Advanced Oxidation Protein Products Induce Hypertrophy and Epithelial-To-Mesenchymal Transition in Human Proximal Tubular Cells through Induction of Endoplasmic Reticulum Stress

Chronic kidney disease (CKD) often leads to end-stage renal disease and requires renal replacement therapy, and it is a global public health problem. CKD is known to progress to end-stage renal disease through the final common pathway of renal interstitial fibrosis [1]. Although the cellular mechanisms that facilitate tubulointerstitial fibrosis and the contribution of tubular epithelial-to-mesenchymal transition (EMT) to renal interstitial fibrosis are still controversial, tubular EMT is through to play a role in the pathogenesis of CKD [2,3,4,5,6,7,8]. Hypertrophy of tubular cells is the main cause of renal hypertrophy, because the tubulointerstitium makes up almost 90% of the kidney volume. Moreover, previous studies demonstrated that initial adaptive tubular hypertrophy will become maladaptive and lead to tubular atrophy and interstitial fibrosis, which is related to the pathogenesis of CKD [9]. Therefore, preventing the EMT and hypertrophy of renal tubular epithelial cells is critical for preventing and interrupting the accelerated progression of CKD.

Advanced oxidation protein products (AOPPs), which were first reported by Witko-Sarsat et al. in 1996, are dityrosine-containing and cross-linking protein products that are mainly formed as a result of the reaction of plasma albumin with chlorinated oxidants during oxidative stress [10,11]. Previous studies have suggested that AOPPs can be observed in the plasma of patients with early-stage CKD and that elevated levels of AOPPs are correlated positively with lesions of renal function [12]. Furthermore, growing evidence suggests that AOPPs play a crucial role in the progression of CKD [13,14,15]. AOPPs have been reported to induce podocyte apoptosis, renal tubular epithelial injury, and mesangial cell proliferation and differentiation [13,14,15,16]. However, although numerous factors might induce hypertrophy in renal tubular epithelial cells and initiate EMT in the kidney, whether AOPPs induce the hypertrophy and EMT of renal tubular epithelial cells remains poorly understood.

The endoplasmic reticulum (ER) functions as the major processing center where newly synthesized, secreted and membrane-associated proteins are correctly folded and assembled. Adverse environmental conditions such as hypoxia affect the function of the ER and cause an imbalance between the protein-folding capacity of the ER and its protein content, which results in the accumulation of unfolded or misfolded proteins in the ER lumen and leads to ER stress [17]. Unfolded protein response, which triggers both adaptive and apoptotic pathways, is initiated after ER stress [18]. ER stress has been increasingly recognized to participate in the development or pathology of several kidney diseases such as diabetic nephropathy, renal tubular interstitial lesions induced by toxins or drugs, renal ischemia-reperfusion injury, and CKD [19,20,21]. Moreover, AOPPs have been shown to induce inflammation in adipocytes by inducing ER stress in vitro [22]. However, whether AOPPs induce ER stress and how ER stress affects renal tubular epithelial cells are questions that remain poorly studied.

In this study, we investigated whether AOPPs induce hypertrophy and EMT in HK-2 cells (a human proximal tubular epithelial cell line), and whether the process is mediated by ER stress. Moreover, we examined whether oxidative stress is involved in this process.

AOPPs-bovine serum albumin (BSA) was prepared as previously described. Briefly, a BSA solution was exposed to HOCl at a molar ratio of 1:140 for 30 min at room temperature and then dialyzed against PBS at 4ºC for 24 h to remove free HOCl. In the control incubation, native BSA was dissolved in PBS alone. All preparations were passed through a Detoxi-Gel column (Pierce, Rockford, IL, USA) in order to remove contaminating endotoxins. The endotoxin levels in the preparations were measured using the Amebocyte lysate assay kit (Sigma, St. Louis, MO, USA) and were determined to be <0.025 EU/mL. The AOPP content was determined by measuring the absorbance at 340 nm in an acidic condition and was calibrated using chloramine-T in the presence of potassium iodide.

