Cigarette Smoke Exposure Increases Glucose-6-phosphate Dehydrogenase, Autophagy, Fibrosis, and Senescence in Kidney Cells In Vitro and In Vivo

Cigarette smoke (CS) is a risk factor for chronic obstructive pulmonary disease. We attempted to investigate fully the possible effects of CS on kidney cells. We found that the viability of a human kidney proximal tubular epithelial cell line (HK-2 cells) was decreased after treatment with CS extract (CSE). In particular, the effects of CSE at low concentrations did not change the expression of apoptosis and necrosis. Furthermore, CSE increased autophagy- and fibrosis-related proteins in HK-2 cells. Senescence-related proteins and the senescence-associated secretory phenotype (SASP) increased after HK-2 cells were treated with CSE. In addition, both RNA sequencing and gene set enrichment analysis data revealed that glucose-6-phosphate dehydrogenase (G6PD) in the reactive oxygen species (ROS) pathway is responsible for the changes in CSE-treated HK-2 cells. CSE increased G6PD expression and its activity. Moreover, the inhibition of G6PD activity increased senescence in HK-2 cells. The inhibition of autophagy reinforced senescence in the CSE-treated cells. In a mouse model of CS exposure, CS caused kidney damage, including tubular injury and glomerulosclerosis. CS increased fibrosis, autophagy, and G6PD expression in kidney tissue sections. In conclusion, CS induced G6PD expression, autophagy, fibrosis, and senescence in kidney cells. G6PD has a protective role in CS-induced nephrotoxicity.


Introduction
One of the main causes of death, chronic obstructive pulmonary disease (COPD), occurs in the United States [1] and worldwide [2]. Clinically, cigarette smoking (CS) is a risk factor leading to the progress of COPD [3]. Fibroblasts in the lung play a vital role in repair, regeneration, and lung homeostasis [4]. Recent studies have indicated that lung fibroblasts of patients with COPD display a decreased growth rate [5]. CS reduces the proliferation of lung fibroblasts and upregulates pathways related to cellular senescence [5] and the p53 [6], p16, and p21 retinoblastoma protein pathways [7]. Moreover, CS also induces senescence-associated secretory phenotype-(SASP-) related inflammation in human epidermal keratinocytes and skin [8]. Apoptosis resulting from smoke extract-induced COPD has been observed in in vitro and in vivo studies [9]. Furthermore, CS, nicotine, and cotinine affect red blood cell hemolysis [10]. Glucose-6-phosphate dehydrogenase (G6PD) deficiency causes substantial oxidant damage to the erythrocyte membrane [11]. A previous study reported that CS [12] and nicotine [13] increase oxidant stress to red blood cells in healthy donor volunteers. In addition, CS also induces oxidative stress as a result of reactive oxygen species (ROS) in the brain [14]. In other cases, ROS have been shown to activate transforming growth factor β (TGF-β) in the modulation of profibrotic effects [15]. ROS accumulation or antioxidant depletion occurs could destroy to cellular elements, containing DNA, RNA, proteins, lipids, and carbohydrates [16]. Early reports have shown that heat shock protein 27 (Hsp27) is a neuroprotective biomarker in ischemic stroke [17]. On the other hand, intraperitoneal injection of recombinant soluble Klotho protein improves the premature aging-related phenotype in mice with the homozygous mutated allele [18]. Klotho reduces kidney senescence and fibrosis [19] by targeting mitochondrial dysfunction in renal tubular cells [20]. CS extract (CSE) reduces the expression and secretion of Klotho in alveolar macrophages and airway epithelial cells in COPD patients [21].
CSE not only induces oxidative stress but also fibrosisrelated gene expresses in orbital fibroblasts in Graves' ophthalmopathy patients [22]. Autophagy is an important and conserved "self-cleansing" pathway [23], and other studies have shown that fibrosis is often companied by autophagy [24,25]. In the kidney, autophagy can protect the proximal tubule from damage [26] to overcome many types of kidney injury [27], aging [28], and disease [29]. Autophagy is an essential cellular process that promotes cell survival by removing protein aggregates during kidney injury [30]. However, autophagy also promotes cell death or enhances apoptosis [31]. Therefore, autophagy has two contrasting outcomes in response to stress [32]. Studies have shown that CS causes autophagy [33] and accelerates lung aging via autophagy [34]. CSE-induced autophagy regulates many cellular processes such as FOXO transcription factors in human lung adenocarcinoma cells (A549) [35] and Galectin-3 in endothelial progenitor cells [36]. In addition, SIRT1 is downregulated by autophagy in senescence and aging [37]. On the other hand, Hsp27 phosphorylation plays a crucial role in the activation of G6PD to reduce cerebral ischemia/reperfusion injury in male Wistar rats [38]. G6PD is a major source of NADPH, which drives many essential cellular processes including antioxidant pathways [39] such as the suppression of oxidative stress in cerebral ischemic male Sprague-Dawley rats [40].
Many diverse diseases may lead to chronic kidney disease (CKD) via irreversibly impaired formation or dysfunctions of the kidney, such as fibrosis [41]. Recently, a metaanalysis suggested that CS is an independent risk factor in the general adult population with CKD [42]. The database of the Korean genome and epidemiology study also revealed that the healthy middle-aged adults who smoke have a high risk of CKD [43]. COPD patients have shown renal function worsening [44]. Additionally, tobacco CS promotes kidney injuries related to biochemical changes in male adult Wistar rats [45]. In this study, we examined the effect of CS on kidney cells in vitro and in vivo. Furthermore, gene set enrichment analysis (GSEA) was performed after HK-2 cells were treated with CSE. We also observed autophagy, fibrosis, senescence, and ROS generation after CS treatment.

