Atorvastatin Decreases C-Reactive Protein-Induced Inflammatory Response in Pulmonary Artery Smooth Muscle Cells by Inhibiting Nuclear Factor-κB Pathway

Authors

  • Jie Li,

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author
  • Jian-Jun Li,

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author
  • Jian-Guo He,

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author
  • Jing-long Nan,

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author
  • Yuan-lin Guo,

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author
  • Chang-ming Xiong

    1. Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100037, China
    Search for more papers by this author

  • inline image

Correspondence
Jian-Jun Li, M.D., Ph.D., or Jian-Guo He, M.D., Ph.D., Department of Cardiology, Cardiovascular Institute & Fu Wai Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, 167 BeiLiShi Road, Beijing, 100037, P.R. China.
Tel.: 86-10-88398943, 86-10-88396077;
Fax: 86-10-68331730;
E-mail: lijnjn@yahoo.com.cn

Abstract

C-reactive protein (CRP) is well-known inflammatory marker, and recognized as a risk predictor of pulmonary arterial diseases. Although statins have a beneficial effect in animal models and patients with pulmonary arterial hypertension (PAH), the underlying mechanisms of their actions have less been investigated. The aims of this study was to examined the effects of CRP on expressions of interleukin-6 (IL-6) and monocyte chemoattractant protein-1 (MCP-1), and the possible mechanisms of atorvastatin on CRP-induced IL-6 and MCP-1 production in cultured human pulmonary artery smooth muscle cells (PASMCs). In a preliminary study, the human PASMCs were stimulated by a variety of concentrations of CRP (5–200 μg/mL) at different time points (0, 3, 6, 9, 12, 18 and 24 h) for the purpose of determining the dose- and time-dependent effects of CRP on inflammatory response of the cells. Then, the cells were pre-incubated for 2 h with atorvastatin (0.1–10 μmol/L) in the presence of CRP. The supernatant levels of both IL-6 and MCP-1 secretion were examined by ELISA. The cellular mRNA expressions of IL-6 and MCP-1 and nuclear factor-κB (NF-κB) activity were determined by real-time reverse transcription and polymerase chain reaction (RT-PCR) and electrophoretic mobility shift assay (EMSA), respectively. CRP resulted in elevated IL-6 and MCP-1 secretion and mRNA expression in a dose- and time-dependent manner. In addition, CRP also significantly activated the NF-κB pathway. Preincubation with 0.1–10 μmol/L of atorvastatin significantly decreased the secretions of IL-6 and MCP-1 induced by CRP. Moreover, 10 μmol/L of atorvastatin completely abrogated CRP-induced increase in IL-6 and MCP-1 by attenuating the activation of NF-κB. The present study demonstrated that inhibiting effect of atorvastatin on CRP-induced inflammatory response in cultured PASMCs was associated with NF-κB pathway. This pathway might represent a promising target for controlling CRP-induced inflammatory response in pulmonary arterial diseases.

Introduction

Inflammation is a typical feature of atherosclerotic diseases. Increasing evidence has suggested that inflammation may play an important role in pulmonary vascular disease. For example, elevated systemic inflammatory markers such as interleukin-6 (IL-6) and monocyte chemoattractant protein-1 (MCP-1) are reported in patients with primary pulmonary arterial hypertension (PAH) [1–4], and associated with higher pulmonary arterial pressure (PAP) [5–11]. Moreover, plasma concentration of C-reactive protein (CRP), a sensitive marker of underlying systemic inflammation, are also elevated in patients with PAH, suggesting that the serum CRP level may be an independent predictor of systolic PAP [12,13]. Although previous studies have demonstrated that CRP could cause endothelial dysfunction, down-regulate nitric oxide synthesis, activate nuclear factor-kappa B (NF-κB), and promote expression of adhesion molecules and plasminogen activator inhibitor-1 (PAI-1) in cultured monocytes, endothelial cells, and smooth muscle cells [14–17], the possible effect of CRP on IL-6 and MCP-1 expression in pulmonary arterial smooth muscle cells (PASMCs) has been limited.

