Kaempferide ameliorates cisplatin-induced nephrotoxicity via inhibiting oxidative stress and inducing autophagy

Abstract

Acute kidney injury (AKI) caused by anti-tumor drugs, such as cisplatin, is a severe complication with no effective treatment currently, leading to the reduction or discontinuation of chemotherapy. Natural products or herbal medicines are gradually considered as promising agents against cisplatin-induced AKI with the advantages of multi-targeting, multi-effects, and less resistance. In this study, we investigated the effects of kaempferide, a natural flavonoid extracted from the rhizome of Kaempferia galanga, in experimental AKI models in vitro and in vivo. We first conducted pharmacokinetic study in mice and found a relative stable state of kaempferide with a small amount of conversion into kaempferol. We showed that both kaempferide (10 μM) and kaempferol (10 μM) significantly inhibited cisplatin-caused injuries in immortalized proximal tubule epithelial cell line HK-2. In AKI mice induced by injection of a single dose of cisplatin (15 mg/kg), oral administration of kaempferide (50 mg/kg) either before or after cisplatin injection markedly improved renal function, and ameliorated renal tissue damage. We demonstrated that kaempferide inhibited oxidative stress and induced autophagy in cisplatin-treated mice and HK-2 cells, thus increasing tubular cell viability and decreasing immune responses to attenuate the disease progression. In addition, treatment with kaempferide significantly ameliorated ischemia-reperfusion-induced renal injury in vitro and in vivo. We conclude that kaempferide is a promising natural product for treating various AKI. This study has great implications for promotion of its use in healthcare products, and help to break through the limited use of cisplatin in the clinic.

Introduction

Medications-induced acute kidney injury (AKI), characterized by a dramatic decrease in urine output and a harmful accumulation of blood urea nitrogen (BUN) and serum creatinine (Cre), is a relatively common side effect that occurs in up to 30% of hospital patients who are exposed to numerous agents and can even lead to death in severe cases [1–3]. Cisplatin is one of the most effective broad-spectrum anti-cancer drugs and is widely used to treat a variety of solid malignancies, such as lung and ovarian cancers [1]. However, around 25% of patients treated with cisplatin suffer severe nephrotoxicity, a major side effect of cisplatin, leading to the reduction or discontinuation of chemotherapy [4]. Up to now, mannitol and preloading magnesium have been shown to significantly decrease the incidence of cisplatin-induced AKI (cis-AKI) in clinical trials [5, 6]. Even so, it is still not known how to effectively alleviate the occurred nephrotoxicity by cisplatin.

The AKI caused by cisplatin is cumulative and dose-dependent, reactive oxygen species (ROS)-dependent renal cell death and inflammation in the proximal tubules are the hallmarks of cis-AKI, leading to the proximal tubular injury including the loss of brush borders, epithelial cell necrosis and sloughing and obstruction [7–9], but the underlying mechanisms have not been fully elucidated. In addition, the cytoprotective role of autophagy in AKI has been brought to our attention recently. Pharmacological inhibition of autophagy by 3-MA in kidney epithelial cell line LLC-PK1, or knockout Atg7 in primary proximal tubular epithelial cells increases their sensitivity to cisplatin and increased cell apoptosis [10, 11]. Autophagy gene knockout mice exhibit more severe loss of renal function and tissue damage after cisplatin treatment [12], suggesting autophagy manipulation may be a potential therapeutic strategy against cisplatin attack.

Natural products or herbal medicines are gradually considered as promising agents against cis-AKI with the advantages of multi-targeting, multi-effects, and less resistance [13, 14]. Kaempferide, a natural flavanol compound found in the rhizome of Kaempferia galanga, Chromolaena odorata, Alpinia conchigera, and Hippophae rhamnoides L [15–17]. Kaempferide is a 4′-O-methyl derivative of kaempferol, which shows great pharmacological activities with nutraceutical and medicinal applications in human healthcare, such as Brazilian green propolis [18]. Given that liver microsomes can metabolize kaempferide to kaempferol with K_m values of 17.8 μM [19], most studies focus on the bioactivity of kaempferol rather than kaempferide. However, two pharmacokinetics and metabolism studies indicated that both kaempferide and kaempferol can be found in plasma after metabolism [18, 20], with the maximum concentration (C_max) values of 276.22 ± 58.51 ng/mL for kaempferide and 54.03 ± 8.24 ng/mL for kaempferol after an oral administration of Sedum sarmentosum Bunge extract [18, 20], suggesting the potential direct activity of kaempferide in vivo. Additionally, although the similar structure of kaempferide and kaempferol leads to many analogous characters, including anti-inflammation and antioxidants, common but differentiated responsibilities can still be found between them [21, 22]. Thus, although the protection of kaempferol on cis-AKI is already known, it is still of great worth exploring the role of kaempferide to treat the nephrotoxicity induced by cisplatin.

In this study, we try to understand the pharmacokinetic properties of kaempferide and then investigate its protective role in AKI. In vitro and in vivo studies demonstrate that kaempferide treatment attenuates cisplatin-induced kidney injury in a dose-dependent manner. The underlying mechanism of its role is supposed to be acted as an anti-oxidant and autophagy activator. Moreover, kaempferide also functions well in the treatment of ischemia-reperfusion-induced AKI, suggesting that kaempferide is a potential reagent in the clinical treatment of various acute kidney injuries.