HK-2 cells were purchased from American Type Culture Collection (Rockville, MD, USA) and cultured Dulbecco's modified Eagle's medium (DMEM/F12) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and maintained at 37°C in humidified air/5% CO₂. Cells were trypsinized using 0.25% trypsin-EDTA and then incubated in 6-well plates until they reached approximately 80% confluence. All experiments were performed using the differentiated HK-2 cells that were 70%-80% confluent. In the blocking experiments, cells were pretreated with the indicated inhibitors or inducer and then incubated with or without AOPPs until the end of the experiments. The inhibitors include diphenyleneiodonium (DPI) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), salubrinal (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and superoxide dismutase (SOD) (Cayman, USA) and the inducer is thapsigargin (Santa Cruz Biotechnology, Santa Cruz, CA, USA).

Total RNA was extracted from control and treated HK-2 cells by using TRIzol reagent (Invitrogen, USA) according to the manufacturer's protocol. Total RNA was reverse transcribed by using the M-MLV reverse transcriptase kit according to the manufacturer's instructions, and the cDNA was synthesized using TRIzol and Oligo dT (Invitrogen). Quantitative real-time PCR was performed using appropriate primers in a PCR-system-9700 and then the products were examined using the SYBR Green assay (TaKaRa, Japan). The amplification conditions were the following: 95ºC for 2 min, followed by 40 cycles of 95ºC for 30 s and 60°C for 35 s. The following sets of primers were used: β-actin: 5'-TGGCACCCAGCACAATGAA-3' (forward) and 5'-CTAAGTCATAGTCCGCCTAGAAGCA-3' (reverse); CCAAT/enhancer-binding protein-homologous protein (CHOP): 5'-CCTGGAAATGAAGAGGAAGAATCAA-3' (forward) and 5'-GGAGGTGCTTGTGACCTCTG-3' (reverse); glucose-regulated protein (GRP) 78: 5'-ACCTACTCCTGCGTCGGCGTGTT-3' (forward) and 5'-CGG CATCGCCAATCAGACGTTCC-3' (reverse); p27: 5'-CACAGTGAGGGGAAGCCC CTGAA-3' (forward) and 5'-CAAGCCTCTTCCAGATCCTTATGC-3' (reverse); α-smooth muscle actin (α-SMA): 5'-TACTACTGCTGAGCGTGAGA-3' (forward) and 5'-CATCAGGCAACTCGTAACTC-3' (reverse); and E-cadherin: 5'- CTGCTGCAGG TCTCCTCTTG-3' (forward) and 5'-TGTCGACCGGTGCAATCTTC-3' (reverse). All data were normalized using the internal control β-actin, and expression levels were analyzed using the 2-DDCt method.

Western blotting was performed as described previously. Briefly, cells were lysed in radioimmunoprecipitation assay (RIPA) buffer, and the lysates were separated using SDS-PAGE. The separated proteins were transferred to polyvinylidene fluoride (PVDF) membranes, and the blots were probed with primary antibodies (overnight, 4ºC). After incubating the blots with appropriate secondary antibodies conjugated to horseradish peroxidase, immunoreactive proteins were visualized using an enhanced chemiluminescence system and quantified densitometrically using the software LabWorks 4.5 (SensiAnsys, China). The following primary antibodies (1:1000) were used: anti-CHOP, anti-GRP78, anti-p27, anti-α-SMA, and anti-E-cadherin (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The internal control was β-actin.

We used flow cytometry to analyze the cell cycle as described previously [23]. Briefly, adherent cells was collected from control and experimental groups and fixed using ice-cold 70% ethanol for 30 min. The cells were washed with PBS and then resuspended in 1 mL of cell-cycle buffer (0.05 mg/mL propidium iodide in PBS containing 1 mg/mL RNase) at 37ºC for 30 min. After staining, the distributions of cell-cycle phases were analyzed using flow cytometry (BD FACSCalibur™, USA) and the percentages of cells were measured using Multicycle software.

Cells were treated with AOPPs and indicated reagents, after which the cells were trypsinized, washed with PBS, and counted using a hemocytometer. Equal numbers of cells were lysed in RIPA buffer containing (in PBS) 0.1% (w/v) SDS, 0.5% (w/v) sodium deoxycholate, and 1.0% (w/v) Nonidet P-40. The total protein content was measured using the bicinchoninic acid method.

All experiments were conducted in triplicate. Results are expressed as means ± SEM. Multiple groups were compared using one-way ANOVA. For comparing 2 groups, the LSD method was used, or when the assumption of equal variance did not hold, the Dunnett's T3 method was used. Differences were considered statistically significant at P < 0.05. Statistical analyses were conducted using SPSS 20.0 software.

