Doxycycline attenuates cisplatin-induced acute kidney injury through pleiotropic effects

Abstract

Cisplatin (CDDP) is a widely-used chemotherapeutic drug for solid tumors, but its nephrotoxicity is a major dose-limiting factor. Doxycycline (Dox) is a tetracycline antibiotic that has been commonly used in a variety of infections. Dox has been shown to possess several other properties, including antitumor, anti-inflammatory, antioxidative, and matrix metalloproteinase (MMP)-inhibiting actions. We, therefore, investigated whether Dox exerts renoprotective effects in CDDP-induced acute kidney injury (AKI). Twelve-week-old male C57BL/6J mice were divided into the following groups: 1) control, 2) Dox (2 mg/ml in drinking water), 3) CDDP (25 mg/kg body weight, intraperitoneally), and 4) CDDP+Dox. After seven days of pretreatment with Dox, CDDP was administered and the animals were killed at day 1 or day 3. We evaluated renal function along with renal histological damage, inflammation, oxidative stress, and apoptosis. MMP and serine protease activities in the kidney tissues were assessed using zymography. Administration of CDDP exhibited renal dysfunction and caused histological damage predominantly in the proximal tubules. Dox did not affect either expression of CDDP transporters or the accumulation of CDDP in renal tissues; however, it significantly ameliorated renal dysfunction and histological changes together with reduced detrimental responses, such as oxidative stress and inflammation in the kidneys. Furthermore, Dox inhibited the activity of MMP-2 and MMP-9, as well as serine proteases in the kidney tissues. Finally, Dox markedly mitigated apoptosis in renal tubules. Thus, Dox ameliorated CDDP-induced AKI through its pleiotropic effects. Our results suggest that Dox may become a novel strategy for the prevention of CDDP-induced AKI in humans.

INTRODUCTION

Cisplatin (CDDP) is a platinum-containing anticancer drug that has been widely used for the treatment of solid tumors in many types of malignancies, including head, neck, lung, testis, ovary, and breast cancers. The use of CDDP, however, is often limited due to serious side effects, such as nephrotoxicity, neurotoxicity, and ototoxicity (31). In particular, nephrotoxicity is the major dose-limiting adverse effect and at times can result in discontinuation of the treatment. Although many studies have revealed that CDDP-induced nephrotoxicity is derived from oxidative stress, inflammation, DNA damage, and apoptosis in renal tubular cells (37), strategies to prevent or mitigate CDDP-induced acute kidney injury (AKI) have yet to be established clinically.

Doxycycline (Dox) is a tetracycline antibiotic that has been commonly used in a variety of infectious diseases. Its main antibiotic mechanism is the inhibition of protein synthesis via prevention of the amino-acyl tRNA from binding to the A site of ribosomes (5). Besides, it has been reported that tetracyclines exert antitumor effects in various cells derived from melanoma, leukemia, prostate, breast, and colorectal cancers (27, 29, 43). In particular, Dox inhibited tumor growth by inducing caspase activation and apoptosis in tumor cells (7, 14, 27, 29, 43), and coadministration of Dox with CDDP preclinically enhanced its cytotoxicity in ovarian cancer (55). Furthermore, Dox has been reported to have pleiotropic effects, including anti-inflammatory, antioxidative, and antiapoptotic actions, which are the pivotal mechanism in the pathogenesis of CDDP-nephrotoxicity, on nontumor cells (23, 42), indicating that coadministration with Dox could mitigate the renal tubular cell damages caused by CDDP. Dox is also known to inhibit matrix metalloproteinases (MMPs) by binding their enzymatic active sites or reducing their mRNA expression (12). A previous report showed that Dox attenuated ischemia reperfusion-induced AKI via the suppression of inflammatory cytokines, MMP-2, and apoptosis in renal tubular cells (13). These results, together with its known carcinostatic properties, suggest that Dox can be a promising strategy to mitigate CDDP-induced AKI. Minocycline, another tetracycline antibiotic, also has anti-inflammatory and antiapoptotic effects. Minocycline has been reported to attenuate apoptosis in cultured renal tubular cells treated with detrimental stimuli (51) and to exert protective effects against CDDP-induced ototoxicity in guinea pigs (24). However, no studies have been reported to elucidate whether tetracyclines could alleviate CDDP nephrotoxicity in vivo. Therefore, we conducted this study to investigate the renoprotective effects of Dox on CDDP-induced AKI and to clarify the underlying mechanisms of its actions in mice.

MATERIALS AND METHODS

Animal studies.