Materials and methods

Mice

C57BL/6 mice (male, 6–8 weeks old) were obtained from Zhejiang Vital River Laboratory Animal Technology Co., Ltd. All mice were housed in a specific pathogen-free environment at 21 ± 1 °C and 60% ± 5% humidity, with a 12 h light-dark cycle. All animal use and studies were performed in compliance with all relevant ethical regulations. The procedures conducted on animals were approved by the Institutional Animal Care and Use Committee (IACUC) of Zhejiang University.

Compounds and drugs

Kaempferide (Cat No. N04195) was purchased from Shanghai Merrill Chemical Technology Co., Ltd, Shanghai, China. Kaempferol (Cat No. T2177) was purchased from TargetMol, Shanghai, China. 3-MA (Cat. No. 189490) was purchased from Sigma-Aldrich, USA. Cisplatin (Cat. No. S20180314) was purchased from Jiangsu Haosen Pharmaceutical Co., Ltd, Jiangsu, China.

Pharmacokinetic study of kaempferide

Mice were randomly divided into eight groups to receive the kaempferide dosage level at 50 mg/kg (ig). The compound was suspended in 60% aqueous PEG-200 before the experiment. Mice were sacrificed at 5 min, 10 min, 30 min, 1 h, 2 h, 4 h, 8 h and 12 h after dosing to collect blood and kidney samples. The blood samples were then centrifuged at 3000 rpm for 5 min to obtain the plasma. The kidneys were collected rapidly and rinsed thoroughly with normal saline (0.9%) and 0.3 g tissue samples were homogenized in 0.9% normal saline (1:3, w/v). All the samples were stored at −80 °C until analysis.

An LC-MS/MS method was developed to determine the concentration of kaempferide and kaempferol in plasma and kidney. Plasma (50 μL) or tissue homogenate was added to 150 μL of internal standard dissolved in acetonitrile to precipitate the protein. Then samples were centrifuged at 12,000 rpm for 5 min. The supernatant layer was collected for LC-MS/MS analysis. Samples were detected by an AB SCIEX 5500 mass spectrometer (AB SCIEX Corp.). The pharmacokinetic parameters were calculated by DAS 3.0 software with non-compartment analysis.

HK-2 cells culture

HK-2 cells (immortalized human proximal tubule epithelial cell line) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). HK-2 cells were cultured in RPMI-1640 medium with 10% fetal bovine serum maintained in a 37 °C, 5% CO₂ cell incubator. HK-2 cells viability was measured by the SRB method as described in previous studies [23].

AKI model establishment and drug treatment

To establish a cisplatin-induced AKI model, mice received a single intraperitoneal injection of cisplatin at 15 mg/kg. Kaempferide was treated by ig at 25 mg·kg⁻¹·d⁻¹ or 50 mg·kg⁻¹·d⁻¹ for continuous 5 days or 3 days, and 3-MA was treated by ip at 30 mg·kg⁻¹·d⁻¹ for 2 day.

To establish an ischemia-reperfusion-induced AKI model, mice were anesthetized by ip injection of 1% pentobarbital sodium, exposed to bilateral renal tips for clamping to induce ischemia for 40 min, and then the clamps were released for reperfusion for 24 h. A single dose of kaempferide (50 mg/kg) was administered (ig) 2 h after ischemia.

Serum biochemistry

The blood of mice was centrifuged at 5000 rpm and 4 °C for 15 min, and the supernatant was aspirated to obtain serum (without anticoagulant). Blood urea nitrogen (BUN) and serum creatinine (Cre) levels were measured by BUN kit (Cat. No. 04460715190, Roche Cobas, IN, USA) and Cre kit (Cat. No. 04810716190, Roche Cobas, IN, USA).

Renal histopathology

Kidneys from control and AKI mice were fixed in 4% formalin, embedded in paraffin and sectioned. Slides were dewaxed and stained with hematoxylin and eosin (H&E) to evaluate the score of kidney damage. For semiquantitative analysis of morphological changes, two sections were randomly selected from each sample of at least 3 for every group and 5 fields were randomly selected from each section. Histopathological changes were evaluated by the percentage of injured/damaged renal tubules, as indicated by tubular lysis, dilation, disruption, and cast formation. The renal tissue damage was evaluated by the following scoring system: 0 score, no damage; 1 score, <25% tubular damaged; 2 score, 25%–50% tubular damaged; 3 score, 50%–75% tubular damaged; 4 score, >75% tubular damaged.

RNA extraction, quantitative PCR and RNA-sequencing

Total RNA was isolated from kidney tissues or HK-2 cells using Trizol reagent (Invitrogen, CA, USA), and cDNA was prepared using a cDNA preparation supermix (TransGen Biotech, Beijing, China). Quantitative PCR was carried out using SYBR Green Supermix (Bio-Rad, CA, USA) with 20 μL reaction mixtures. PCR reactions were performed on the Quant Studio 6 Flex Real-Time PCR System (Applied Biosystems, CA, USA). The gene expression was normalized to the reference gene encoding Gapdh/β-Actin using the delta-delta CT method. PCR primer sequences are available upon request. Total RNA was isolated from the renal tissues of control and kaempferide-treated AKI mice, and then submitted for subsequent cDNA library preparation and RNA sequencing.

