AKF-PD alleviates diabetic nephropathy via blocking the RAGE/AGEs/NOX and PKC/NOX Pathways

Diabetic nephropathy (DN) is a major complication of diabetes. Currently, drugs are not available to effectively control the disease. Fluorofenidone (AKF-PD) is a recently developed drug; it possesses activities in reducing DN progression in preclinical research. Nonetheless, its renal protection and the underlying mechanisms have not been thoroughly investigated. We report here that AKF-PD significantly alleviatesrenal oxidative stress (OS) in db/dbmice through downregulation of Nicotinamide Adenine Dinucleotide Phosphate (NADPH) oxidase and upregulation of glutathione peroxidase and superoxide dismutase, thereby protecting kidney from DN pathogenesis. AKF-PD likely reduces OS through the advanced glycation end products (AGE) and protein kinase C (PKC) pathways. While renal AGEs, PKCα, PKCβ, and NADPH oxidase 4 (NOX4) were all substantially upregulated in db/db mice compared to db/m animals, AKF-PD robustly downregulated all these events to the basal levelsdetected in db/m mice. In primary human renal mesangial cells (HMCs), high glucose (HG) elevated receptor for advanced glycation endproducts (RAGE), PKCα, PKCβ and NOX4 activity, and induced the production of reactive oxygen species (ROS); these events were all inhibited by AKF-PD. Furthermore, HG led to mitochondrial damagein HMCs;AKF-PD conferred protection on the damage. Knockdown of either PKCα or PKCβ reduced HG-induced ROS production and mitochondrial damage in HMCs. The knockdown significantly enhanced AKF-PD-mediated inhibition of ROS production and mitochondrial damage in HG-treated HMCs. Collectively, our study demonstrates that AKF-PD protects renal function under diabetes conditions in part through inhibition of OS during DN pathogenesis. AKF-PD can be explored for clinical applications in DN therapy.

Scientific RepoRts | (2019) 9:4407 | https://doi.org/10.1038/s41598-018-36344-w elevation of SCR in the long-term users of ACEI/ARB. The elevation resulted in patients at risk of developing renal dysfunction 12 . Plenty preclinical experiments have been conducted to examine the anti-fibrosis pharmacology of numerous drugs, including statins, rhubard and tripterygium, poricoic acid, ergone and so on [4][5][6][7][13][14][15][16][17] . These drugs display activities in regulating cytokines secretion and reducing inflammatory reactions; their safety and clinical efficacy are under investigation. It is emerging from a series of recent studies that OS is a critical cause of DN 8,9 . In mouse models for DN and patients, ROS and the metabolic product of peroxidation such as malondialdehyde (MDA) and 8-iso-PGF2a were elevated concurrently with decreases in antioxidase activity including SOD and GSH-Px, supporting the importanceof ROS during the course of DN. At early stages, ROS impairs the vascular permeability and glomerular hemodynamics, destroys the glomerular electrostatic and filtration barrier, activates nuclear factor kappa beta (NF-κB) and activator protein 1(AP-1), and regulates the secretion of multiple inflammatory mediators 10 . At advanced stages of DN, ROS can promote the transdifferentiation of renal tubular cells, induce apoptosis in renal podocytes and mesangial cells, alter the balance of ECM production and degradation, and accelerate the formation of glomerular sclerosis 11 . ROS thus promotes DN initiation and progression by affecting multiple DN processes. These properties indicate targeting ROS being attractive in DN therapy. However, this potential has yet to be realized owing to the lack of medications.
The basic pathological changes of DN is renal fibrosis caused by the accumulation of extracellular matrix (ECM), mainly including collagen I and collagen IV, fibronectin and vementin 18,19 . The reversal and prevention of renal fibrosis is the key to DN therapy. accumulated studies have confirmed that renal fibrosis is associated with oxidative stress (OS) and inflammation, which is related to tissue injury resulted in imbalance between the production and elimination of reactive oxygen species (ROS) [20][21][22][23][24][25] . Therefore, how to reduce ROS formation is of clinical significance to preclude DN of diabetic patients. Our research group has recently invented a very promising novel drug AKF-PD, also known as Fluorofenidone; AKF-PD delivers satisfactory results in the treatment of renal interstitial fibrosis in preclinical research 26,27 . AKF-PD dramatically delayed the progression of DN in db/db mouse, indicating its attractive potential in management of patients with renal fibrosis. But less is known about the therapeutic mechanism of AKF-PD in DN. In view of the important role of OS in the pathogenesis of DN, our previous research suggested that AKF-PD likely inhibited the progression of DN through suppression of the expression of NADPH oxidase 27,28 . It is known that the advanced glycation end products (AGEs) and the protein kinase C (PKC) signal pathway play a central role in the expression of NADPH oxidase/OS 29 , which might be the targets of AKF-PD in DN treatment. In this study, we aimed to detect the therapeutic effects of AKF-PD in DN and to explore the related molecular mechanism both in vitro and in vivo.

