Early Elimination of Uremic Toxin Ameliorates AKI to CKD Transition

Acute kidney injury (AKI)-related fibrosis is emerging as a major driver of chronic kidney disease (CKD) development. Aberrant kidney recovery after AKI is multifactorial and still poorly understood. The accumulation of indoxyl sulfate (IS), a protein-bound uremic toxin, has been identified as a detrimental factor of renal fibrosis. However, the mechanisms underlying IS-related aberrant kidney recovery after AKI is still unknown. The present study aims to elucidate the effects of IS on tubular damage and its involvement in the pathogenesis of AKI to CKD transition. Our results showed that serum IS started to accumulate associated with the downregulation of tubular organic anion transporter, but not observed in the small-molecule uremic toxins of the unilateral ischemia-reperfusion injury (UIRI) without a contralateral nephrectomy model. Serum IS is positively correlated with renal fibrosis and Binding immunoglobulin protein (BiP) and CAAT/enhancer-binding protein (C/EBP) homologous protein (CHOP) expression induction in the UIRI with a contralateral nephrectomy model (UIRI+Nx). To evaluate the effects of IS in the AKI to CKD transition, we administered indole, a precursor of IS, at the early stage of UIRI. Our results demonstrated IS potentiates renal fibrosis, senescence-associated secretory phenotype (SASP), and activation of endoplasmic reticulum (ER) stress, which is attenuated by synergistic AST-120 administration. Furthermore, we clearly demonstrated that IS exposure potentiated hypoxia-reperfusion (H/R) induced G2/M cell cycle arrest, epithelial-mesenchymal transition (EMT), and aggravated ER stress induction in vitro . Finally, the ER chemical chaperon, 4-phenylbutyric acid (4-PBA), successfully reversed the above-mentioned AKI to CKD transition. Taken together, early IS elimination in the early stage of AKI is likely to be a useful strategy in the prevention and/or treatment of the AKI to CKD transition. nephrectomy AST-120


INTRODUCTION
Acute kidney injury (AKI) is characterized by a rapid decline of renal function [1], which associates with the increasing risks of subsequential comorbidity, including acute myocardial infarction, heart failure, and sepsis. Recently, it has been identified that AKI is a long-term risk factor for chronic kidney disease (CKD), end-stage renal disease (ESRD), and death [2]. Some patients can fully recover from AKI, but others cannot because maladaptive repair after AKI leads to chronic kidney fibrosis [3]. The rising burden of CKD after AKI has pushed the study to focus on the early treatment of AKI and enhance adequate renal function recovery. However, even the normal serum creatinine (Scr) level after AKI does not always truly reflect the complete recovery from AKI [4].
With renal function deterioration, the accumulation of various endogenous metabolites is generally named uremic toxins [5]. Indoxyl sulfate (IS), a protein-bound uremic toxin, is the product of indole sulfation in the liver via the portal circulation after uptaken metabolite of tryptophan by the gut bacteria [6].
Because of its high protein bond characteristics, IS is secreted through the organic anion transporter (OTA) of renal tubular epithelial cells (RTECs) and systemically accumulates in CKD patients [7]. Recently, it has caught more attention for its systemically detrimental effects in CKD patients [6,8,9], but not in the AKI population.
It is well documented that the unfolded protein responses (UPRs) signaling is activated by the derangement of endoplasmic reticulum (ER) proteostasis and participates in various kidney injuries [10]. Our previous study demonstrated the ER stress implicated in the development of renal fibrosis [11]. Furthermore, abundant pieces of evidence reported a high association between ER dysfunction and premature Downloaded from http://portlandpress.com/clinsci/article-pdf/doi/10.1042/CS20210858/924836/cs-2021-0858.pdf by guest on 29 November 2021 aging [10], and CKD is increasingly being accepted as a type of renal aging. During AKI to CKD transition, senescence-associated secretory phenotype (SASP), which is associated with the secretion of interleukin-6 (IL-6), tumor necrotic factor-α (TNF-α), and plasminogen activator inhibitor 1 (PAI-1), actively contributed to the progression of CKD [12,13]. This study hypothesized that the early uremic toxin removal in AKI would modify the harmful effects through modulation of UPRs and SASP and retarded AKI to CKD transition. To clarify this question, we use the unilateral ischemia-reperfusion with/without contralateral nephrectomy model to explore the early removal of IS by AST-120, an oral spherical carbonaceous adsorbent, on AKI to CKD progression in vivo. Furthermore, we also investigated the molecular mechanisms of IS exposure to H/R conditions in vitro. Our results support that the early removal of IS attenuates renal fibrosis after ischemic AKI.