Western blot analysis

Protein extracts were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, 8%–12% gels) and blotted onto PVDF membranes. After blocking with 5% fat-free milk, the membranes were incubated at 4 °C overnight with the following primary antibodies: anti-cleaved PARP (1:1000, 9544 s, Cell Signaling Technology); anti-P62 (1:1000, ab56416, Abcam); anti-OPTN (1:1000, ab23666, Abcam); anti-LC3 (1:1000, 12741 s, Cell Signaling Technology); anti-AMPK (1:1000, 12741 s, Cell Signaling Technology); anti-p-AMPKα1 (Thr172) (1:1000, sc-101630, Santa Cruz); anti-AKT (1:1000, 9272 s, Cell Signaling Technology); anti-p-AKT (ser 473) (1:1000, 4060 s, Cell Signaling Technology); anti-GAPDH (1:5000, db106, Diag Biotechnology); anti-β-Actin(1:5000, db6001, Diag Biotechnology). The bound antibodies were detected using horseradish peroxidase-conjugated IgG (Multi Sciences) and visualized with enhanced chemiluminescence detection reagents (FDbio Science Biotech Co., Ltd.). GAPDH/β-Actin was used as a loading control. Images were taken by GE AI600.

mCherry-GFP-LC3 dots quantification

HK-2 cells were transfected with mCherry-GFP-LC3 plasmid, and 12 h after transfection, cells were treated with kaempferide and cisplatin for another 12 h. Afterwards, cells were observed under a confocal microscope and the number of mCherry-LC3 and GFP-LC3 dots formed per cell was counted. Colocalization between GFP and mCherry denotes autophagosomal structures that have not already fused with a lysosomal compartment (phagophores or autophagosomes) whereas mCherry signal without GFP signal denotes autophagosomal acidified organelle that have fused with lysosomes (amphisomes or autolysosomes).

Immunostaining

Kidney cryosections were rinsed 3 times with PBS for 5 min each, then blocked with PBS containing 5% goat serum and 0.3% Triton X-100 for 1 h. Then sections were incubated with primary antibody TNF-α (ab6671, Abcam) overnight at 4 °C, and followed by fluorescent secondary antibodies for 1 h and subsequent DAPI for 5 min, washed in PBS, and mounted in Fluoromount-G.

TUNEL assay

Fluorescent terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay was performed using a kit from Beyotime (Cat No. C1088, Shanghai, China) according to the manufacturer’s instructions. The staining results were observed with a fluorescence microscope.

Intracellular ROS measurement

HK-2 cells were incubated with cisplatin (5 μM) and kaempferide (10 μM) for 24 h. Intracellular ROS was detected by a flow cytometer (BD Biosciences) using a fluorescent probe (2,7-dichlorodihydrofluorescein diacetate, DCFH-DA), a redox-sensitive fluorescent probe that measures hydroxyl, peroxyl, and other ROS level in cells.

MDA and SOD assays

To assess the level of oxidative stress in HK-2 cells and kidneys, SOD and MDA levels were measured using a commercially available kit (Beyotime, Shanghai, China). Briefly, HK-2 cells or the renal tissues were homogenized in ice-cold phosphate buffer (0.1 M, pH 7.4), then the homogenate was filtered and centrifuged at 4 °C. The obtained supernatant was used to determine SODase activity and lipid peroxidation levels by measuring the MDA content. SODase activity was shown as units of activity per mg of protein, and MDA content was shown as micromoles per gram of protein.

Hypoxia-reoxygenation in HK-2 cells

Normal oxygen cultured HK-2 cells were switched to sugar-free DMEM medium and immediately placed in a sealed incubator, then N₂ gas was loaded to reduce the oxygen concentration to 1% and incubated at 37 °C with 5% CO₂ for 10 h. Then cells were transferred back to regular RPMI-1640 medium and 21% oxygen in a conventional incubator for 12 h reperfusion. Cells cultured in regular medium and conventional incubators were regarded as controls.

Statistical analysis

Data were analyzed using ImageJ (version 1.8.0) and GraphPad Prism software (version 8.0). Unpaired two-tailed Student’s t-test test was used between two groups, and one-way ANOVA Tukey’s post-hoc analysis was used for multiple comparisons to determine statistical significance. All data are from at least three independent experiments and presented as means ± SEM. The statistical significance was considered when P < 0.05 (represented as *P < 0.05, **P < 0.01, ***P < 0.001, or not significant (n.s.)).

Results

Stable kaempferide is detected in vivo revealed by a pharmacokinetic study

Given kaempferide is able to convert into kaempferol in liver microsomes [19], a pharmacokinetic study of kaempferide was performed to see the exact conversion proportion in vivo. Results showed that kaempferide was quickly absorbed and distributed into tissues like kidney (Fig. 1a–c). As excepted, kaempferol was also detected in plasma and kidney of mice treated with kaempferide (Fig. 1a–c), suggesting a conversion from kaempferide to kaempferol really happaned in vivo. In details, the C_max of kaempferide was 58.87 ± 30.69 ng/mL in plasma and 1765.33 ± 587.75 ng/g in kidney, and the area under curve (AUC) of kaempferide was 48.57 ± 19.99 ng/mL*h in plasma and 4947.11 ± 146.03 ng/g*h in kidney, while the C_max of kaempferol was 10.27 ± 2.37 ng/mL in plasma and 346 ± 96.26 ng/g in kidney, and the AUC of kaempferol was 16.08 ± 5.42 ng/mL*h in plasma and 732.97 ± 110.19 ng/g*h in kidney (Fig. 1d–g). Overall, these data indicated that the majority of kaempferide is maintained in its primary state after absorption in vivo, and only a small amount of kaempferide is metabolized into kaempferol.