AKF-PD reduces mesangial expansion of DN.
In comparison to db/m mice, renal mesangial matrix expansion was clearly observed in db/db mice (Fig. 1B). The addition of AKF-PD or losartan at week 5 (5w) and 8w significantly alleviated this expansion (Fig. 1B), which was evidenced by lowering renal glomerular expansion index (Fig. 1C). Therapeutic effect of the intervention initiated at week 8 seems better than weeks 5, evident by the non-significant glomerular expansion index observed in the treated db/db mice compared to db/m mice (Fig. 1C).
AKF-pD reduces Albumin/creatine ratio. To demonstrate the above renal protection delivered by AKF-PD being functional, we examined whether renal function was also preserved. As expected, the renal function of db/db mice was significantly compromised compared with db/m mice ( Table 1). Administration of AKF-PD lowered 24-hour urine albumin/creatinine ratio (ACR) ( Table 1), regardless whether intervention was initiated at weeks 5 or 8-. This provides the first demonstration that kidney function can be partially preserved by AKF-PD at least in db/db mice. On the other hand, the administration of losartan aggravated renal injury with a significant increase in serum creatinine SCR (Table 1). Collectively, evidence suggests that AKF-PD is better than losartan indeceasing urine protein at 8 weeks (p < 0.05), at the same time we confirmed that AKF-PD did not influence uric acid and serum serum creatinine in AKF-PD treated mice compared to db/db mice.
Protective effects of AKF-PD on renal function are associated with the inhibition of OS. OS is a major cause of DN under diabetes conditions 30 . Significant elevations of the renal NAPDH oxidase activity was demonstrated in mock-treateddb/db mice at weeks 5 or 8 compared to db/m mice ( Fig. 2A). Administration of either AKF-PD or losartan to both 5 weeks and 8 weeks db/db mice significantly reduced renal NAPDH oxidase activity to the basal level detected in db/mmice ( Fig. 2A). ROS is regulated by cell's activities in ROS production and clearance, suggesting that AKF-PD could also affect ROS elimination. Consistent with this possibility, the activity of renal glutathione peroxidase (GSH-Px) was dramatically increased in db/db mice when animals were received either AKF-PD or losartanat at weeks 5 and 8 (Fig. 2B). Similar observations were also obtained for superoxide dismutase (SOD) activity (Fig. 2C). Collectively, the above observations suggest the notion that AKF-PD offer a better antioxidation effects during diabetes nephropathy progression.
Oxidative stress is plays a pivotal role in the progression of diabetes renal injury. OS is functionally related to advanced glycation end products (AGEs), their receptor RAGE, fibronetin, PKCα, and PKCβ 17 . To further investigate the impact of AKF-PD on renal OS, we hypothesized that AKF-PD driven blockage of oxidative stress in db/db mice was likely mediated by antagonizing AGEs/RAGE and PKC signaling. We detected a dramatic upreglaton of AGEs, RAGE, fibronectin, PKCa, and PKCβ in the kidneys of db/db mice compared to db/m mice; all these increases were dramatically reduced to the basal levels observed in db/m when AKF-PD was given to either 5 weeks or 8 weeks db/db mice (Figs 3 and 4). These results are even more significant when they are interpreted in the context of experimental duration of 24 weeks, i.e. starting of the AKF-PD intervention at 8 weeks at time when renal damage was clear (Fig. 1) and kidney functions were substantially compromised (Table 1)  To determine the mechanisms underlying AKF-PD-derived inhibition of the AGE-RAGE-NOX and PKC-NOX pathways, we demonstrated that AKF-PD reduced RAGE, NOX4, PKCα, and PKCβ mRNA expression in HMCs treated with HG ( Fig. 6), suggesting that AKF-PD downregulates the AGE-RAGE-NOX and PKC-NOX pathways in part via inhibition of transcription.