Material and Methods
2.1. Cell Line. HK-2 cells, the proximal tubular epithelial cell line from human kidney, were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cells were kept in keratinocyte-serum-free medium (K-SFM) with bovine pituitary extract (BPE) and human recombinant EGF (Invitrogen, CA), and the cells were incubated at 37°C and 5% CO 2 . The culture medium was refreshed two or three times per week. The ATG5 KD HK-2 cells were incubated and maintained in the K-SFM medium with rEGF and BPE at 37°C with 5% CO 2 and cultured every two or three days. Lentivirus with control shRNA and ATG5 siRNA were purchased from the National RNAi Core Facility at Academia Sinica in Taiwan [46].

Preparation of Cigarette Smoke
Extract. CSE solutions were prepared using a modification of standardized methods [47]. Three cigarette types (Longlife, Taipei, Taiwan; 11 mg of tar and 0.9 mg of nicotine) were subsequently collected by a liquid impinger device and then mixed with 15 ml of K-SFM. One of the cigarettes was dissolved in 5 ml of K-SFM, which was designated as 100% CSE solution.
2.3. Cell Viability Assay. Cell proliferation was accessed with sulforhodamine B (SRB, Sigma-Aldrich, St. Louis, MO). HK-2 cells (5 × 10 3 /well) were plated in 96-well plates and cultured with or without CSE solutions in a 37°C and 5% CO 2 incubator overnight. After 24 h of incubation at 37°C and 5% CO 2 , the plates were moved out, washed with PBS twice, and then fixed with iced 10% trichloroacetic acid (TCA, Sigma) at 4°C for 1 h. Each well was washed 2 times with distillation-distillation H 2 O, and then, 0.1% SRB/1% acetic acid was incubated for 1 h. The wells were rinsed 2 times with 1% acetic acid and dried in an oven at 60°C for 20 min. In the end, the dye form SRB-positive cells were redissolved in 20 mM Tris buffer (Sigma) for 30 min. The   , and GAPDH (Proteintech). After the hybridization process with the abovementioned antibodies on the PVDF membrane, the membrane was rinsed with TBS-T for 15 min three times. Subsequently, the PVDF membrane was further treated with anti-mouse (Jackson) or anti-rabbit (Jackson) secondary antibody for 2 h and rinsed with TBS-T for 15 min over three times. The protein bands of the PVDF membrane were visible by performing an enhanced chemiluminescence system (Amersham, Little Chalfont, United Kingdom).
2.7. RNA Sequencing and Analysis. RNA sequencing used to characterize and analyze the transcriptome (RNA sequenc-ing, Tools, Taiwan). Briefly, the purity and quantification of RNA were detected with SimpliNano™-Biochrom Spectrophotometers (Biochrom, MA, USA). The levels of RNA degradation and integrity were detected by a BiOptic Qsep100 DNA/RNA Analyzer (BiOptic Inc., Taiwan). The sequencing library was established with the KAPA mRNA HyperPrep Kit (KAPA Biosystems, Roche, Basel, Switzerland). mRNA was extracted from total RNA with magnetic oligo-dT beads and incubated at a high temperature in KAPA buffer that contained magnesium. cDNA was generated with random hexamer priming. cDNA fragments with a length of 300~400 bp were selected, and library fragments were extracted with the KAPA Pure Beads system (KAPA Biosystems, Roche, Basel, Switzerland). The library was increased with KAPA HiFi HotStart ReadyMix (KAPA Biosystems, Roche, Basel, Switzerland). Finally, the library was extracted with the KAPA Pure Beads system and qualified with the Qsep100 DNA/RNA Analyzer (BiOptic Inc). The library data were detected with high-throughput sequencing (Illumina NovaSeq 6000 platform), which was transformed into raw sequenced reads with CASAVA base calling and then stored in FASTQ format. The FASTQ files were used with FastQC and MultiQC [48]. The raw paired-end reads were filtered with Trimmomatic (v0.38) [49]. The obtained high-quality data were aligned to the reference genome  Oxidative Medicine and Cellular Longevity  7 Oxidative Medicine and Cellular Longevity (e.g., H. sapiens, GRCh38) with HISAT2 software (v2.1.0) [50,51]. featureCounts (v1.6.0) was used to count the read numbers mapped to individual genes [52]. The RNA series dataset was uploaded to the Gene Expression Omnibus (Accession: GSE182541). GSEA was analyzed with 1000 permutations to identify enriched biological functions and activated pathways from the molecular signature database [53] (MSigDB) (https://www.gsea-msigdb.org/gsea/msigdb).