Animal models and protocol
C57BL/6 mice aged from 8 to 10 weeks were used in this study. All procedures in this study were approved by the Institutional Animal Care and Use Committee of the National Taiwan University College of Medicine (IACUC NO. 20170164) and all the surgeries were conducted at the operation room in National Taiwan University College of Medicine Laboratory Animal Center. The AKI to CKD transition mouse model: Unilateral ischemic-reperfusion injury without contralateral nephrectomy (UIRI) and two-stage unilateral ischemic-reperfusion injury model with contralateral nephrectomy (UIRI+Nx) at day10 were introduced in this experiment [14,15]. For both UIRI models, mice were anesthetized with intraperitoneal injection of Zoletil (50 The right kidney was kept intact. The wound was sutured after the release of the clip, and animals were allowed to recover. At the beginning of stage II (day 10), the right kidneys were removed, and animals were sacrificed to observe CKD transition at indicated time points or the end of observation on day 15. The next day after nephrectomy was expressed as NxD1, and the observative endpoint was named as UIRI+Nx. All mice in each group are divided into indole (IS precursor, 10µg/g/day via gavage) or AST-120 (4mg/g/day via oral feeding) treatment at the 5th day of two-stage UIRI, and were allocated into 5 groups, including (1) Sham, (2) UIRI, (3) UIRI + indole, (4) UIRI + indole+ AST-120, (5) UIRI + AST-120 groups. At the end of the experimental procedure, all the animals were anesthetized and euthanized by cardiac puncture then cervical dislocation. The animal serum and kidney samples were collected for further analysis.

Reagents and antibodies
IS and sodium phenylbutyrate were purchased from Sigma-Aldrich. Indole and anaerobic bags were purchased from Merck KGaA (Darmstadt, Germany).

RNA extraction, reverse transcription, RT-PCR, and quantitative real-time PCR
Total RNA of cells or freshly isolated kidneys were extracted by GENEzol™ TriRNA Pure Kit (Geneaid, New Taipei City, Taiwan). One microgram of RNA was reverse transcribed with iScript reverse transcription supermix (Bio-Rad, Hercules, CA). The resulting cDNA products were amplified with specific primer pairs to detect mRNA abundance using a StepOnePlus real-time PCR system (ThermoFisher Scientific). The relative expression of the target genes was calculated using the comparative threshold cycle (CT) method (ΔΔCT). The sequences of the primer pairs are listed in Table 1.

Histological analysis
Kidneys were removed from euthanized mice and fixed in 10% formalin with phosphate-buffered saline (PBS). Hematoxylin and eosin-stained sections were used to estimate renal histological injury. Masson's trichrome-stained sections were used to evaluate renal fibrosis which is measured by imageJ blue-stained area to cross the whole image represented as fibrosis area fraction for the degree of interstitial collagen deposition. Twenty cortical tubulointerstitial fields that were randomly selected at 400X magnification were assessed in each mouse, and the average for each group was analyzed.

MTS assay
CellTiter 96® Aqueous One Solution Cell Proliferation kit (Promega, Wisconsin, USA) was used in MTS assay. normalized to controls and represented as the proliferation rate of the controls.