Fig. 1. Pharmacokinetic study of kaempferide and its protection in HK-2 cell toxicity.

a Chemical structure of kaempferide (KF) and kaempferol (KFR). b, c Concentration-time curve in plasma (b) and kidney (c), n = 3 animals/group. d, e The C_max of kaempferide and kaempferol in plasma (d) and kidney (e). f, g The AUC of kaempferide and kaempferol content in plasma (f) and kidney (g). h Relative mRNA expression of KIM-1 in HK-2 treated with vehicle, cisplatin (5 μM), cisplatin plus kaempferide (10 μM), cisplatin plus kaempferol (10 μM), respectively. i Western blot analysis of the Clv-PARP expression in HK-2 treated with vehicle, cisplatin, cisplatin plus kaempferide, cisplatin plus kaempferol. β-Actin served as a loading control. j Relative mRNA expression of Bax and Bcl2 in HK-2 treated with vehicle, cisplatin, cisplatin plus kaempferide, cisplatin plus kaempferol, respectively. Data are presented as means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

Kaempferide has a protective effect in HK-2 cells suffered from cisplatin

Since the renal proximal tubule epithelial cells are regarded as the target cells of the nephrotoxicity caused by cisplatin [24], the immortalized proximal tubule epithelial cell line HK-2 was applied to examine if kaempferide can directly antagonize the tubule cell toxicity induced by cisplatin. Results showed that the gene expression of kidney injury molecular-1 (KIM-1), an early indicator of kidney tubular injuries, was dramatically upregulated in cisplatin-treated HK-2 cells, and then were downregulated significantly upon kaempferide or kaempferol incubation (Fig. 1h). Further, both kaempferide and kaempferol inhibited cisplatin-induced HK-2 cell apoptosis as evidenced by reduced expression of Cleaved-PARP (Clv-PARP) and Bax/Bcl2 ratio (Fig. 1i, j). Taken together, these data revealed the protective role of kaempferide itself in renal proximal tubule cell toxicity caused by cisplatin.

Kaempferide prevents the renal side effects caused by cisplatin in vivo

To investigate if kaempferide can reduce the occurrence and severity of cis-AKI in vivo, a single dose of cisplatin (15 mg/kg) was intraperitoneally injected into mice to induce AKI, and oral administration of kaempferide (25, 50 mg·kg⁻¹·d⁻¹) was pretreated for continuous 5 days (Fig. 2a). The serum and renal tissues were collected for analyses 3 days after cisplatin injection. Biochemistry analyses showed that the mice suffering from cis-AKI exhibited severe nephrotoxicity, as evidenced by their significantly accumulated BUN and serum Cre levels in mice treated with cisplatin. Pretreatment of kaempferide substantially lowered the elevated levels of BUN and Cre in cis-AKI mice (Fig. 2b, c). Histopathology analyses revealed that the cis-AKI mice exhibited a dramatic drop in blood flow to the kidneys, which were characterized by obvious necrosis, renal tubule detachment, brush border loss, tubular degeneration, and vacuolization. However, kaempferide dramatically reduced this renal tissue damage (Fig. 2d, e). Consistently, kaempferide decreased the significantly upregulated mRNA expression of KIM-1 and Neutrophil gelatinase-associated lipocalin (NGAL), two biomarkers of kidney tubular injuries (Fig. 2f, g) [25, 26]. In addition, kaempferide ameliorated cisplatin-induced tubular apoptosis, as indicated by decreased expression of Clv-PARP in the renal cortex (Fig. 2h). Overall, these results suggested that kaempferide protects against cisplatin-induced nephrotoxicity.

Fig. 2. Kaempferide prevents the nephrotoxicity caused by cisplatin.

a Schematic illustration of kaempferide (KF) treatment schedule for AKI prevention induced by cisplatin. b, c BUN and serum Cre levels of mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. n = 6 animals/group. d, e Representative H&E staining and the morphology of kidneys from each group. Arrows indicated tubules with necrosis, anoikis, or brush border loss, asterisks indicated cast formation. Scale bar: 100 μm. The tubular injury degree was scored in each group. n = 6 animals/group. f, g Relative mRNA expression of KIM-1 and NGAL in the renal cortex from mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. h Western blot analysis of the Clv-PARP expression after cisplatin and kaempferide treatment. GAPDH served as a loading control. Data are presented as means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

To further evaluate the therapeutic role of kaempferide on cis-AKI, kaempferide (50 mg/kg) was repeatedly gavaged 2, 24, and 48 h after cisplatin injection, and then the serum and renal tissues were collected for analyses (Fig. 3a). As expected, the elevated BUN and serum Cre in cis-AKI mice almost got back to normal levels after the administration of kaempferide (Fig. 3b, c). In addition, the anatomical study on the morphology of the kidneys showed a sufficient blood supply in kaempferide treated cis-AKI mice. H&E staining results also indicated kaempferide treatment improved the pathological variation including necrosis and tubular degeneration (Fig. 3d, e). Besides, the significantly upregulated gene expression of KIM-1 and NGAL, and the number of apoptotic proximal tubule cells in the renal tissues from cis-AKI mice was decreased by kaempferide administration as determined by qPCR and TUNEL staining, respectively (Fig. 3f–h). Taken together, all these data revealed that kaempferide treatment improves the renal damage caused by cisplatin.

