase (Santa Cruz Biotechnologies; sc-271803, dilution 1:500), HO-1 (Enzo Life Sciences, Farmingdale, NY; ADI-SPA-896F, dilution 1:1000), NQO1 (Santa Cruz Biotechnologies; sc-16464, dilution 1:500), SOD1 (Santa Cruz Biotechnologies; sc-8637, dilution 1:300), and SOD2 (Santa Cruz Biotechnologies; sc-137254, dilution 1:500).

Small interfering RNA (siRNA) transfections

HPAECs were transfected with either 50 nM control siRNA (Dharmacon, Lafayette, CO; d-001810) or AhR specific siRNA (Dharmacon; L-004990) using LipofectamineRNAiMAX (Life Technologies, Grand Island, NY; 13778030). siRNA mediated knockdown of AhR was validated by determining AhR mRNA expression by real time RT-PCR analysis. Following transfection, whole-cell protein extracts of cells treated with leflunomide or DMSO were obtained to quantify catalase, NQO1, and SOD2 protein expression. Additionally, the transfected cells were treated with leflunomide and DMSO and exposed to normoxia or hyperoxia for 24 h, after which cells were harvested to quantify the H₂O₂ levels.

Analyses of data

The results were analyzed by GraphPad Prism 5 software. The data are expressed as mean ± SD. One-way ANOVA was used to determine the effect of leflunomide on cell viability, H₂O₂ levels and expression of anti-oxidant enzymes, while two-way ANOVA was used to determine the effects of leflunomide and AhR gene, and their associated interactions for the outcome variables, H₂O₂ levels and expression of anti-oxidant enzymes. Post hoc multiple t-tests with Bonferroni corrections were performed if statistical significance of either variable or interaction was noted by ANOVA. A p value of <0.05 was considered significant.

Results and Discussion

In this study, we investigated the effects and mechanisms by which leflunomide modulates oxidative stress in HPAECs. In wild-type HPAECs exposed to normoxia or hyperoxia in vitro, leflunomide increased catalase, NQO1, and SOD2 expression and decreased H₂O₂ levels compared with controls. siRNA-mediated knockdown of the AhR gene abrogated the effects of leflunomide on NQO1, but not on catalase, SOD2 or H₂O₂, implying that leflunomide increases NQO1 expression via AhR-dependent mechanisms and decreases H₂O₂ generation and upregulates catalase and SOD2 expression via AhR-independent mechanisms.

Hyperoxia-induced ROS generation can injure and disrupt the reparative processes in the developing lungs and contribute to BPD pathogenesis [18]. Therefore, we investigated whether leflunomide modulates the level of a reliably measured ROS such as H₂O₂ using HPAECs. Leflunomide in concentrations greater than 30 μM causes mitochondrial dysfunction and cytotoxicity [19, 20]. Further, drug-induced toxicity can induce oxidative stress and affect the expression of several anti-oxidant enzymes. Hence, we used lower leflunomide concentrations and tested for its cytotoxicity by MTT assay. Leflunomide in concentrations up to 10 μM did not affect HPAEC viability (Fig. 1A), indicating that concentrations up to 10 μM can be used to investigate its effects on the levels of H₂O₂ and anti-oxidant enzymes. To determine whether leflunomide modulates oxidative stress, H₂O₂ levels were measured by ROS-Glo™ H₂O₂ assay. Leflunomide at concentrations of 5 or 10 μM decreased H₂O₂ levels at basal conditions (Fig. 1B). To determine the mechanims associated with this effect, we quantified the levels of the antioxidant enzymes, including catalase, HO-1, NQO1, SOD1 and SOD2. While leflunomide increased the mRNA and protein levels of catalase (Figs. 2A, F, and G), NQO1 (Figs. 2C, F, and I), and SOD2 (Figs. 2E, F, and K), it did not affect the expression of HO-1 (Figs. 2B, F, and H) and SOD1 (Figs. 2D, F, and J). Having seen that leflunomide modulates H₂O₂ levels and anti- oxidant enzymes at basal conditions, our logical next step was to elucidate whether it exerts similar effects in hyperoxic conditions. Hyperoxia exposure increased H₂O₂ levels in HPAECs; however, leflunomide attenuated the effects of hyperoxia on H₂O₂ generation (Fig. 3A). Consistently we observed that leflunomide increased the mRNA and protein levels of catalase (Figs. 3B, G, and H), NQO1 (Figs. 3D, G, and J), and SOD2 (Figs. 3F, G, and L), without affecting the expression of HO-1 (Figs. 3C, G, and I) and SOD1 (Figs. 3E, G, and K) in normoxic conditions. Hyperoxia exposure increased the expression of HO-1 (Figs. 3C, G, and I) and NQO1 (Figs. 3D, G, and J). Interestingly, the leflunomide treated cells had increased levels of NQO1 (Figs. 3G and J) and SOD2 (Figs. 3G and L) protein compared with vehicle-treated cells upon hyperoxia exposure. Although, the HO-1 mRNA levels (Fig. 3C) were increased in leflunomide treated cells exposed to hyperoxia, there were no corresponding changes in HO-1 protein levels (Figs. 3G and I).