Cell hypertrophy is characterized by an enlargement of cells that occurs because of an increase in protein and RNA content in the absence of DNA replication [24,25]. Normal renal tubular epithelial cells are quiescent in G0 phase, and when the cells prepare to proliferate, they enter the cell cycle at G1 phase and protein synthesis increases. G1 arrest is widely recognized to relate to antiproliferation or hypertrophy [26], and cell-cycle transit is known to be blocked by the cyclin-dependent kinase (CDK)-interacting protein/kinase-inhibitor protein(CIP/KIP) family of CDK inhibitors (CKIs:p21,p27, and p57). Tubular-cell hypertrophy is associated with the expression of p27 and p21 [27]. To investigate whether AOPPs induce HK-2-cell hypertrophy, we measured he total protein content in HK-2 cells. AOPP stimulation significantly increased the total protein content in HK-2 cells (Fig. 1). Additionally, we also measured the protein and mRNA expression of p27 and cell-cycle distribution of HK-2 cells. Treatment of the cells with AOPPs induced the overexpression of p27 at the protein (Fig. 2, A & B) and gene (Fig. 2, C & D) levels and increased the percentage of HK-2 cells arrested in G1 phase (Fig. 2 E & F) in a dose- and time-dependent manner. Although the maximal expression of p27, the total protein content, and the maximal number of cell arrested in G1 phase were not detected in the 24-h AOPP-treatment group, their levels in this group were significantly higher than that in the control group. Moreover, this effect was not observed in either the control group or the unmodified-BSA group. These results indicated that the overexpression of p27, the accumulation of total proteins, and the increased percentage of HK-2 cells arrested in G1 phase were all associated with the advanced oxidation of BSA. Taken together, these data suggested that AOPPs induced HK-2-cell hypertrophy through induction of the overexpression of p27 and G1 phase arrest.

During EMT, renal tubular cells lose their epithelial phenotype and acquire the characteristics of mesenchymal cells. The first phase of EMT is characterized by a loss of epithelial cell adhesion, including the loss of proteins associated with epithelial membrane junctions and tight junctions, such as E-cadherin. In the intermediate phase, the cells express a combination of epithelial and mesenchymal markers such as α-smooth muscle actin (α-SMA). E-cadherin and α-SMA are the most critical proteins involved in EMT [28]. Thus, to determine whether AOPPs induce tubular EMT, we measured the expression of E-cadherin and α-SMA. Treatment with AOPPs downregulated the mRNA and protein levels of E-cadherin but upregulated those of α-SMA in a concentration- and time-dependent manner (Fig. 3). The AOPP-induced increase in the mRNA levels of α-SMA peaked at 12 h and then decreased by 24 h, but the level was still significantly more than that in the control (P < 0.05). The expression of E-cadherin and α-SMA was not changed in control cells and in cells treated with unmodified BSA, which indicated that the loss of E-cadherin and the overexpression of α-SMA were associated with the advanced oxidation of BSA. Collectively, these results indicated that AOPPs induced EMT in HK-2 cells.

ER stress is a self-protection mechanism of cells that can restore homoeostasis within the ER. However, prolonged or severe ER stress can lead to apoptosis. GRP78 is involved in the unfolded protein response and the protection mechanism that functions during ER stress. Conversely, CHOP is associated with the injury mechanism induced by excessive ER stress. These represent 2 distinct mechanisms that are induced during ER stress and are regarded as markers of ER stress [29,30]. In this study, to examine whether AOPPs triggered ER stress, we examined the expression of GRP78 and CHOP by performing RT-PCR and western-blot assays. Compared with the levels in control cells and in cells treated with unmodified BSA, the protein and mRNA levels of GRP78 and CHOP were significantly increased in AOPP-treated HK-2 cells (Fig. 4). Although the maximal expression of CHOP and GRP78 at both gene and protein levels was not in the 24-h AOPP-treatment group, the expression levels were still at significantly higher than those in the control group. Collectively, these data indicated that AOPPs induced ER stress in HK-2 cells.