Fig. 3. Kaempferide treatment attenuates cisplatin-induced AKI.

a Schematic illustration of kaempferide (KF) treatment schedule for AKI therapy induced by cisplatin. b, c BUN and serum Cre levels of mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. n = 4 animals/group. d, e Representative H&E staining and the morphology of kidneys from each group. The dashed area indicated the tubular damage, arrows indicated tubules with necrosis, anoikis, or brush border loss, and asterisks indicated cast formation. Scale bar: 200 μm. The tubular injury degree was scored in each group. n = 3 animals/group. f, g Relative mRNA expression of KIM-1 and NGAL in the renal cortex from mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. h Representative TUNEL staining images in kidneys from mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. Scale bar: 50 μm. Quantification of TUNEL positive nuclei in kidneys of each group. n = 3 animals/group. Data are presented as means ± SEM; **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

Immune response is inhibited in cis-AKI mice treated with kaempferide

In order to explore the underlying mechanism of the protective role of kaempferide in cis-AKI, we collected the renal tissues from cis-AKI mice treated with or without kaempferide and submitted them for RNA-sequencing. Gene ontology (GO) analysis indicated that differentially expressed genes that were down-regulated in kaempferide-treated kidneys were enriched in the inflammatory response and cell death, suggesting the protective role of kaempferide in cis-AKI (Fig. 4a). The anti-inflammatory effect of kaempferide was validated by qPCR for anti-inflammatory cytokines and pro-inflammatory cytokines. Results showed that kaempferide significantly upregulated the gene expression of anti-inflammatory factors including Il-4 and Il-10 (Fig. 4b, c), and reduced the increased gene expression of pro-inflammatory cytokines including Tnf-α, Il-6 and Il-1β (Fig. 4d–f). And the reduced pro-inflammatory cytokine levels of TNF-α and IL-6 after kaempferide treatment were further measured by immunostaining and ELISA (Fig. 4g, h). Thus, the above data indicated that kaempferide alleviates the immune response in cis-AKI.

Fig. 4. The immune response is reduced in kaempferide-treated cis-AKI mice.

a GO analysis of the downregulated genes in the renal tissues from kaempferide treated mice. b–f Relative mRNA expressions of Il-4, Il-10, Tnf-α, Il-1β, and Il-6 in the renal cortex from each group. g Representative images of immunofluorescence staining for TNF-α antibody in kidneys. Scale bar: 50 μm. n = 3 animals/group. h ELISA to detect the secretion level of IL-6 in kidney tissues from each group. Data are presented as means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

Kaempferide regulates oxidative stress to confront cis-AKI

Then we would like to know how kaempferide functions in confront cis-AKI. GO analysis indicated that the upregulated gene signatures in kidneys from kaempferide treated cis-AKI were rich in oxidoreductase activity and antioxidant activity (Fig. 5a). Given the fact that elevated intracellular levels of reactive oxygen species (ROS) cause cell damage. Malondialdehyde (MDA) is one of the final products of lipid peroxidation by ROS in cells, and its concentration is widely used as a marker of oxidative stress [27]. Flow cytometry analysis with DCFH-DA probe showed that the relative ROS level was remarkably higher in HK-2 cells after cisplatin incubation for 24 h when compared with control, while kaempferide treatment was able to block the intracellular ROS release caused by cisplatin (Fig. 5b, c). Consistently, kaempferide also reversed the raised level of MDA in HK-2 cells induced by cisplatin (Fig. 5d), indicating the opposed role of kaempferide on oxidative stress. Meanwhile, the lower activity of superoxide dismutase (SOD), an antioxidant enzyme, was significantly raised by kaempferide (Fig. 5e), in keeping with previous studies. We then investigated the antioxidant activity of kaempferide in cis-AKI models. As expected, the MDA level in the renal tissues almost got back to normal after kaempferide administration into the cis-AKI mice, as well as the reversed SOD activity (Fig. 5f, g). Therefore, kaempferide is able to control oxidative stress in renal tubule cells to exert a protective effect against cis-AKI.

Fig. 5. Cisplatin-induced oxidative stress is inhibited after kaempferide treatment.

a GO analysis of the upregulated genes in the renal tissues from kaempferide treated mices. b Representative ROS images in HK-2 cells detected by flow cytometry after cisplatin and kaempferide incubation. c Statistical results of ROS level in HK-2 cells with cisplatin and kaempferide treatment. d, e The MDA level and SOD activity from HK-2 cells exposed to cisplatin and kaempferide. f, g Detection of MDA level and SOD activity in renal cortex from mice treated with vehicle, cisplatin, cisplatin plus kaempferide, respectively. n = 3 animals/group. Data are presented as means ± SEM; **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