Figure 1. Leflunomide decreases H₂O₂ levels in HPAECs.

HPAECs were treated with dimethylsulfoxide (DMSO) or leflunomide (L) at concentrations of 1 (L1), 5 (L5) or 10 (L10) μM for up to 24 h, after which cell viability was assessed by MTT (3-(4, 5-dimethylthiazolyl-2)- 2, 5-diphenyltetrazolium bromide) assay (A) and H₂O₂ level was measured by ROS-Glo™ H₂O₂ assay (B). Data are representative of at least three independent experiments. Values are presented as mean ± SD (n=4). One-way ANOVA showed an effect of leflunomide for the dependent variable, H₂O₂ level. Significant differences between DMSO- and leflunomide –treated cells are indicated by *, p < 0.05.

Figure 2. Leflunomide-treated HPAECs display increased catalase, NQO1, and SOD2 expression.

HPAECs were treated with leflunomide (L) at doses of 5 (L5) or 10 (L10) μM or DMSO for 24 h, and the cells were harvested for real-time RT-PCR and immunoblot analyses. A–E. Real-time RT-PCR analyses-based determination of catalase (A), HO-1 (B), NQO1 (C), SOD1 (D), and SOD2 (E) mRNA-expression levels. F. Determination of catalase, HO-1, SOD1, SOD2, and NQO1 protein levels by immunoblotting. G-K. Quantification and normalization of catalase (G), HO-1 (H), NQO1 (I), SOD1 (J), and SOD2 (K) band intensities to those of β-actin. The data shown are representative of three independent experiments. Values are presented as the mean ± SD (n = 3/group). One-way ANOVA showed an effect of L5 and L10 for catalase, NQO1, and SOD2 mRNA and protein expression. Significant differences between DMSO-, L5- and L10-treated cells are indicated by *, p < 0.05.

Figure 3. Leflunomide treated HPAECs display decreased H₂O₂ levels and increased NQO1 and SOD2 protein expression upon exposure to hyperoxia.

HPAECs were treated with leflunomide (L) at doses of 5 (L5) or 10 (L10) μM or DMSO and exposed to normoxia or hyperoxia for 24 h, and the cells were harvested for H₂O₂ estimation (A), and real-time RT-PCR and immunoblot analyses. B–F. Real-time RT-PCR analyses-based determination of catalase (B), HO-1 (C), NQO1 (D), SOD1 (E), and SOD2 (F) mRNA-expression levels. G. Determination of catalase, HO-1, NQO1, SOD1 and SOD2 protein levels by immunoblotting. H-L. Quantification and normalization of catalase (H), HO-1 (I), NQO1 (J), SOD1 (K), and SOD2 (L) band intensities to those of β-actin. The data shown are representative of three independent experiments. Values are presented as the mean ± SD (n = 3/group). Two-way ANOVA showed an effect of leflunomide and hyperoxia on the dependent variables, H₂O₂ level, and HO-1 and NQO1 expression, and an interaction between leflunomide and hyperoxia for the dependent variables, H₂O₂ level, and NQO1 and SOD2 expression. Significant differences between vehicle and leflunomide treated cells exposed to normoxia are indicated by p, p < 0.05. Significant differences between normoxia and hyperoxia-exposed cells are indicated by *, p < 0.05. Significant differences between vehicle and leflunomide treated cells exposed to hyperoxia are indicated by †, p < 0.05.

The biological effects of leflunomide are concentration- or dose-dependent. It exerts toxic effects, including mitochondrial dysfunction, cytotoxicity, and disrupted angiogenesis when used in high concentrations [19–21]. On the other hand, leflunomide also exerts beneficial anti-inflammatory and anti-oxidant properties in several pathological states [11, 22, 23]. Our study supports the hypothesis that low concentrations of leflunomide has beneficial anti-oxidant properties without any cytotoxic effects in hyperoxic conditions. This conclusion is supported by recent studies, which have demonstrated that leflunomide protects lungs against several insults by upregulating the expression of catalase and SOD [10, 24, 25].