Quantitative Polymerase Chain Reaction (Q-PCR).
Total RNA of the kidney will be extracted using the TRIzol reagent (Invitrogen). Purity and quantification of RNA will be detected. Complementary DNA (cDNA) will be synthesized using the Easy Fast RT Kit (TOOLS, Taiwan). Q-PCR will be detected using SYBR Green (TOOLS, Taiwan). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) will be used as an internal control. The 2 −ΔΔCt method will be used to calculate the expression changes. All primers are listed as follows: G6PD, Hsp27, and GAPDH.
2.9. Senescence β-Galactosidase Staining. The cells were treated with CSE for 24 h, washed, fixed, and cultured in a 37°C/5% CO 2 incubator with X-gal chromogenic substrate at pH 5.5 overnight by following the protocol for SA-β-gal staining (BioVision, Milpitas, CA). The images for β-galactosidase were collected using a digital microscope. The positive cell intensity was counted in 3 fields of view (>50 cells/ field). Polydatin (MCE, Monmouth Junction, NJ) is an inhibitor of G6PD activity [54].
2.10. Enzyme-Linked Immunosorbent Assay. The amount of the NADPH-producing enzyme G6PD was measured using ELISA kits specific for human G6PD according to the manufacturer's protocol (Cayman Chemical, Ann Arbor, MI). The fluorescent product was measured under an excitation wavelength of 530/540 and an emission wavelength of 585-595 nm.
2.11. BrdU Cell Proliferation Assay. HK-2 cells were plated in a 96-well plate and incubated. BrdU was measured using BrdU Cell Proliferation Assay Kit (BioVision). Briefly, cells were added 1x 5-bromo-2-deoxyuridine (BrdU) solution and incubated plate at 37°C. Cells were fixed and denatured. Cells were hybrid with BrdU detection antibody solution.
Finally, cells were added 3,3 ′ ,5,5 ′ -tetramethylbenzidine (TMB) substrate and measured the absorbance at 650 nm.  Details of the CS exposure system were reported in a previous study [55]. Briefly, the system consisted of a CS generator in a whole-body exposure chamber (TECNIPLAST, VA, Italy) with a particulate matter monitor. A side stream was placed into the whole-body exposure chamber at a flow rate of 15 l/min. Sixteen commercial cigarettes (Longlife) were combusted for 8 h/day and 5 days/week for 4 months. The mass concentration of particulate matter of <2.5 μm in aerodynamic diameter was monitored using a DustTrak monitor (TSI, Shoreview, MN). The mice were sacrificed by CO 2 , and the kidneys were excised and fixed with 10% neutral formalin.
2.13. Histological Analysis. The kidney tissues were embedded, dehydrated, sectioned into 2 μm thick slices slicing a microtome, and then stained with hematoxylin and eosin (H&E) (Sigma) for histological analysis. The glomerulosclerosis and tubular injury scores were measured. Details of the glomerulosclerosis and tubular injury scores are provided in a previously reported study [56].
2.14. Immunohistochemical (IHC) Staining Analysis. Kidney sections were maintained in an oven at 60°C. The kidney sections were sequentially washed with xylene (Sigma), 100% ethanol (Sigma), 95% ethanol, and 75% ethanol. Finally, the kidney sections soaked in MQ water and boiled with sodium citrate buffer (0.01 M, pH 6.0, 1% Tween 20). The sections then washed with PBS, soaked in 3% H 2 O 2 / methanol, and finally with PBS. UltraVision protein block buffer was applied to analyze the kidney after treatment with G6PD (Proteintech), LC3 (MBL) or β-gal (Invitrogen) antibody in 3% BSA overnight at 4°C. The sections were washed with PBS, treated with Trekkie Universal Link for 20 min, and then mixed with poly-HRP reagent for 20 min. The DAB coloring agent was used to stain the sections, followed by placement in MQ water to terminate the reaction. For the next step, hematoxylin was also used as a contrast dye for the second staining assay. In the end, the mounting buffer was added to the kidney sections, which were covered with a cover slip. Masson's trichrome staining was used according to the protocol (TRM-2-IFU, ScyTek). After the sections solidified with mounting buffer, the slices were recorded with Motic Digital Slide Assistant (Motic VM3.0, New York, NY).