NF-κB is a key transcription factor in inflammatory responses, which controls the transcription of many genes with an established role in atherosclerosis, such as cytokines, chemokines, and adhesion molecules. It has been demonstrated that NF-κB inhibitor, pyrrolidine dithiocarbamate decreased vascular cell adhesion molecule-1 (VCAM-1) expression and macrophage infiltration, and ameliorated pulmonary hypertension in a PAH rat model injected with the toxin monocrotaline [18], suggesting that the activation of NF-κB may be associated with the development of PAH. Additionally, statins have used for the treatment of cardiovascular disease for many years, and have an established role in the treatment of atherosclerotic disease. Similarly, some recent studies have also suggested that statins have beneficial effects for PAH [19–26]. However, the underlying mechanisms of their beneficial actions have less been investigated.

In the present study, therefore, we examined the effects of CRP on IL-6 and MCP-1 expression in cultured human PASMCs, and evaluated the underlying mechanisms of atorvastatin on CRP-induced inflammatory cytokines in the cells.

Methods

Human PASMCs Culture

Human PASMCs (Cascade Biologies, Inc., Portland, OR, USA) were cultured in DMEM containing 10% fetal bovine serum (FBS), 100 μg/mL penicillin, and 100 μg/mL streptomycin on 24-well plates or T25 culture bottles. Cells were used between passages 3 to 8. Confluent monolayers of PASMCs were cultured in FBS-free medium for 24 h and used for the experiments.

Highly purified CRP used in all studies was purchased from Sigma (St. Louis, MO, USA). As reported by the manufacturer, the purity of the compound was >75% (determined by SDS-PAGE and visualized by silver stain) and endotoxin level was <1.0 EU/1 μg of CRP (determined by the Limulus amoebocyte lysate method). Lipopolysaccharide (LPS) was purchased from R&D systems (Minneapolis MN, USA). DMEM and FBS were from GIBCO (Grand Island NY, USA). Atorvastatin was a gift obtained from Calbiochem (San Diego, CA, USA).

Measurements of IL-6 and MCP-1 Concentration

For dose-dependent studies on CRP-induced inflammatory response, cells were incubated in the presence of CRP 0, 5,10, 20, 50, 100, 200 μg/mL (n = 3 for each point, respectively) for 24 h. And then, 100 μg/mL of CRP was used for time-dependent studies (0, 3, 6, 9, 12, 18, 24 h, n = 3 for each time point). To evaluate the effect of atorvastatin on IL-6 and MCP-1 production, cells were pretreated with atorvastatin (0.1, 1, 10 μM) for 2 h, and then co-incubated with CRP (100 μg/mL) for 24 h (n = 3 for each time point).

Both IL-6 and MCP-1 concentrations in the supernatants were determined with ELISA Kits according to the manufacturer's instructions (R&D systems Inc, MN, USA). The absorbance at 450 nm for IL-6 and MCP-1 were measured, and their concentrations were determined by interpolation of standard calibration curves. The limits of detection of IL-6 and MCP-1 were 0.48–1500 pg/mL and 31.2–1000 pg/mL. Both intra- and inter-assay variability was <10%.

Real-Time Reverse Transcription and Polymerase Chain Reaction (RT-PCR)

Quiescent cells were incubated with CRP (0, 5, 10, 20, 50, 100 μg/mL). After incubated with CRP for 2 h, cells were washed with phosphate buffered solution (PBS). Total RNA was isolated with TRIzol reagent (GIBCO, Grand Island, NY, USA) according to the manufacturer's instructions.

cDNA was obtained by RT-PCR using 1 μg of total mRNA. Specific IL-6 and MCP-1 cDNA were obtained by PCR. The primers used were following: forward 5′-CCCAGTACCCCCAGGAGAAG-3′, reverse 5′-CGAGGATGTACCGAATTTGTTTG-3′; amplifying a 101 bp-length fragment for human IL-6; forward 5′-AACTGAAGCTCGCACTCTCTCG-3′, reverse 5′-TCAGCACAGATCTCCTTGGC-3′, corresponding to a 257 bp-length fragment for MCP-1. Human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified as a reference for quantification. The primers for GAPDH were: forward 5′-CCTGGCACCCAGCACAAT-3′, reverse 5′-CCGATCCACACGGAGTACTTG-3′, corresponding to a 374 bp-length fragment. The ratios between the amplified DNA fragments and GAPDH for each sample RNA were quantified.