Statistical Analysis
All data examined are expressed as means ± SEMs. Statistical analyses of the data were performed using Prism 6 software. Comparison between groups was made using one-way ANOVA followed by Newman-Kuels test. P ＜ 0.05 was considered statistically significant.

Accumulation of serum IS in UIRI without contralateral nephrectomy model
As shown in Figure 1A, the UIRI protocol demonstrated the time-point of observation through the study period. The tissue and blood samples were collected for histopathology and biochemistry analysis at indicated days post-UIRI. As shown in Figure 1B, the H&E staining showed an apparent tubular dilation, the hallmark feature of AKI, at day 11 post-UIRI (UD11). Small molecular uremic toxins including, blood urea nitrogen (BUN), and Scr were not raised, but IS was significantly accumulated in UIRI mice compared with that of sham-operated mice ( Figure 1C and 1D). Because IS was secreted by organic anion transporter, we further demonstrated renal OAT-1 expression significantly decreased in UIRI kidney tissue ( Figure 1E). In addition, BUN and Scr levels were significantly accumulated at NxD1 mice ( Figure   1F and 1G). These findings suggested that serum IS accumulated since the early stage of UIRI while impairing organic acid secretion of the injured kidney, but BUN and Scr accumulated only after impaired glomerular filtration developed while the non-ischemic kidney is removed.

Serum IS accumulation was associated with post-AKI fibrosis
To monitor the fibrosis progression after UIRI, we stained kidney tissue sections with Masson's Trichrome (MT), which represented the extracellular matrix deposits [16]. As shown in Figures 2A and 2B, in the UIRI group, the MT staining area gradually increased, and more than 20% collagen deposition was formed from UD11 to UD15.
Concomitantly, the protein expression of α-SMA and vimentin were also substantially increased in UIRI mice kidneys ( Figure 2C-E), and similar was also found in Col11 mRNA expression ( Figure 2F). We hypothesize that IS plays as a trigger for fibrosis development, then further evaluate the relationship between serum IS level and fibrotic/epithelial-mesenchymal transition (EMT) parameters. As shown in Figure 2G and 2H, serum IS level was significantly positive correlated with α-SMA (Pearson's r = 0.7436, P < 0.001), and also positively correlated with fibrosis fraction (Pearson's r = 0.6561, P < 0.05). These results suggest that IS may be a possible detrimental factor in the transition from AKI to CKD in the UIRI kidney.

AST-120 administration attenuated renal fibrosis in two-stage UIRI with contralateral nephrectomy
In order to demonstrate the early removal of protein-bound uremic toxin attenuated the renal fibrosis in the two-stage UIRI mice, we randomly assigned into indole (10 µg/g/day via gavage) or AST-120 (4 mg/g/day via oral feeding) treatment 5 days after UIRI+Nx+indole+AST-120; and (5) UIRI+Nx+AST-120 groups as shown in Figure   3A. First, the UIRI+Nx group showed significantly serum IS accumulation, further increasing after indole administration ( Figure 3B). Furthermore, tubulointerstitial damage and fibrosis were also demonstrated by H&E and Masson's trichrome staining (MT) in Figure 3C. These results suggested that high-protein intake through indole supplement might enhance kidney injury during AKI to CKD transition. As shown in Figures 3B and 3C, AST-120 treatment effectively reduced the serum IS accumulation, which concomitantly abolished the tubulointerstitial damage and fibrosis. Further study revealed that transforming growth factor-beta 1(TGF-) and connective tissue growth factor (CTGF), which are the critical mediators in promoting renal fibrosis, and extracellular matrix gene col1a1 mRNA expression in the indole gavage group were significantly upregulated ( Figure 3D-F) and decreased in IS removal group (indole+AST-120 and AST-120 treatment group). Moreover, EMT plays a critical role in AKI to CKD transition, and we further analyzed the effects of AST-120 on CTGF, vimentin, and α-SMA protein expression. As shown in Figure   3G-3J, AST-120 significantly attenuated CTGF, vimentin, and α-SMA expression in UIRI+Nx+indole+AST-120 and UIRI+Nx+AST-120 groups, and the downstream target of TGF-: p-Smad2/3 protein expression had similar trends in IS accumulation and elimination group ( Figure 3G and 3K). These results suggested the therapeutic potential of early IS removal by AST-120 in the AKI to CKD transition.