The aryl hydrocarbon receptor (AhR), a member of basic - helix – loop – helix family of transcriptional regulators , modulates oxidant stress by inducing detoxifying enzymes or via “cross-talk” with other signal transduction pathways [26–29]. We and others have shown that leflunomide is an AhR agonist [9, 30]. Therefore, the question we asked was that whether leflunomide attenuates oxidative stress via AhR-dependent mechanisms. To answer this question, we knocked down AhR by siRNA-mediated transfections and determined H₂O₂ generation and anti-oxidant enzyme expression in the transfected cells treated with DMSO or leflunomide. As expected, AhR siRNA significantly decreased the AhR mRNA expression (Fig. 4A). Leflunomide decreased H₂O₂ generation in normoxic and hyperoxic conditions in both AhR-sufficient and –deficient cells (Fig. 4B). Likewise, leflunomide increased catalase (Figs. 4C and D) and SOD2 (Figs. 4C and F) protein levels in both AhR-sufficient and –deficient cells exposed to normoxia. However, leflunomide-induced NQO1 protein (Figs. 4C and E) expression was significantly attenuated in AhR-deficient cells, indicating that leflunomide induces NQO1 via AhR-dependent mechanisms, a finding consistent with our previous study [9]. Our AhR- knock down experiments suggest that leflunomide decreases H₂O₂ levels via catalase- and SOD2-dependent, but AhR- and NQO1-independent mechanisms. It is possible that leflunomide modulates oxidative stress via nuclear factor erythroid 2–related factor 2-dependent pathways; further studies are necessary to address this speculation.

Figure 4. Leflunomide decreases H₂O₂ levels and increases catalase and SOD2 expression in HPAECs via an AhR-independent mechanism.

Twenty-four hours after HPAECs were transfected with either 50 nM control (SiC) or AhR (SiAhR) siRNA, the RNA was extracted to determine AhR (A) mRNA expression. Further, 24 h after transfection, the cells were treated with leflunomide (L) at doses of 5 (L5) or 10 (L10) μM or a vehicle control (DMSO) and exposed to normoxia or hyperoxia for 24 h and the H₂O₂ level was estimated by ROS-Glo™ H₂O₂ assay (B). In a separate set of experiments, the transfected cells were treated with L10 or DMSO as described above in normoxic conditions, and the cells were harvested for immunoblot analyses. C. Determination of catalase, HO-1, NQO1, and SOD2 protein levels by immunoblotting. D-F. Quantification and normalization of catalase (D), NQO1 (E), and SOD2 (F) band intensities to those of β-actin. The data shown are representative of two independent experiments. Values are presented as the mean ± SD (n = 3/group). Two-way ANOVA showed an effect of leflunomide, but not the AhR gene for the dependent variables, H₂O₂ level, catalase and SOD2 protein expression. However, there was an effect of both leflunomide and the AhR gene for the dependent variable, NQO1 protein expression. Significant difference in the AhR mRNA expression between control and AhR siRNA-transfected cells are indicated by *, p < 0.05. Significant differences between vehicle and leflunomide treated cells exposed to normoxia are indicated by p, p < 0.05. Significant differences between normoxia and hyperoxia-exposed cells are indicated by *, p < 0.05. Significant differences between vehicle and leflunomide treated cells exposed to hyperoxia are indicated by †, p < 0.05.

In summary, we demonstrate for the first time that leflunomide attenuates H₂O₂ generation by inducing multiple anti-oxidant enzymes in HPAECs. Thus, our results suggest that leflunomide can be beneficial in the prevention and treatment of oxidative stress-induced disorders like BPD, retinopathy of prematurity, and necrotizing enterocolitis in preterm infants.

Highlights.

• Leflunomide decreases oxidative stress in fetal human lung endothelial cells.

• Leflunomide induces catalase, NQO1, and SOD2 enzymes in fetal human lung cells.

• Leflunomide effects on oxidative stress, catalase, and SOD2 are AhR-independent.

• AhR deficiency abrogates leflunomide-mediated induction of NQO1 enzyme.

Abbreviations

AhR

aryl hydrocarbon receptor

BPD

bronchopulmonary dysplasia

DMSO

dimethylsulfoxide

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

HO-1

heme oxygenase-1

H₂O₂

hydrogen peroxide

HPAECs

human pulmonary artery endothelial cells

L

leflunomide (PubChem CID: 3899)

L 1

leflunomide 1 μM

L 5

leflunomide 5 μM

L 10

leflunomide 10 μM

MTT

3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide

NQO1

NAD(P)H quinone dehydrogenase 1

ROS

reactive oxygen species

SOD

superoxide dismutase