FGF21 is induced in cisplatin nephrotoxicity to protect against kidney tubular cell injury

Abstract

Cisplatin, a widely used cancer therapy drug, induces nephrotoxicity or acute kidney injury (AKI), but the underlying mechanism remains unclear, and renal protective approaches are not available. Fibroblast growth factor (FGF)21 is an endocrine factor that regulates glucose uptake, metabolism, and energy expenditure. However, recent work has also implicated FGF21 in cellular stress response under pathogenic conditions. The role and regulation of FGF21 in AKI are unclear. Here, we show that FGF21 was dramatically induced during cisplatin treatment of renal tubular cells in vitro and mouse kidneys in vivo. The inductive response was suppressed by pifithrin (a pharmacological inhibitor of P53), suggesting a role of P53 in FGF21 induction. In cultured renal tubular cells, knockdown of FGF21 aggravated cisplatin-induced apoptosis, whereas supplementation of recombinant FGF21 was protective. Consistently, recombinant FGF21 alleviated cisplatin-induced kidney dysfunction, tissue damage, and tubular apoptosis in mice. Mechanistically, FGF21 suppressed P53 induction and activation during cisplatin treatment. Together, these results indicate that FGF21 is induced during cisplatin nephrotoxicity to protect renal tubules, and recombinant FGF21 may have therapeutic potential.-Li, F., Liu, Z., Tang, C., Cai, J., Dong, Z. FGF21 is induced in cisplatin nephrotoxicity to protect against kidney tubular cell injury.

Keywords: P53; acute kidney injury; apoptosis.

Figure 1.

FGF21 is induced during cisplatin treatment in BUMPT cells. A) Representative cellular nuclear morphology. BUMPT cells were treated with cisplatin (20 μM) for 24 h and then stained with Hoechst 33342. Cellular and nuclear morphology were recorded by microscopy (original magnification, ×200). Scale bar, 50 μm. B) Apoptosis percentage. Fewer than 200 cells in each group were evaluated to determine the percentage of cells with typical apoptotic morphology. Means ± sd (n = 3). *P < 0.0001 vs. untreated control cells. C, D) BUMPT cells were treated with different concentration of cisplatin (0, 5, 10, and 20 μM) for 24 h (C) or with 20 μM cisplatin for 0, 8, 12, 16, or 24 h (D). Whole cell lysates were collected for immunoblot analysis of C-caspase 3, FGF21, and β-actin as loading control.

Figure 2.

FGF21 is induced in kidney tissues after cisplatin-induced AKI. C57Bl/6 mice (male, 8–10 wk old) were treated with cisplatin (20, 25, or 30 mg/kg body weight) and euthanized at 24, 48, or 72 h. A, B) Blood samples were collected for measurements of BUN (A) and serum creatinine (B). Means ± sd (n = 5). *P < 0.05 vs. untreated control group. C, D, E) Renal tissues were collected for histologic analysis and for immunoblot as well as immunohistochemical analysis of FGF21 expression. Scale bar, 50 μm. C, D) Representative immunoblotting of FGF21 and β-actin (loading control). E) Representative immunohistochemical staining of FGF21 within kidney tissues.

Figure 3.

Suppression of cisplatin-induced FGF21 expression by pifithrin-α. BUMPT cells were treated with cisplatin (20 μM) in the presence or absence of pifithrin-α (20 μM). A) Representative images of cellular and nuclear morphology. Scale bar, 50 μm. B) Percentage of apoptosis. Means ± sd (n = 3). *P < 0.0005 vs. untreated control cells, ^#P < 0.005 vs. cisplatin-only treated KD group cells. C) Representative immunoblots of FGF21, p-P53, P53, C-caspase 3, and β-actin as protein loading control.

Figure 4.

Silencing of FGF21 sensitizes BUMPT cells to cisplatin-induced apoptosis. A) BUMPT stable cell lines were generated by transfecting shRNA pLKO.1-FGF21 plasmids (KD1, KD2) to silence endogenous FGF21 expression or by transfecting NC shRNA vector as control. KD efficiency was determined by immunoblot analysis of FGF21 and β-actin (loading control). B) BUMPT stable cell lines were treated with cisplatin (20 μM) for 24 h to record cell morphology. Scale bar, 100 μm. C) BUMPT stable cell lines were pretreated with or without recombinant FGF21 (50 ng/ml) for 1 h and cultured in the presence or absence of cisplatin (20 μM) for 24 h. Representative immunoblotting of C-caspase 3 and β-actin as protein loading control. D) Percentage of apoptosis. BUMPT stable cell lines were treated with cisplatin (20 μM) for 24 h to assess apoptosis by cell morphology. Means ± sd (n = 3). *P < 0.01 vs. cisplatin treated NC group cells.

Figure 5.