Statistical Analysis.
The results were analyzed by SPSS (SPSS Software, CA, San Diego) and plotted as the mean ± standard deviation. The statistical significance between groups was determined by Student's t-test. Comparisons of three or more groups were calculated by ANOVA. Significance was confirmed at p < 0:05.

Cell Viability, Cell Death, and Apoptosis-Related Protein
Expression in CSE-Treated HK-2 Cells. The viability of CSEtreated HK-2 cells was significantly decreased in a concentration-dependent manner, as shown in Figure 1(a). After HK-2 cells were treated with CSE at low concentrations of 0.1%, 0.2%, 0.4%, and 0.6%, the cell viability decreased to 94.6%, 92.8%, 81.3%, and 67.9%, respectively. Flow cytometry analysis revealed that the CSE-treated cells at low concentrations of 0.1%, 0.2%, 0.4%, and 0.6% had no significant differences in either index of apoptosis or necrosis (Figures 1(b) and 1(c)). In addition, LDH assay revealed that the necrosis index did not have any significant differences (Figure 1(d)). Moreover, western blotting analysis showed that low concentrations of CSE did not increase Bax and cleaved caspase 3 protein expression (Figures 1(e) and 1(f)).

Expression of Autophagosome-Related and Fibrosis-Related Proteins in HK-2 Cells after Treatment with CSE.
The expression levels of LC3 determined by immunofluorescence were concentration-dependent (Figure 2(a)). Statistical analysis also showed that LC3 levels were CSE concentration-dependent in HK-2 cells (Figure 2(b)). Approximately 35 ± 2:5% of HK-2 cells expressed LC3 9 Oxidative Medicine and Cellular Longevity signals after treatment with 0.6% CSE. Furthermore, CSEtreated HK-2 cells revealed higher expression of the autophagy-related proteins LC3, p62, ATG5, and SIRT1, as determined by western blotting (Figure 2(c)). The expression levels of LC3, p62, ATG5, and SIRT1 were also concentration-dependent in HK-2 cells after CSE treatment (Figure 2(d)). On the other hand, HK-2 cells exposed to CSE exhibited higher expression of the fibrosis-related proteins collagen type 1, PAI-1, and CTGF, as determined by western blotting (Figure 2(e)). Statistical analysis also revealed that the levels of collagen type 1, PAI-1, and CTGF after treatment with CSE in HK-2 cells were also concentration-dependent (Figure 2(f)).