Kaempferide is able to induce autophagy in HK-2 cells

Interestingly, kaempferide is recently identified as an autophagy enhancer in adipocytes by targeting a mitochondrial translation factor TUFM to induce autophagosome formation [28, 29]. We then examined if kaempferide can induce autophagy in the renal tubular cells. A decreased expression of phosphorylated protein kinase B (p-AKT) and an increased expression of phosphorylated AMP-activated protein kinase (p-AMPK) were detected in kaempferide treated HK-2 cells by Western blotting assay (Fig. 6a), both of which can promote autophagy directly [30, 31]. As a result, the gene expression of Beclin1, an autophagosome marker, was significantly upregulated in HK-2 cells with kaempferide incubation (Fig. 6b). And an increased expression of Optineurin (OPTN) and microtubule-associated protein 1 light chain 3 (LC3 II) were observed in kaempferide treated HK-2 cells, which were specifically associated with autophagosome formation and autolysosome degradation (Fig. 6c). The autophagic flux was further measured in HK-2 cells transfected with mCherry-GFP-LC3 II. Results indicated kaempferide increased the autophagic flux significantly as analyzed by the increased number of autolysosomes (red dots) in kaempferide treated HK-2 cells (Fig. 6d). All these data suggested that autophagy can be induced by kaempferide in HK-2 cells.

Fig. 6. Autophagy is induced after kaempferide treatment in HK-2 cells.

a Western blot analysis of the p-AKT, AKT, p-AMPK and AMPK expression in HK-2 cells treated with vehicle and kaempferide. GAPDH served as a loading control. b Relative mRNA expression of Beclin-1 in HK-2 cells treated with vehicle, kaempferide, respectively. c Western blot analysis of the OPTN, P62 and LC3 II expression in HK-2 cells treated with vehicle, kaempferide. GAPDH served as a loading control. d HK-2 cells transfected with mCherry-GFP-LC3 plasmid were treated with kaempferide for 12 h and the GFP, mCherry signal was detected by immunofluorescence. Scale bars: 10 μm. e Western blot analysis of the OPTN, P62 and LC3 II expression in the renal cortex treated with vehicle, cisplatin, cisplatin plus kaempferide. GAPDH served as a loading control. f HK-2 cells transfected with mCherry-GFP-LC3 plasmid were treated with cisplatin, cisplatin plus kaempferide, cisplatin and kaempferide plus 3-MA for 12 h and the GFP and mCherry signals were detected by immunofluorescence. Scale bars: 10 μm. g The survival rate of HK-2 cells upon treatment with cisplatin (5 μM), kaempferide (10 μM) and 3-MA (10 μM) for 24 h. Data are presented as means ± SEM; n.s. not significant, *P < 0.05, **P < 0.01, ***P < 0.001; Student’s t test or one-way ANOVA with multiple comparisons test.

Autophagy induction by kaempferide is required to alleviate cis-AKI

Research has demonstrated that autophagy plays a key role in the progression of cis-AKI [32, 33], autophagy induction in tubular cells has been documented in cis-AKI kidneys, while proximal tubule-specific Atg7 knockout increased sensitivity to cis-AKI [10, 11], indicating a protective role of tubular cell autophagy in cis-AKI. We then guess autophagy may also be involved in the protective role of kaempferide on cis-AKI. As expected, Western blotting results showed that cisplatin induced autophagy in HK-2 cells, and kaempferide treatment further activated autophagy with cisplatin as indicated by the degradation of autophagy receptor OPTN and P62, the increased expression of LC3 II and the increased autophagic flux (Fig. 6e, f). 3-Methyladenine (3-MA), a widely used autophagy inhibitor because of its inhibitory effect on class III PI3K [34], was further applied to examine whether autophagy activity is involved in the protective role of kaempferide in alleviating cisplatin-induced HK-2 cell damage. Results showed that 3-MA significantly inhibited kaempferide-induced autophagy, causing the invalid protection of kaempferide in HK-2 cells (Fig. 6f, g). These data indicated kaempferide-induced autophagy was required for its protective effect on HK-2 cell viability.

Similar results were observed in cis-AKI model. In details, Western blotting results showed that kaempferide induced autophagy in cis-AKI kidneys as indicated by the degradation of OPTN and P62 and the increased expression of LC3 II (Fig. 7a). To further validate the key role of autophagy in the protection of kaempferide, a pretreatment of kaempferide (50 mg/kg) for total of 5 days to prevent cis-AKI, and an intraperitoneal injection of 3-MA (30 mg/kg) was given on the second and third day to inhibit the autophagy induced by kaempferide (Fig. 7b). Results showed that kaempferide cannot reduce the elevated BUN and serum Cre levels after autophagy inhibition by 3-MA administration (Fig. 7c, d). Moreover, the reduced expression of Clv-PARP and the protection of the damaged renal tissue by kaempferide was abandoned upon 3-MA treatment (Fig. 7e–g). Taken together, these data suggest that in vivo inhibition of autophagy significantly counteracts the protective role of kaempferide in cis-AKI mice.