To investigate the role of ER stress in AOPP-induced hypertrophy and EMT in HK-2 cells, we measured the expression of CHOP, GRP78, p27, E-cadherin, and α-SMA in cells treated without or with AOPPs and an ER stress inhibitor (salubrinal), or in cells treated with an ER stress inducer (thapsigargin) alone. AOPP again induced the overexpression of CHOP, GRP78, p27, and α-SMA and suppressed the expression of E-cadherin at the transcriptional and translational levels, and these effects of AOPPs were partly reversed when the cells were treated with both AOPPs and the ER stress inhibitor. By contrast, the aforementioned effects of AOPPs were reproduced when cells were treated with only the ER stress inducer (Fig. 5). Furthermore, the AOPP-induced increases in the total protein content in HK-2 cells and the percentage of HK-2 cells arrested in G1 phase were also partly reversed following treatment with the ER stress inhibitor, whereas both of these increases were produced in cells after treatment with the ER stress inducer alone (Fig. 6). Collectively, these results indicated that AOPP-induced hypertrophy and EMT in HK-2 cells was mediated by the activation of ER stress.

Oxidative stress has been reported to activate ER stress [31,32,33]. Here, to determine the role of oxidative stress in ER stress in HK-2 cells, we measured the expression of CHOP, GRP78, p27, α-SMA, and E-cadherin in the presence or absence of AOPPs and either an NADPH oxidase inhibitor (diphenyleneiodonium, DPI) or superoxide dismutase (SOD), a cell-permeable scavenger of reactive oxygen species (ROS). Our results showed that AOPP-induced loss of E-cadherin and overexpression of α-SMA, CHOP, and GRP78 were partly suppressed after treatment with the NADPH oxidase inhibitor and the ROS scavenger (Fig. 5). Moreover, the increases in the total protein content in HK-2 cells and the percentage of HK-2 cells arrested in G1 phase after AOPP treatment were also partly reversed by DPI and SOD (Fig. 6). Taken together, these data suggested that oxidative stress played a role upstream or downstream of ER stress and participated in the process of AOPP-induced HK-2-cell hypertrophy and EMT.

In this study, we firstly demonstrated that AOPPs induced tubular hypertrophy and confirmed that AOPPs induced EMT in cultured human proximal renal tubular cells. Furthermore, our data also firstly revealed that AOPP-induced hypertrophy and EMT in human renal tubular cells were associated with ER stress, which was mediated by oxidative stress.

The first key finding of this study is that AOPPs induced hypertrophy in HK-2 cells. Cellular hypertrophy, an increase in cell size, is attributed to an increase in protein and RNA content in the absence of DNA replication. Previously, angiotensin II was shown to induce the overexpression of the CDK inhibitor p27 and cause cell-cycle arrest in G1 phase and lead to hypertrophy [27,34,35]. In accord with previous studies, we showed that AOPP treatment increased the total protein content in HK-2 cells, the expression of p27 at the gene level and protein levels, and the percentage of cells arrested in G1 phase. Based on these lines of evidence, we can conclude that AOPPs induce HK-2-cell hypertrophy via induction of overexpression of p27 and G1 phase arrest, which is the main cause of renal hypertrophy and which is associated with the pathogenesis of CKD.

Our results suggested that AOPPs induced not only hypertrophy but also EMT in HK-2 cells. Accumulating evidence indicates that renal tubular EMT is associated with tubulointerstitial fibrosis [2,3,36]. However, the contribution of tubular EMT to renal interstitial fibrosis is still controversial. In a previous study, it was shown that the number of myofibroblasts derived from EMT is only 5% [7]. In addition, Wilhelm et al. demonstrated that EMT might not be an in vivo process in renal fibrosis [6]. The discrepancy about the effect of EMT in kidney fibrosis may be attributed to the experimental conditions such as the mouse strain and type of disease model [37]. It is worth noting that, many studies revealed that tubular EMT plays an important role in the pathogenesis of CKD and is regarded as a potential therapeutic target for CKD [8]. The process of tubular EMT is characterized by the loss of an epithelial phenotype and the gain of profibrotic features that are characteristic of mesenchymal cells. E-cadherin and α-SMA are the most critical proteins known to be involved in EMT, during which the expression of E-cadherin is decreased and that of α-SMA is increased [4,28]. Our results showed that AOPP treatment induced tubular EMT, which was revealed by a loss of E-cadherin and the overexpression of α-SMA. By contrast, HK-2 cells treated with unmodified BSA did not exhibit marked changes in E-cadherin and α-SMA expression. These results showed that AOPPs, but not BSA, caused HK-2 cells to undergo EMT. Consistent with our finding, AOPPs were recently reported to induce EMT in human proximal renal tubular cells through oxidative stress [38]. Thus, we confirmed that AOPPs induce human proximal renal tubular cells to undergo EMT, which is related to the pathogenesis of CKD.