Reverse transcription was performed using M-MLV (GIBCO) and 1 μg of the RNA samples from each culture conditions. The relative quantitative RT-PCR was performed in triplicate in 50 μl reactions by using an ABI PRISM 7900 HT sequence detector system (Applied Biosystems, Foster City, CA, USA). Samples were incubated at 50 °C for 2 min, followed by 95 °C for 10 min and then 40 cycles of 95 °C for 15 s, 60 °C for 1 min. SYBR-green chemistry was used to detect fluorescence and an internal standard (Applied Biosystems). A total of 3 different experiments were performed for each experimental condition.

Electrophoretic Mobility Shift Assay (EMSA)

The activity of NF-κB in nuclear extracts from cells was analyzed by EMSA. The cells starved in serum-free medium for 24 h were co-incubated with CRP at different concentrations (0, 5,10, 20, 50, 100 μg/mL) for 60 min. For evaluating effects of atrovastatin on NF-κB activation, the cells were pretreated with 10 μmol/L of atrovastatin for 2 h, and co-incubated with CRP100 μg/mL for 60 min. The cells co-incubated with 50 ng/mL LPS were served as positive control. Preparation of nuclear extract was performed according to methods described previously [27].

Nuclear proteins were subjected to EMSA using biotin-labeled NF-κB oligonucleotides (5′-ACTTGAGGGGACTTTCCCAGGC-3′) (Viagene). First, nuclear proteins were incubated with oligonucleotide for 30 min, and then subjected to gel electrophoresis and finally autoradiographed. For competitive experiment, the extracts were incubated with 100-fold excess unlabeled probe as a competitor.

Statistical Analysis

All data are presented as mean ± S.D. Differences in terms of IL-6 and MCP-1 expression of cells stimulated with different concentrations of CRP were determined by two-way ANOVA. If an F value was found to be significant, by a Student's t test with Bonferroni's correction. Differences of concentrations between two groups were compared using unpaired t test. A P value < 0.05 was considered statistically significant.

Results

Effects of CRP on IL-6 and MCP-1 Production

Quiescent human PASMCs were cultured in serum-free medium for 24 h, and then stimulated with different concentrations of CRP up to 24 h. As shown in Figure 1, CRP dose-dependently increased IL-6 and MCP-1 secretion with a peak at 200 μg/mL. Compared with control group, increased production of IL-6 and MCP-1 were detected at a concentration of CRP 10 μg/mL. After a 24-h incubation with CRP 200 μg/mL, the levels of IL-6 and MCP-1 were increased 3.8- and 2.9-fold compared with control groups (14204.03 ± 694.40 pg/mL vs 3718.06 ± 241.35 pg/mL; 20260.57 ± 1362.85 pg/mL vs 6976.60 ± 454.88 pg/mL, P < 0.01, respectively). In addition, CRP enhanced both IL-6 and MCP-1 production in PASMCs presented as a time-dependent manner (Figure 2A,B).

Figure 1.

Dose-dependent induction of IL-6 and MCP-1 secretion induced by CRP in human cultured PASMCs. Cells were incubated in the presence of different concentration of CRP (0, 5, 10, 20, 50, 100, 200 μg/mL) for 24 h (n = 3 for each points), and levels of IL-6 and MCP-1 in cultured supernatants were measured by ELISA. Results are mean ± SD, and the error bars indicate the SD of the mean. *P < 0.01 compared with control (CRP 0 point). #P < 0.05: Compared with control (CRP 0 point). IL-6 = interleukin-6; MCP-1 = monocyte chemoattractant protein-1; CRP = C-reactive protein; PASMCs = pulmonary arterial smooth muscle cells; SD = standard deviation.

Figure 2.