Fig. 7. Autophagy induction is required in the renal protection of kaempferide.

a Western blot analysis of the OPTN, P62 and LC3II expression in the renal cortex treated with vehicle, cisplatin, cisplatin plus kaempferide. GAPDH served as a loading control. b Schematic diagram of the model of AKI mice treated with cisplatin, kaempferide and 3-MA. c, d BUN and serum Cre levels of mice treated with vehicle, cisplatin, cisplatin plus kaempferide, cisplatin plus 3-MA, cisplatin plus kaempferide and 3-MA. n = 6 animals/group. e Western blot analysis of Clv-PARP expression in the renal cortex from each group. GAPDH served as a loading control. f, g Representative H&E sections of the kidney in each group. Arrows indicate the location of tubular damage, including tubular necrosis and brush border loss, and asterisks indicate protein cast deposition. Scale bar: 200 μm. The tubular injury degree was scored in each group. n = 6 animals/group. Data are presented as means ± SEM; n.s. not significant, *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

Kaempferide protects against ischemia-reperfusion induced AKI

Given that kaempferide has a great effect on cisplatin-induced AKI, we then wonder if kaempferide is effective on other AKI, including ischemia-reperfusion induced AKI (IR-AKI), which represent the most common reasons for AKI in hospitalized patients [35]. To mimic the condition of ischemia-reperfusion in vitro, HK-2 cells with hypoxia-reoxygenation (HR) were applied in the study. Results indicated that the cell survival was significantly decreased upon HR, while obviously reversed after kaempferide treatment (Fig. 8a). Besides, HR stimulated HK-2 cell apoptosis compared with control, which was remarkably blocked by kaempferide (Fig. 8b). Similar results were observed in the gene expression of KIM-1 in kaempferide treated cells (Fig. 8c). Overall, these data suggested that kaempferide protects against HR-induced HK-2 cell injury.

Fig. 8. Kaempferide treatment attenuates ischemia-reperfusion-induced AKI.

a Cell viability of HK-2 cells with HR treated with kaempferide for 24 h. b Western blot analysis of the Clv-PARP expression in HK-2 cells after HR and kaempferide (10 μM) treatment. GAPDH served as a loading control. c Relative mRNA expression of KIM-1 in HK-2 cells from each group. d Schematic diagram of kaempferide treatment schedule for IR-AKI. e BUN and serum Cre levels of IR-AKI mice treated with or without kaempferide. n = 9 animals/group. f Representative H&E staining of kidneys from each group. The dashed area indicated the tubular injury. The arrows indicated the loss of tubular structure and epithelial necrosis, asterisks indicated cast formation. Scale bar: 200 μm. n = 9 animals/group. g The tubular injury degree was scored in each group. h Relative mRNA expression of KIM-1 and NGAL in the renal cortex from each group. i Western blot analysis of the Clv-PARP expression from the renal tissues of each group. β-Actin served as a loading control. j Relative mRNA expression of IL-6 and IL-1β in the renal cortex from each group. k Detection of MDA level and SOD activity in the renal cortex from each group. n = 3 animals/group. l Western blot analysis of the OPTN, P62 and LC3II expression from the renal tissues of each group. β-Actin served as a loading control. Data are presented as means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with multiple comparisons test.

To further investigate the effect of kaempferide in vivo, a model of IR-AKI was constructed using a bilateral renal vascular ligation. Kaempferide (50 mg/kg) was orally administered 2 h after ligation, mice were harvested, and serum and renal tissues were collected at 24 h (Fig. 8d). Interestingly, the administration of kaempferide significantly reduced the BUN and serum Cre levels in IR-AKI mice (Fig. 8e). H&E staining results showed that compared with the control group, ischemia-reperfusion induced pathological changes such as renal tubular necrosis and vacuolization, and kaempferide significantly reduced the injured area (Fig. 8f, g). Consistently, the gene expression of kidney tubular injury biomarkers, like KIM-1 and NGAL, was downregulated in kaempferide treated IR-AKI mice (Fig. 8h). The cell apoptosis and inflammation in IR-AKI kidneys were also inhibited by kaempferide treatment as evidenced by the decreased protein expression of Clv-PARP and gene expression of pro-inflammatory cytokines such as Il-6 and Il-1β (Fig. 8i, j). The above results suggested that kaempferide significantly ameliorates IR-AKI in mice.

Given that oxidative stress and autophagy are also important in the progression of IR-AKI [36, 37], we next investigated whether oxidative stress and autophagy were included in the IR-AKI protection of kaempferide. Preliminary data showed that the MDA level in the renal tissues from IR-AKI almost got back to normal after kaempferide administration, while the SOD activity was increased significantly in kaempferide treated IR-AKI kidneys, indicating an anti-oxidative role of kaempferide in IR-AKI (Fig. 8k). Besides, we also detected a degradation of OPTN and P62 and an increased expression of LC3 II (Fig. 8l), suggesting autophagy was activated in kaempferide treated IR-AKI kidneys. These data revealed a possible regulation of oxidative stress and autophagy in kaempferide’s protection of IR-induced renal injury, and more in vitro and in vivo studies are needed to get a further validation.

Discussion

In this study, we first observed a relative stable state of kaempferide with only a small amount of conversion into kaempferol by a pharmacokinetic study. Then, both in vitro and in vivo studies indicated kaempferide had a therapeutic role on either cisplatin or ischemia-reperfusion induced acute kidney injury. And the protective role of kaempferide might be through inhibiting oxidative stress and inducing autophagy (Fig. 9). These data suggested that kaempferide can be a promising drug for treating various acute kidney injuries.

Fig. 9. Schematic diagram of kaempferide’s protection against acute kidney injury.

Kaempferide can inhibit oxidative stress and induce autophagy in tubular cells, leading to reduced cell death and immune response in kidneys, thus protecting against both cisplatin-induced and ischemia-reperfusion-simulated acute kidney injury.