Various adverse conditions such as viral infection and glucose deprivation can result in the accumulation of unfolded proteins in the ER and then lead to ER stress [39]. Zhou et al. showed that AOPPs induced adipocyte perturbation through ER stress in vitro [22]. Considerable evidence suggests that ER stress is involved in a wide range of kidney diseases [19,40,41,42,43]. ER stress has been reported to play a crucial role in cardiac hypertrophy [44]. Moreover, the results of Wu et al. indicate that ER stress is closely correlated with renal tubular interstitial fibrosis in patients with diabetic nephropathy [45]. In this study, our results provided several lines of evidence suggesting that AOPP-induced hypertrophy and EMT in tubular cells was mediated by ER stress. First, exposure of HK-2 cells to AOPPs increased the expression of GRP78 and CHOP at both mRNA and protein levels, which indicated that AOPPs induced ER stress. Second, AOPP-induced hypertrophy and EMT of HK-2 cells were potently blocked by salubrinal, an inhibitor of eukaryotic initiation factor-2α dephosphorylation, which can inhibit ER stress. Lastly, thapsigargin, an inducer of ER stress, stimulated HK-2 cell hypertrophy and EMT. These results support the previous finding that ER stress might lead to EMT in HK-2 cells [46] and are consistent with the evidence suggesting that ER stress is involved in the pathological process in the progression of kidney diseases [20]. Collectively, these data indicated that AOPPs induced HK-2-cell hypertrophy and EMT by inducing ER stress.

ROS accumulation has been reported to activate ER stress [15,16,17]. However, although studies have suggested cross-talk between ER stress and oxidative stress, the mechanistic link between them is incompletely understood, and the potential effect of oxidative stress on AOPP-induced ER stress in HK-2 cells has not been previously investigated. Our results showed that an NADPH oxidase inhibitor and an ROS scavenger could partly prevent ER stress and also alleviate hypertrophy and EMT in HK-2 cells, suggesting that oxidative stress is involved in the pathway through which AOPPs induce ER stress. These results agree with the evidence indicating that oxidative stress can function upstream or downstream of ER stress [47]. Furthermore, oxidative stress can induce ER stress in neural and tumor cells [32,33], as well as in adipocytes [22]. Therefore, we conclude that oxidative stress and ER stress are associated.

Although we confirmed that AOPPs induced HK-2-cell hypertrophy and EMT through the induction of ER stress, which was mediated by ROS, whether ROS is upstream or downstream of ER stress has not been fully elucidated. Moreover, the signal transduction pathways that are activated by AOPP-induced ER stress in HK-2 cells remain to be fully elucidated. Three pathways are widely recognized to be involved in the unfolded protein response: the activating transcription factor 6 pathway, the inositol-requiring enzyme 1 pathway, and the double-stranded RNA-activated protein kinase-like ER kinase pathway [18]. Further investigation must be conducted to determine whether all of these pathways are involved in AOPP-induced hypertrophy and EMT in HK-2 cells and to identify the pathway that is most critical.

In conclusion, our results demonstrate that AOPPs induce hypertrophy and EMT in HK-2 cells. Importantly, AOPPs triggered hypertrophy and EMT in cultured HK-2 cells probably through the induction of ER stress. The induction of oxidative stress might also be involved, at least partially, in AOPP-induced tubular hypertrophy and EMT. AOPPs commonly accumulate in CKD patients and play a crucial role in the development and progression of CKD. Thus, our results provide a framework for developing novel nephroprotective therapies by targeting ER stress.

The authors declare no competing financial interests.

This work was supported by the Natural Science Foundation of Guangdong Province (No. 10151051501000030, No. S2011010004053, No. S2011040003566), the Science and Technical Plan of Guangzhou, Guangdong, China (No. 11C22120703), the Science and Technical Plan of Guangdong, China (No. 2013B021800149), and the National Natural Science Foundation of China (No. 81202842).

X. Tang, G. Rong contributed equally to this work.