Time-dependent induction of IL-6 and MCP-1 secretion induced by CRP in human cultured PASMCs. Cells were seeded in 24-well plates, and incubated in serum-free medium in the presence of CRP (100 μg/mL) for different time points (0, 3, 6, 9, 12, 18, 24 h). The levels of IL-6 and MCP-1 in cultured supernatants were measured by ELISA. (A) Time course of production of IL-6 in the presence of CRP (n = 3 for each points). (B) Time course of production of MCP-1 in the presence of CRP (n = 3 for each points). Results are mean ± SD, and the error bars indicate the SD of the mean. *P < 0.01 compared with baseline. #P < 0.01 compared with baseline and control group. IL-6 = interleukin-6; MCP-1 = monocyte chemoattractant protein-1; CRP = C-reactive protein; PASMCs = pulmonary arterial smooth muscle cells; SD = standard deviation.

Effects of CRP on IL-6 and MCP-1 mRNA Expression

The effects of CRP on IL-6 and MCP-1 mRNA expression in human PASMCs were shown in Figure 3. Cells were incubated with different concentrations of CRP for 120 min, and then collected for quantification of mRNA expression of IL-6 and MCP-1. The results of relative quantitative RT-PCR showed that IL-6 and MCP-1 mRNA was expressed at high levels in serum-deprived, quiescent cells. After incubation with CRP (100 μg/mL) for 120 min, levels of IL-6 and MCP-1 mRNA expression were increased by 6.4- and 4.3-fold in stimulated cells compared to that in unstimulated cells, respectively.

Figure 3.

The levels of IL-6 and MCP-1 mRNA expression induced by CRP in human cultured PASMCs. Cells were incubated for 120 min in the presence of different concentrations of CRP (1, 5, 10, 20, 50, 100 μg/mL). The levels of IL-6 and MCP-1 mRNA expression were determined by RT-PCR (n = 3 for each point). Ctrl indicate control group and Atr do 10 μmol/L of atorvastatin group. Data are mean ± SD, and the error bars indicate the SD of the mean. *P < 0.01 compared with control group (CRP 0 point). #P < 0.001 compared with group in the presence of 100 μmol/L of CRP group. IL-6 = interleukin-6; MCP-1 = monocyte chemoattractant protein-1; PASMCs = pulmonary arterial smooth muscle cells; CRP = C-reactive protein; RT-PCR = real-time reverse transcription and polymerase chain reaction; SD = standard deviation.

Atorvastatin Decreases CRP-Induced IL-6 and MCP-1 Production

The possible impact of atorvastatin (0.1–10 μmol/L) on the production of IL-6 and MCP-1 enhanced by CRP was explored in the current study. As shown in Figure 4, atorvastatin decreased the levels of IL-6 and MCP-1 secretion in the supernatants of cultured PASMCs in the presence with 100 μg/mL of CRP in a dose-dependent manner, and 10 μmol/L of atorvastatin abrogated this effect on both IL-6 (14204.03 ± 850.68 pg/mL vs 3814.82 ± 128.50 pg/mL, P < 0.01) and MCP-1 secretion induced by CRP (19453.63 ± 1312.28 pg/mL vs 6292.23 ± 311.93 pg/mL, P < 0.01).

Figure 4.

Inhibiting effects of atrovastatin on the secretion of IL-6 and MCP-1 stimulated by CRP in human cultured PASMCs. Cells were pretreated with atorvastatin (0.1,1,10 μmol/L) for 2 h, and then co-incubated with 100 μg/mL of CRP for 24 h. The levels of IL-6 and MCP-1 in cultured supernatants were measured by ELISA (n = 3 for each points). Ctrl represent control group, and Atr 0.1, 10, 10 indicate the groups in the presence of 0.1, 10, 10 μmol/L of atorvastatin groups. Results are mean ± SD, and the error bars indicate the SD of the mean. *P < 0.01 compared with group in the presence of 100 μg/mL of CRP. #P < 0.01 compared with control group. IL-6 = interleukin-6; MCP-1 = monocyte chemoattractant protein-1; CRP = C-reactive protein; PASMCs = pulmonary arterial smooth muscle cells; SD = standard deviation.