Although many studies held the view that kaempferol is the active metabolite of kaempferide and investigated the bioactivity of kaempferol only, a small number of studies compared the function of kaempferol and kaempferide at the same time and explored their different roles in diseases [21, 22]. In this study, we showed that only a small amount of kaempferide was converted into kaempferol, and the similar renal protection between kaempferide and kaempferol was found in HK-2 cells (Fig. 1). Thus, we can say kaempferide prevents from AKI mainly by itself and may partially by its conversion to kaempferol. In view of the high dose (200 mg/kg) and long time (14 day) of kaempferol pre-treatment for the prevention of cis-AKI, we found a similar function with a lower dose (50 mg/kg) and a shorter time period (5 days) of kaempferide. However, our head-to-head comparison in vitro showed a similar efficiency between kaempferide and kaempferol (Fig. 1h–j). Thus, further head-to-head comparisons in vivo are needed to see which one functions better in AKI protection. On the other hand, kaempferol is also reported to have protective effects on other kidney diseases such as calcium oxalate crystal-induced renal injury, renal fibrosis, and diabetic nephropathy [38–40]. But for kaempferide, little is known if it protects against other chronic kidney diseases, and further studies need to be carried out.

A natural compound may have multiple active targets rather than one unique target. Generally, the flavanol compounds can control the oxidative stress to attenuate disease progress, kaempferide is no exception [41–43]. Here we also investigated the regulation of oxidative stress after kaempferide treatment. Inhibited oxidative stress were observed in kaempferide-treated cis-AKI and IR-AKI mices (Figs. 5, 8). Till now, we only know kaempferide protected the renal cells from death both in vivo and in vitro, many detailed mechanisms are out of master. For example, as we still do not know the direct target of kaempferide during oxidative stress, even though we have examined it in HK-2 cells. More studies are needed to find out the signaling pathways that kaempferide participates in to inhibit oxidative stress, and how kaempferide intervenes in the relevant signals.

Autophagy is a conserved lysosomal pathway used to degrade cytoplasmic components [44]. Basal autophagy in renal cells is essential for the maintenance of renal homeostasis, structure, and function [33], while autophagy dysfunction on renal tubules caused impaired autophagosome formation process and cell apoptosis, leading to severe tubular injury [45, 46]. Specially, given that the autophagy induced by cisplatin is protective for the renal injury [32], we believe the autophagy induced by cisplatin in the renal tissues is more like a feedback activation of kidney injury rather than a direct effect of cisplatin. In addition, since kaempferide is found to be able to induce autophagy [29], here we also examined autophagy is involved in kaempferide’s protection of cis-AKI and IR-AKI, and manipulation of autophagy could intervene in the renal protection of kaempferide (Figs. 6–8). Interestingly, it is said autophagy is closely related to oxidative stress, ROS can induce autophagy, and autophagy in turn can modulate ROS production [47, 48]. Here, for the oxidative stress inhibition and autophagy enhancer of kaempferide, it is possible kaempferide targets a certain protein target in either the oxidative stress pathway or the autophagy pathway, and then causes the alteration of the other pathway. Elsewise, kaempferide has two targets that target the oxidative stress pathway and autophagy pathway, respectively. Both two hypotheses make sense, and more studies are required to make a judgment.

In addition, inflammation was also included in the renal protection of kaempferide. However, only HK-2 cells were applied to examine the oxidative stress and autophagy affected by kaempferide are direct roles on renal tubular cells in this study. But for the inflammation, we only showed inhibited immune response in kaempferide treated cis-AKI and IR-AKI kidneys (Figs. 4, 8). Thus, it is absolutely possible that the anti-inflammatory role was a secondary effect of kaempferide followed by low cell death in the renal tissues. Though it is reported that kaempferide has the capacity to anti-inflammation [49], it still needs further determination whether the reduced inflammation here is a direct role by kaempferide, or just an accompanied effect because of the relieved damage. Taken together, our results here suggested that the oxidative stress inhibition and autophagy activation, two paralleled pathways, were direct effects of kaempferide. And the reduced inflammation might be a kind of protective phenotype from kaempferide.

Since cisplatin is a chemotherapy drug, when we pay attention to its nephrotoxicity, we can not miss its major activity in anti-tumor in vivo. Many reagents that can protect cis-AKI have been reported to attenuate cisplatin-mediated anti-tumor effects. For example, NAC and vitamin E have been indicated to act as antioxidants and contribute to the development of lung cancer [50]. Or things are not even worse is the results that the renal injury preventive drugs, like ceria nanoparticles, have no effect on the anti-tumor activity of cisplatin [23]. Of course, it is better to see the antagonistic role of nephrotoxicity and the synergetic role of the anti-tumor effects. Surprisingly, our preliminary data revealed that kaempferide had a synergetic role with cisplatin on its anti-tumor effects (data not shown). We will do more studies to validate it and explore the underlying mechanism, which will have great implications on the promotion of its use in healthcare products, and help to break through the limited use of cisplatin in the clinical setting.

Author contributions

QJH, JJW and YFS designed and interpreted experiments. YFS, BBT, YHD, CYF, LH performed experiments. CXS, ZXY, YPQ, JCW contributed to the data acquisition. QJH, JJW, YFS and TBB analyzed data and prepared the paper. BY and QJW critically reviewed the paper.

Competing interests

The authors declare no competing interests.