RNA Sequencing, Gene Set Enrichment Analysis, and
Interpretative Phenomenological Analysis in CSE-Treated HK-2 Cells. The RNA sequencing data after treatment of HK-2 cells with CSE are shown in Figure 4(a). The most upregulated gene was MMP3 in CSE-treated cells, while the most downregulated gene was SEMA5B. In addition, the data were further analyzed by GSEA (Figure 4(b)). The results showed higher expression of the ROS pathway in HK-2 cells after treatment with CSE ( Figure 4(b)). The enrichment plot showed higher enrichment of the ROS pathway after cells were exposed to CSE (Figure 4(c)). Moreover, ROS-related gene expression is shown in Table 1. The most upregulated gene was G6PD in HK-2 cells after CSE treatment as shown in Table 1.

Expression
Profile of Heat Shock Protein 27, Glucose-6phosphate Dehydrogenase, and Senescence-Associated β-Galactosidase Assay Regulation with Autophagy in CSE-Treated HK-2 Cells. Hsp27 and G6PD were observed to be concentration-dependently increased, as determined by real-time polymerase chain reaction (Q-PCR) assay ( Figure 5(a)) and western blotting ( Figure 5(b)). The expression of Hsp27 and G6PD in HK-2 cells after treatment with CSE was also concentration-dependent ( Figure 5(c)). The G6PD activity after HK-2 cell treatment with CSE was also concentration-dependent ( Figure 5(d)). G6PD activity was higher in HK-2 cells after CSE treatment at concentrations of 0.6% and decreased after polydatin treatment at concentrations of 10, 20, 50, and 100 μM ( Figure 5(e)). 3.6. Tubular Injury and Glomerulosclerosis Score Analysis of Senescence, Senescence-Associated β-Galactosidase, Autophagy Protein LC3, and Glucose-6-phosphate Dehydrogenase after treatment of Mice with CS. A mouse model was established by exposure of mice to CS for 4 months. H&E staining displayed varying degrees of tubular injury and glomerulosclerosis when comparing the CS-treated group to the normal group (Figure 6(a)). The tubular injury and glomerulosclerosis scores were significantly increased after treatment of mice with CS ( Figure 6(b)). The kidney sample also revealed varying degrees of fibrosis staining with Masson's trichrome (purple) (Figure 6(c)). IHC staining showed higher expression of SAβgal, LC3 and G6PD than that of the normal group at 4 months (Figures 6(d)-6(f)).

Discussion
A previous study showed that CS altered cell viability in gingival mesenchymal cells at a concentration of 250 μg/ml [58]. The cell viability of mouse embryonic fibroblasts and NIH3T3 cells decreased below 50% after exposure to 400 μl of 4% CSE solution [59]. In addition, the cell viability of human lung bronchial epithelial cells (BEAS-2B) decreased to 50% after treatment with 10-20% CSE [60]. Our results showed that cell viability decreased by over 50% in HK-2 kidney cells after treatment with CSE (0.8% and 1%) (Figure 1(a)). Early reports indicated that apoptosis was significantly induced in BEAS-2B cells by CSE [61]. However, our results showed that low concentrations CSE (0.1%-0.6%) did not induce apoptosis in HK-2 cells (Figures 1(b)-1(e)). A previous study showed that the CSE-induced autophagy in A549 cells is associated with many cellular processes [35]. Our results indicated that CSEtreated HK-2 cells not only induced autophagy but also induced SIRT1 after CSE (0-0.6%) exposure (Figures 2(a)-2(d)). Hence, our results indicated that autophagy was regulated by SIRT1, which has been reported in other studies [62][63][64].
CSE is a risk factor for the development of lung fibrosis [65]. Renal fibrosis is involved in various kidney diseases [66]. Previous studies have demonstrated that through the autocrine and paracrine stimulation of cells by TGF-β1, CTGF is released and synthesized, which plays a role in fibrogenesis [67]. Furthermore, PAI-1 is the major physiologic inhibitor of the plasmin-based pericellular cascade and a causative factor in the fibroproliferative disorders [68]. The upregulation of CTGF and PAI-1 caused extracellular matrix (ECM) accumulation [69,70]. In the current study, CSE exhibited higher expression of the fibrosisrelated proteins including collagen type 1, PAI-1, and CTGF in HK-2 cells (Figures 2(e) and 2(f)). In in vivo study, the  Figure 7: Schematic of the putative mechanism illustrating the CSE-induced G6PD, autophagy-related, fibrosis-related, and senescencerelated protein expressions in kidney cells. Kidney cells were induced by CSE via several related pathways that increase the expression of the fibrosis-related proteins, autophagy-related proteins, and senescence-related proteins in kidney cells. CSE regulates senescence and fibrosis via inhibition of Hsp27, G6PD, Klotho, and autophagy.