Atorvastatin Inhibits IL-6 and MCP-1mRNA Expression and NF-κB Activation Induced by CRP

As shown in Figure 3, CRP also dose-dependently enhanced IL-6 and MCP-1 mRNA expression with a peak at 100 μg/mL detected by RT-PCR assay. However, our data indicated that 10 μmol/L of atorvastatin significantly suppressed CRP-induced IL-6 and MCP-1 mRNA expression (Figure 3). Additionally, to elucidate the mechanisms by which CRP stimulates the productions of IL-6 and MCP-1 of the cells, we also examined the effects of CRP on activation of NF-κB pathway, which was involved in CRP-induced cytokine upregualtion in monocytes and endothelial cells according to our previous studies [27]. We employed EMSA on nuclear extracts obtained from cells stimulated with or without CRP. As shown in Figure 5, NF-κB of cells was significantly activated after incubation with CRP, while no significant shift could be observed in control cells. Using 50 ng/mL of LPS as positive control, the data showed that the band could be abolished by 100-fold excess unlabeled NF-κB oligonucleotide probe. Therefore, we confirmed that these shifted bands were specific band of NF-κB. As indicated in Figure 5, the effects of CRP on NF-κB activation were also in a dose-dependent manner. However, atorvastatin significantly inhibited the NF-κB shifted band, suggesting that atorvastatin-mediated anti-inflammatory action was definitely associated with NF-κB pathway.

Figure 5.

The results of electrophoretic mobility shift assay for assessing the dose-dependent effects of CRP and atorvastatin (10 μmol/L) on NF-κB activity in human cultured PASMCs. Cells were stimulated with different concentrations of CRP (0, 5, 10, 20, 50, 100 μg/mL) or with medium alone (control) for 60 min. Nuclear proteins were extracted and subjected to EMSA using biotin-labeled NF-κB oligonucleotides for 30 min, then subjected to gel electrophoresis and autoradiographed. For competitive experiment, the extracts were incubated with 100-fold excess unlabeled probe as a competitor. LPS 50 ng/mL as positive control. 100 +Atr indicate 100 μg/mL of CRP + 10 μmol/L of atorvastatin group. Comp = competitive test; CRP = C-reactive protein; NF-κB = nuclear factor-κB; PASMCs = pulmonary arterial smooth muscle cells; LPS = lipopolysaccharide.

Discussion

The role of inflammation in pulmonary arterial disease and effect of statins on inflammatory response has been increasingly recognized. However, the possible impact and mechanism of statin on CRP-induced inflammatory response in PASMCs has less been investigated. In the present study, data suggested that CRP could indeed enhance the expression of IL-6 and MCP-1 in PASMCs, and atorvastatin could inhibit the CRP-induced expression of IL-6 and MCP-1, which were related to suppression of NF-κB pathway. Obviously, our study confirmed and extended the previous studies.

Over the past few years, it has become clear that inflammation plays an important pathophysiological role in the development of atherosclerosis. In particular, CRP, formerly regarded as solely a biomarker for inflammation, is currently considered as a prominent partaker in endothelial dysfunction and atherosclerosis. Interestingly, recent data have also reported that plasma levels of CRP were higher in patients with PAH related to conditions such as systemic lupus erythematosus [12] and chronic obstructive pulmonary disease, and was an independent predictor of systolic PAP [13]. These results suggested that CRP might be involved in the pathogenesis of pulmonary hypertension. Accordingly, the effects of CRP on tissues of pulmonary arteries and its possible mechanisms are worthy of further investigation.