Recombinant FGF21 decreases cisplatin-induced BUMPT cell apoptosis. A, B) BUMPT cells were pretreated with or without recombinant FGF21 (50 ng/ml) for 1 h and cultured in the presence or absence of cisplatin (20 μM) for 24 h. Cells were stained with Hoechst 33342 to record cellular and nuclear morphology. A) Representative images of cellular and nuclear morphology. Scale bar, 100 μm. B) Percentage of apoptosis. Means ± sd (n = 3). *P < 0.0005 vs. untreated control cells; ^#P < 0.005 vs. cisplatin-only treated cells. C, D) BUMPT cells were pretreated with or without recombinant FGF21 (0, 12.5, 25, 50, and 100 ng/ml) for 1 h and incubated with cisplatin (20 μM) for 24 h. Whole cell lysates were collected for immunoblot analysis of C-caspase 3 and β-actin as protein loading control. C) Representative immunoblots. D) Densitometry of C-caspase 3 signals. The cleaved-caspase 3 signals were normalized to the β-actin signal of the same samples to determine the ratios. The ratios of control group (in the absence of cisplatin and FGF21) were arbitrarily set as 1. Means ± sd (n = 3). *P < 0.0001 vs. untreated control cells, ^#P < 0.001 vs. cisplatin-only–treated cells.

Figure 6.

Recombinant FGF21 attenuates cisplatin-induced AKI in mice. C57Bl/6 mice (male, 8–10 wk) were administered cisplatin (30 mg/kg) with or without pretreatment of FGF21 (0.1 μg/g) or with saline as control. Mice were euthanized at 24, 48, or 72 h. A, B) Blood samples were collected for measurements of serum creatinine (A) and BUN (B). C–E) Renal tissues were collected for histologic analysis and tubular cell death analysis. C) Upper panel: Representative histology of kidney cortex by hematoxylin and eosin (H&E) staining. Scale bar, 50 μM. Lower panel: Representative images of TUNEL staining of kidney tissues. Scale bar, 50 μM. D) Quantification of TUNEL-positive cells in kidney tissues. E) Histopathological score of tubular damage. F, G) Renal tissues were collected for immunoblot analysis of KIM-1, C-caspase 3, and β-actin as protein loading control. F) Representative immunoblotting. G) Densitometry of KIM-1 and C-caspase 3 signals. The KIM-1 and C-caspase 3 signals were normalized to the β-actin signal of the same samples to determine the ratios. Means ± sd (n = 4). *P < 0.01 vs. the control group (A, B, D, E), ^#P < 0.05 vs. cisplatin-only group; *P < 0.05 vs. the control group; ^#P < 0.01 vs. cisplatin-only group (G).

Figure 7.

FGF21 suppresses P53 during cisplatin nephrotoxicity in vitro and in vivo. A, B) BUMPT cells were pretreated with or without recombinant FGF21 (0, 12.5, 25, 50, and 100 ng/ml) for 1 h and incubated with cisplatin (20 μM) for 24 h. Whole cell lysates were collected for immunoblot analysis of p-P53, P53, and β-actin (loading control). A) Representative immunoblots. B) Densitometry of p-P53 and P53 signals. The p-P53 and P53 signals were normalized to the β-actin signal of the same samples to determine the ratios. The ratios of the control group (in the absence of cisplatin and FGF21) were arbitrarily set as 1. Means ± sd (n = 3). *P < 0.01 vs. cisplatin-only treated cells, ^#P < 0.005 vs. cisplatin-only treated cells, *^#P < 0.05 vs. cisplatin plus FGF21 (50 ng/ml) treated cells. C, D) BUMPT stable cell lines were pretreated with or without recombinant FGF21 (50 ng/ml) for 1 h and cultured in the presence or absence of cisplatin (20 μM) for 24 h. Whole cell lysates were collected for immunoblot analysis of p-P53 and β-actin as protein loading control. C) Representative immunoblots. D) Densitometry of p-P53 signals. The p-P53 signals in C were normalized to the β-actin signal of the same samples to determine the ratios. The ratios of the NC control group (in the absence of cisplatin and FGF21) were arbitrarily set as 1. Means ± sd (n = 3). *P < 0.001 vs. cisplatin-only treated NC group cells, ^#P < 0.05 vs. cisplatin-only treated NC group cells. E, F) Recombinant FGF21 (0.1 μg/g) was administered intraperitoneally to the mice (male C57BL/6, 8–10 wk) every 8 h, with the first dose 1 h before cisplatin injection. A comparable volume of PBS was given to the no-FGF21 (recombinant FGF21 was dissolved in PBS) animals. Mice were euthanized at 72 h. Kidney tissues were collected for immunoblot analysis of P53, p-P53, and β-actin (loading control). E) Representative immunoblotting. F) Densitometry of p-P53 and P53 signals. The p-P53 and P53 signals were normalized to the β-actin signal of the same samples to determine the ratios. The ratios of control group in the absence of cisplatin and FGF21 were arbitrarily set as 1. Means ± sd (n = 3). *P < 0.0005 vs. cisplatin-only treated group, ^#P < 0.0001 vs. cisplatin-only treated group.

Figure 8.

Diagram depicting FGF21 in cisplatin-induced renal tubular apoptosis. Upon cisplatin treatment, P53 is phosphorylated and activated to induce tubular cell apoptosis. P53 also induces FGF21 expression. Upon induction, FGF21 represses the expression and phosphorylation of P53 to reduce renal tubular apoptosis.