12
Oxidative Medicine and Cellular Longevity significant accumulation of collagen fibers in the kidney tissues of the CS group ( Figure 6(c)). These results indicated that CS may cause kidney fibrosis. Previous research has shown that CS induces SASP-related inflammation in human epidermal keratinocytes and skin [8]. The expression of fibrosis-related genes is induced in orbital fibroblasts from patients with Graves' ophthalmopathy [22]. In addition, previous study revealed that CS reduced the proliferation of lung fibroblasts by upregulating signaling pathways such as cell senescence, p53, and p16-retinoblastoma [5]. The p53 [6], p16, and p21 pathways [7] were related to cellular senescence, including fibrosis accompanied by senescence, senescence-related proteins, and SASP-related inflammation (Figures 3(a)-3(f) and Figure 6(d)). Early reports showed that CSE reduces the expression and secretion of Klotho in alveolar macrophages and airway epithelial cells in COPD patients [21]. The present study demonstrated that Klotho significantly decreased in human kidney cells after the CSE (0%-0.6%) treatment ( Figure 3(c)).
Recently, a meta-analysis suggested that CS is an independent risk factor for the general adult population with CKD [42] and healthy middle-aged adults [43]. In addition, GSEA revealed that three pathways were involved in CS-treated BEAS-2B cells, namely, cell matrix adhesion and the TGF-β receptor signaling pathway, RNA catabolic processes, and the regulation of cell cycle phase transition, as well as calcium-mediated signaling and regulation of cell-cell adhesion [71]. Deficiency of G6PD in the erythrocyte membrane causes substantial oxidant damage [11]. G6PD is a major source of NADPH that is involved in antioxidant pathways [39]. Furthermore, GSEA showed that the ROS pathway is the primary regulator in HK-2 cells after treatment with CSE. In particular, the expression of G6PD increased in the ROS-related gene expression pathway after treatment with CSE (Figures 4(a)-4(c) and Table 1). G6PD and Hsp27 were highly expressed in CSE-treated HK-2 cells (Figures 5(a)-5(c)). On the other hand, Hsp27 phosphorylation plays an important role in the activation of G6PD [38], and the phosphorylation of G6PD results in a reduction of NADPH, subsequently causing oxidative stress which may lead to metabolic syndromes [72,73]. Polydatin is a new inhibitor of G6PD that can block the pentose phosphate pathway [54]. Our results showed that polydatin decreased the activity of G6PD in a dose-dependent manner ( Figure 5(e) and Table 1). The expression of senescence-related factors was increased after treatment with polydatin in HK-2 cells (Figure 5(f)). These data showed that G6PD plays an important role in the protection of kidney cells. A previous study indicated that CS accelerated lung aging [34] and kidney injury [74] via autophagy. Our results showed that CSE induced autophagy (Figures 2(a)-2(d)). Here, we used ATG5 shRNA to increase senescence expression in HK-2 cells (Figure 5(g)) and found that CSE-induced autophagy may inhibit senescence and has a protective role in kidney cells.

Conclusions
We found that CSE induced autophagy, fibrosis, senescence, and SASP in kidney cells (Figure 7). In contrast, Klotho expression was decreased in kidney cells after CSE treatment. Furthermore, RNA sequencing and GSEA revealed that G6PD played an important role in ROS pathway regulation in kidney cells after CSE exposure. G6PD expression and G6PD activity increased in CSE-treated kidney cells. In addition, G6PD inhibited senescence in kidney cells. In an animal model after CS exposure for 4 months, CS caused tubular injury and glomerulosclerosis and induced fibrosis, autophagy, and G6PD. In conclusion, CS induced G6PD, autophagy, fibrosis, and senescence and decreased Klotho in kidney cells. These findings offer more precise molecular mechanism of CS and the chance to find potential preventive or therapeutic strategies for CS-related renal injury. In the current study, we focus on in vitro study and an animal model. In the future, we hope to utilize clinical data and samples for validation of the research that we currently perform on in vitro and animal models.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon reasonable request.

Disclosure
Experiments and data analysis were performed in part through the use of the Medical Research Core Facilities Center, Office of Research & Development at China Medical University, Taichung, Taiwan.