MCP-1 is recognized as a potent chemotactic and activating factor for monocytes and leukocytes, and is produced by various cell types within the arterial wall, including macrophages, vascular smooth muscle cell, endothelial cells, and fibroblasts. Besides the effect of induction of monocyte recruitment, MCP-1 was also secreted by pulmonary arterial endothelial cells, and had a direct role in monocyte infiltration into the injured vessel wall as well as in proliferation of PASMCs [28]. In animal models secondary to chronic hypoxic [7], monocrotaline-induced [8,9] or chronic air embolism [10], the elevated levels of IL-6 and MCP-1 has definitely demonstrated to be linked with PAH. Moreover, injection with recombinant IL-6 could cause PAH in rats, whereas inhibition of IL-6 with dexamethasone might protect from PAH induced by monocrotaline. Additionally, transduction with naked plasmid encoding a 7-NH terminus-deleted dominant negative inhibitor of the MCP-1 gene could significantly inhibit the progression of MCP-1-induced PAH [11]. Those data might provide an insight into a contribution of IL-6 and MCP-1 to the pathogenesis of PAH. That is a reason why we choose IL-6 and MCP-1 as inflammatory markers to perform this in vitro study. In the present study, our data showed that CRP could enhance production of IL-6 and MCP-1 in a dose-dependent manner, which was associated with activation of the NF-κB pathway in cultured human PASMCs.

In addition, our previous studies have demonstrated that statins could inhibit inflammatory response in cultured human monocyte induced by CRP as well as LPS in a dose-dependent manner [26], and increased serum level of anti-inflammatory cytokine such as IL-10 in patients with unstable angina [29]. Similarly, a few data have showed that statins could prevent from the development [19–21] and progression [22–24] of PAH induced by hypoxia, high pulmonary blood flow, and acute pulmonary embolism in animal models. At same time, in human studies statin therapy showed a definitely beneficial effect in patients with idiopathic and secondary causes of PAH. Treatment with simvastatin (20 to 80 mg/day) improves 6-min walk performance, cardiac output, and decreases right ventricular systolic pressure. Moreover, statin also improved the survival in patients with world health organization class IV PAH. [25]. The underlying mechanisms of those beneficial effects have been demonstrated to be associated with cell apoptosis [22,23,30], reduction of neointimal smooth muscle cell proliferation [19,31], and improvement of nitric oxide synthasis expression in endothelial cells [31]. Recently, data from Lee et al. [32] have also confirmed a direct evidence of statin on PAH by depending on its antiinflammation action. In their study, simvastatin prevented from the pulmonary vascular remodeling and the changes in endothelial nitric oxide synthases expression induced by cigarette smoking, and inhibited peribronchial and perivascular infiltration of inflammatory cells and induction of matrix metalloproteinase-9 (MMP-9) activity in lung tissue. These findings suggested that statin ameliorated the structural and functional derangements of the lungs, partly by suppressing inflammation response in pulmonary vascular wall.

Although available data suggested the beneficial impact of statins on pulmonary arterial disease, the involving signal pathway of statin-mediated beneficial effects on PAH has not been well defined. As a key transcription factor in inflammatory responses, NF-κB controls the transcription of many genes with an established role in atherosclerosis, such as cytokines, and also modulates proliferation and apoptosis in many cells. A previous study indicated that NF-κB inhibitor, pyrrolidine dithiocarbamate decreased vascular cell adhesion molecule-1 (VCAM-1) expression and macrophage infiltration, and ameliorated pulmonary hypertension in a rat model with PAH [18], suggesting that the NF-κB nuclear localization was also associated with the development of PAH. The data of the present study showed that CRP induced the activation of NF-κB pathway by which subsequently increased the expression of IL-6 and MCP-1 in PASMCs. However, pretreatment with atorvastatin could dose- dependently inhibit CRP-induced production of IL-6 and MCP-1 in PASMCs by attenuating activation of NF-κB pathway. Consequently, the data confirmed and extended the previous studies regarding the mechanisms in CRP-related pathogenesis of pulmonary artery disease, and provided an insight into the statin-mediated antipathogenesis of pulmonary arterial disease. In a word, our present study suggested that NF-κB pathway might represent a promising target for controlling CRP-induced inflammatory response in pulmonary arterial disease.

Acknowledgments

This article is partly supported by National Natural Scientific Foundation (30670861, 30871055), Beijing Natural Scientific Foundation (7082081), and Specialized Research Fund for the Doctoral Program of Higher Education of China (20060023044, 20070023047), and Grant of National Project in the 11th five-year period entitled “Research on strategy of diagnosis and treatment of pulmonary Arterial hypertension (2006BAI01A07).”

Conflict of Interest

The authors have no conflict interest.

Ancillary