AKF-pD confers protection on mitochondrial damage. In line with the above observations, HG resulted in an increase in cellular ROS in HMCs and AKF-PD reduced ROS accumulation (Fig. 7). Mitochondria is the central source of ROS, implying a role of AKF-PD in protecting mitochondrial integrity. We observed that in HG-induced HMCs caused an elevation of ROS along with an increase in mitochondrial potential, indicative of mitochondrial damage (Fig. 8). AKF-PD reduced ROS production and mitochondiral potential in HMCs treated with HG (Fig. 8); this suggests that AKF-PD attenuates ROS production in response to HG environment in part via maintaining mitochondrial integrity. On the other hand, while losartan was able to reduce ROS in HG-treated HMCs (Fig. 7), it had no effects on mitochondrial damage (Fig. 8), supporting AKF-PD being superior to losartanin protecting renal OS.
AKF-PD preserves mitochondrial integrity via complex mechanisms. Our observed concurrent occurrence between PKC (alpha and beta) upregulations and NOX4 increases in vivo (Figs 3 and 4) and in vitro    Interestingly, the combination of AKF-PD with knockdown of either PKCα or PKCβ further reduced ROS in HG-stimulated HMCs to the basal level (Fig. 9B). We subsequently determined the contributions of PKCα and PKCβ to AKF-PD-derived protection of mitochondrial damage. The similar relationship among AKF-PD and knockdown of either PKCα of PKCβ was observed with respect to mitochondrial potential (Fig. 9E) and ROS levels (Fig. 9B) in HMCs in the presence of HG. Nonetheless, the combination of AKF-PD with knockdown of either PKCα or PKCβ partially prevented increases of mitochondrial potential (Fig. 9E). Collectively, the above results suggest that AKF-PD inhibits ROS production and protects mitochondria from damage in HG-treated HMCs through PKCa-dependent and PKCβindependent mechanisms.

Discussion
Our research provides the first thorough evidence for AKF-PD being such a drug. AKF-PD is a pyridyl ketone compound with a broad spectrum of antifibrosis activity 3,31 . Previous research has revealed its utility in inhibition of mouse renal fibrosis caused by diabetes and unilateral ureter obstruction 3,13 . In this study, we observed that AKF-PD coordinately reduces renal OS in db/db mice through preventing upregulations of the NADPH oxidase and downregulations of the anti-oxidative enzyme system like GSH-Px and SOD. Considering ROS promoting DN development through four recognized pathways: AGEs, PKCs, Polyalcohol, Hexosamine pathways 30,32 , we demonstrated that AKF-PD inactivates the pathways of AGE and PKC, supporting AKF-PD possessing potent activities in inhibiting renal OS. However, we cannot exclude the possibility that AKF-PD also antagonizes the other two pathways; this possibility deserves investigations in the future.
In patients with diabetes, high glucose enhances the generation of ROS in mitochondria, induces OS within histiocyte, and ultimately promotes the development of various complications. ROS is largely originated from mitochondria as O − , which can inhibit the activity of the enzyme complex II, II, and III in the electron transport chain and then lead to energy synthesis dysfunction. In this process, the mitochondria inner membrane is vulnerable, probably due to its special biological and anatomic characteristic 33 . Our previous studies have verified that AKF-PD could depress the production of ROS and suppress the expression of lipid peroxidation product (MDA and 8-iso-PGF2a), which raises the hypothesis that AKF-PD attenuates DN through alleviating the renal mitochondrial oxidative injury 27 . This hypothesis is supported by our results that AKF-PD almost normalized JC-1 aggregation in the membrane of mitochondrialin HMCs while Losartan partly reversedmitochondrial damage induced by HG. The mechanisms underlying AKF-PD-derived protection and therapeutic value towards DN need further investigation. Nonetheless, our research suggests these mechanisms are PKCα and PKCβ-dependent and PKCα, PKCβ-independent. AKF-PD was able to HG-caused mitochondrial damage and ROS levels in HMCs with knockdown of either PKCα or PKCβ. These observations supports the contributions of pathways independent of either PKCα or PKCβ to AKF-PD activities in maintaining mitochondria integrity; alternatively PKCα or PKCβ is able to compensate the knockdown of another.
In conclusion, we demonstrate that AKF-PD displays an impressive preservation of renal function in 24 weeks old db/db mice when intervention was started from the age of 5 weeks or 8 weeks. In view ofthe further reduction of ROS levels and protection of mitochondrial potential alterations when AKF-PD was combined with knockdown of PKCα or PKCβ, the clinical potential of this combination could be explored in the future.