AST-120 treatment modulated UPRs and SASP in AKI to CKD transition
Our previous work and others study demonstrated a significant association between kidney disease progression and proteostasis imbalance [10]. Therefore, we further observed UPR signaling of the study groups. As shown in Figure 4A cyclin D1 protein ratio, which implicated cell cycle arrest [19], was also significantly upregulated in the H/R condition treated with the IS group ( Figure 5C). These results supported that IS potentiated maladaptive repairing processes through prolonged cell cycle arrest. Current evidence suggested that ROS attacked DNA, which forming γ-H2AX, is an early cellular response to the induction of DNA double-strand breaks, resulting in cell cycle arrest [20]. As shown in Figures 5D and 5E, pretreatment of IS aggravated ROS generation, accompanied by γ-H2Ax protein induction in H/R injury.
In maladaptive RTECs, cell cycle G2/M phase arrest would enhance EMT and profibrotic factor upregulation [21]. In Figure 5 F-I, the H/R condition pre-treated with IS had higher profibrotic factor CTGF, -SMA protein expression, and lower E-cadherin expression level in HK2 cells. In addition, UPRs were also involved in the pretreatment of IS in the H/R condition, as shown that BiP, p-eIF2, and CHOP protein expression were significantly upregulated ( Figure 5J-M). These findings showed that IS potentiated EMT and UPRs in H/R tubular cells, and we further designed the study to clarify the effects of UPRs modulation on the profibrotic phenotype.

4-PBA, an ER chemical chaperon, reversed IS-potentiated cell cycle arrest and EMT in vitro
In order to clarify the roles of IS on the ER proteostasis on the two-stage UIRI kidney