Materials and Methods
Mouse model. Four-week old male db/db mice (n = 36) were fed and randomly divided into six groups. Mice in the AKF-PD groups were treated with 500 mg/kg/day AKF-PD at the age of 5 weeks (AKF-PD-5w group) or 8 weeks (AKF-PD-8w group); the dosage was chosen according to on our previously defined conditions 34,35 . Mice in the positive control groupswere treated with 20 mg/kg/d losartan at the age of 5 weeks (LOS-5w group) or 8 weeks (LOS-8w group). Mice in the negative control groups were treated with placebo at the age of 5 weeks (Ctrl-5w group) or 8 weeks (Ctrl-8w group). Mice with the db/m genotype (db/m group, n = 6) served as untreated blank control. All treatments were performed daily by oral gavage. After the treatment ended, mice were sacrificed at 24 weeks old (Fig. 1A), and kidney tissues were collected for pathological, protein and mRNA detection. The protocol was approved by the Committee on the Ethics of Animal Experimentation and Care of the Central South University Xiangya Hospital. pathology and immunohistochemistry. Histological Evaluation were performed as previously described study 36,37 . Hematoxylin-eosin (HE) and Periodic Acid-Schiff (PAS) staining were carried out for pathological analysis. Mesangial expansion was determined by glomerular matrix expansion index (GMI) which was calculated as our previous studies 3 . The expression of FN in kidney tissue was detected by western blot analysis.
Cell culture. Human renal mesangial cells (HMC) were provided by Changsha Central Hospital laboratory.
HMCs cells were cultured in DMEM supplemented with 8% FBS, penicillin (100 U/ml) and 100 μg/ml streptomycin (Invitrogen), at 37 °C in a humidified atmosphere of 5% CO2 and 95% air. The cells were seeded on six-well culture plates to 60-70% confluence in complete medium containing 5% FBS for 24 h. Cells were divided into 5 groups to receive different drug interventions respectively: N group treated with 5.6 mM glucose, OC (mannitol)  Western blot analysis. Total protein extracts from kidney tissue were prepared using the Protein Kit (Qiagen, Valencia, CA, USA) according to manufacturer's instructions and previous publications 39,40 . Tissues were processed in liquid nitrogen and lysed with SDS-PAGE sample buffer. To detect the protein expression of AGEs, RAGE, p-PKCα, p-PKCβ, NADPH-NOX4, Fibronetin, 30 μg protein lysates were loaded per well, separated on 8% or 12% SDS-polyacrylamide gels and transferred to PVDF membranes (Immobilon P, Millipore, Bedford, MA). The membranes were blocked with 5% fat free milk in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 30 minutes at room temperature, and then probed with antibodies, followed by incubation with horseradish peroxidase-conjugated secondary antibodies (1:6000) at room temperature for 1 hour. The immune complexes were visualized by chemiluminescence using an ECL kit (Pierce, USA). Positive bands were analyzed with Quantity One software (Bio-Rad). GAPDH was used as the internal control, and all the experiments were repeated for three times.
os measurement. Kidney tissue suspension was extracted and cytochrome C reduction method was used to measure the NAPDH oxidase activity. SOD and GSH-Px activity was detected by NBT method and DNTB method, respectively. The metabolic product peroxidation were examined by western blot analysis. The measurement of intracellular ROS level changes we reperformed as previously described 13 . Each experiment was performed in triplicate.  PKCβ-siRNA HMCs were treated as indicated, followed by measuring ROS with Dichlorodihydrofluorescein diacetate assay, NOX4 and Fibronectin expressions were determined. (E) Ctrl-siRNA, PKCα-siRNA, or PKCβ-siRNA HMCs were treated as indicated, followed by measuring ROS with Dichlorodihydrofluorescein diacetate assay. Typical images were acquired using a fluorescence microscope. Experiments were repeated three times, mean fluorescence intensities ± SD are graphed. (C) Ctrl-siRNA, PKCα-siRNA, or PKCβ-siRNA HMCs were treated as indicated. Mitochondrial potential was determined. Experiments were repeated three times; mitochondrial potentials are graphed.
Scientific RepoRts | (2019) 9:4407 | https://doi.org/10.1038/s41598-018-36344-w statistical analysis. All the data analysis was conducted using SPSS 17.0 software (SPSS Inc, Chicago, IL, the USA). Continuous and categorical data was presented as umber (percent) or mean ± SD (standard derivation). Normal distribution test and homogeneity test of variances should be conducted for the data of different groups at first. If data meet the normal distribution (P > 0.20) and homogeneity of variance (P > 0.10), one-way analysis of variance (ANOVA) assay was used to compare difference groups. If not, the rank sum test should be conducted. The P-values of less than 0.05 were considered to be statistically significant.