DISCUSSION
AKI is encountered in one-third of in-hospital patients [22]. Current diagnostic criteria of AKI are generally dependent on the kidneys' filtration function, such as urine output and Scr, but do not include the secretory function of the kidney [23].
Current evidence suggests that IS, a representative protein-bound uremic toxin, mainly excretes by tubular secretion and acts as a pathologic factor that triggers the ROS generation, organelle stress, apoptosis, and inflammation leading to cardiovascular and kidney dysfunction [6,24,25] and OAT3, which is responsible for increased serum IS level of the injured kidney [21]. In order to demonstrate the different accumulation of uremic toxins between small molecular and protein-bound uremic toxins in AKI, we conduct UIRI without nephrectomy, which maintains adequate glomerular filtration function in the normal kidney and impaired excretory function in the injured kidney. Figures 1C and 1D show that BUN and Scr levels remain within normal limits, but the IS level accumulates as the OAT1 expression is downregulated ( Figure 1E, 1F, and 1G).
These results support that persistent IS accumulation in AKI even after glomerular filtration function being restored. To elucidate the potentially harmful effects of IS in AKI, we evaluate the correlation between serum IS level and tissue fibrosis. Our results show that a higher serum IS level is positively associated with higher -SMA expression and fibrosis staining in the injured kidney ( Figure 2G and 2H). These results further support the adverse effects of IS in renal fibrosis during AKI to CKD transition.
Second, previous studies have shown that higher dietary protein food will also be underestimated by using Scr as a biomarker in AKI to CKD transition [26,27] . IS is a notorious uremic solute, generally accumulated in CKD patients with deteriorated renal function, impaired neovascularization, and aggravated uremic sarcopenia [8,28].
A previous study has found that the metabolic profiles change and IS accumulates in the early stage of AKI [29]. The present study further demonstrates that IS is significantly accumulated in the UIRI+Nx group and further boosted by oral indole gavage in the UIRI+Nx+Indole group, as shown in Figure 3B. Recently, an increasing number of articles explored the functional roles of microbiota [30]; IS-producing microflora has aroused more attention in CKD patients [31,32]. Thus, a low-protein diet has been recommended to non-dialysis CKD patients to reduce the kidney burden from tryptophan-derived uremic toxins [33]. Applying AST-120 successfully attenuated the fibrotic area ( Figure 3C) and expression of profibrotic factors, TGF1 signaling, andEMT in AKI to CKD tissue ( Figure 3D-3K). However, there is a point to mention that AST-120 administration would absorb IS precursor and other PBUT precursors such as p-cresol in the intestine. To emphasize the specific deteriorated effect of IS in AKI to CKD, we designed the analysis into the UIRI+Nx+Indole group and UIRI+Nx+Indole+AST-120 group to clear up this concern. Overall, our findings further support the early elimination of IS attenuates the AKI to CKD insult.
Third, our previous studies and others clearly demonstrate that activation of ER stress is critical for tubular cell loss and renal fibrosis in the AKI to CKD models. This finding connects a possible causative role for ER stress with the development of progressive kidney fibrosis [10]. In this study, we also evaluate the key indicators of ER stress, including BiP and CHOP, in the UIRI+Nx model. As shown in Figure   4A-C, BiP and CHOP protein have been upregulated in UIRI+Nx and UIRI+Nx+Indole group, which is attenuated by AST-120 administration in trend. In addition, serum IS level is positively correlated with the expression of BiP and CHOP.
A similar observation has shown that IS-treated HK2 cells potentiate the H/R-induced BiP, CHOP, and eIF2 phosphorylation ( Figure 5J-5M). Previous studies also confirmed that the chemical chaperon 4-PBA protects against ER stress-mediated renal fibrosis [34,35]. This study, we successfully demonstrate 4-PBA to abolish the IS-induced ER stress in the H/R injury model in vitro (Figure 6A and 6B). According to the above findings, removal of IS in the early stage of AKI might be beneficial for AKI to CKD injury through blunting ER stress-mediated kidney injury both in vivo and in vitro. In addition, age-related organ dysfunction and tissue fibrosis have suggested that chronically senescent cells with a shared SASP are involved in tissue inflammation and fibrosis [22]. A great number of articles indicate AKI as a condition causing premature kidney senescence [36][37][38]. Adijiang A et al. demonstrates that IS reduces Klotho expression and promotes senescence in the kidney through ROS production, nuclear factor kappa B activation, and enhanced the SASP in proximal tubular cells [39]. In this study, early IS removal by AST-120 successfully suppresses the IL-6 and PAI-1 expression as shown in the UIRI+Nx+indole+AST-120 group ( Figure 4F and 4G). Therefore, removing IS, a well-known proinflammatory activator, would take into concern in the early stage of AKI patients to prevent the development of kidney senescence.
Finally, in response to renal insults surviving renal tubular cells can activate an intrinsic repair process by reentering mitosis and restoring kidney architecture and function. However, the renal repair is limited; even if the renal function returns to baseline after an acute insult, residual inflammatory and fibrotic processes in the kidney can contribute to CKD development [3]. A recent study has linked epithelial cell cycle arrest in G2/M to kidney fibrosis after injury. It is highly associated with proinflammatory and profibrotic cytokines secretion [19]. Our study demonstrates that IS and H/R induce HK-2 cells in G2/M cell cycle arrest and increase cyclin B1/D1 ratio, and IS further potentiates the effects ( Figure 5A, 5B, and 5C). The phenomenon is probably associated with ROS generation and following DNA damages by IS exposure in the H/R injured RTECS ( Figure 5D and 5E). Furthermore, 4-PBA also attenuates the IS-induced cell cycle arrest in the H/R injury model in vitro ( Figure 6E and 6F).