All animal procedures were conducted in accordance with the guidelines for care and use of laboratory animals approved by Kumamoto University (no. 29-115). Twelve-week-old male C57BL/6J mice were purchased from Charles River Laboratories Japan (Yokohama, Kanagawa, Japan) and were divided into the following four groups: 1) control (n = 4), 2) doxycycline (Dox; n = 4), 3) cisplatin (CDDP; n = 6), and 4) CDDP+Dox (n = 6). We used all animals of each group for all techniques. However, due to limited well numbers in SDS-PAGE, representative animals were finally used for figures of immunoblotting and zymography. Mice were housed in a room maintained at constant temperature, humidity, and light cycle (12:12-h light-dark) with free access to food and water. An individual mouse was kept in the metabolic cage to measure water intake and urine volume. Doxycycline (Sigma-Aldrich, St. Louis, MO) was administered at a concentration of 2 mg/ml in drinking water supplemented with 2% sucrose. To our knowledge, the present dose of Dox was the highest one used for mice orally (40). We applied this dose to maximize the effect of Dox. Sucrose was supplemented to mitigate the bitterness of Dox as used previously (40). Because the water ingestion was decreased for an initial few days after starting Dox (data not shown) and because CDDP-induced AKI could be exaggerated under the dehydrated state, we took seven days pretreatment with Dox to allow animals to acclimate Dox/sucrose water and recover from the dehydrated state. In addition, this duration of pretreatment was also applied for the previous study to demonstrate the protective effect of Dox on ischemic tissue injury (41). Cisplatin (Sigma-Aldrich) was dissolved in phosphate-buffered saline (PBS) at a concentration of 1.0 mg/ml and given with a single intraperitoneal injection of a toxic dose (25 mg/kg body wt) that had been conventionally used to induce AKI (16). Following seven days of pretreatment with Dox, CDDP was administered by a single intraperitoneal injection, and the animals were killed 24 or 72 h later. Dox treatment was continued until the animals were killed. Under anesthetic conditions, blood samples were collected from the inferior vena cava of the mice, and both kidneys were removed for analyses, as described below. The serum creatinine (Cr) and blood urea nitrogen (BUN) concentrations were measured by a commercial laboratory (SRL, Tokyo, Japan).

Histological assessment.

In periodic acid-Schiff (PAS) and immunohistological staining, the kidneys were fixed using 4% paraformaldehyde and embedded in paraffin. Kidney samples (3 µm thick) were stained with PAS for morphological analysis. Tubular injury score was calculated on a scale of 0–5 based on the percentage of tubules with necrosis, dilation, cast formation, or loss of brush border: 0, 0%; 1, 1–10%; 2, 11–25%; 3, 26–45%; 4, 46–75%; and 5, 76–100% (36). At least five randomly selected fields under the microscope were evaluated for each mouse. Immunohistochemical detection of F4/80-positive macrophage was performed for the histological clarification of renal inflammation (19). The number of F4/80-positive cells in the kidney cortex were counted in 10 random views from each slide.

Renal reactive oxygen species (ROS) production was evaluated using dihydroethidium (DHE) staining. A slice of kidney tissues was immediately frozen in tissue compound with liquid nitrogen under unfixed condition. Frozen sections (10 µm thick) were incubated with 10 µM DHE (Wako Pure Chemical, Osaka, Japan) in a light-protected and humidified box for 30 min at 37°C. Images were obtained using Olympus BX50 with BH2-RFL-T3 (Olympus, Tokyo, Japan). Quantification of fluorescence intensity was conducted using Image-J (National Institutes of Health, Bethesda, MD). Apoptosis in renal tissues was evaluated using unfixed renal section through terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays (In Situ Cell Death Detection Kit, Roche, Tokyo, Japan). TUNEL-positive cells were counted in the renal cortex. At least six areas were selected for each slide.

Platinum accumulation in the kidney.

Mice were divided into four groups as described above and killed 24 h after CDDP injection. Renal accumulation of platinum was evaluated as described previously (56). Briefly, kidney samples were minced with scissors and razor blade and incubated at 60°C for 1 h in 0.5% nitric acid (HNO₃). The elution extracted from the sample was diluted with HNO₃. To measure the amount of platinum, inductively-coupled plasma mass spectrometry (ICP-MS) was performed using a Thermo Fisher Scientific Element (Waltham, MA).

Real-time quantitative polymerase chain reaction.

A piece of kidney tissues was placed in an RNAlater (Sigma-Aldrich) at 4°C overnight. Total RNA was extracted using TRIzol (Life Technologies, Grand Island, NY) and RNeasy Micro Kit (Qiagen, Valencia, CA). TaqMan probes for mouse organic cation transporter 1 (OCT1), OCT2, copper transporter 1 (Ctr1), multidrug and toxin extrusion 1 (MATE1), NADPH oxidase 2 (Nox2), p47phox, Rac1, p67phox, TNF-α, IL-1β, IL-6, MCP-1, TGF-β, TNF receptor 1 (TNFR1), TNFR2, FasL, Fas, and GAPDH were purchased from Applied Biosystems (Foster City, CA), and probes for SOD1, SOD2, and SOD3 were purchased from Sigma-Aldrich. Real-time polymerase chain reaction (PCR) was performed using an ABI PRISM 7900 Sequence Detector System (Applied Biosystems). Quantitative analysis was performed using the ΔC_t value (C_{t gene of interest} − C_{t GAPDH}). Relative gene expression was obtained using the ΔΔC_t method (C_{t sample} − C_{t calibrator}).

Immunoblotting.

A slice of kidney tissues was homogenized in ice-cold T-PER solution (Thermo Fisher Scientific, Rockford, IL) with a protease inhibitor cocktail (Sigma-Aldrich). Aliquots of 30 mg of protein were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted with antibodies against phosphorylated NF-κB p65 (p-NF-κB p65), TNF-α, NOD-like receptor family pyrin domain containing 3 (NLRP3), cleaved caspase-1, cleaved IL-1β, caspase-3, cleaved caspase-3, Bax, Bcl-2, and β-actin (Cell Signaling Technology, Danvers, MA).

Gelatin zymography.

A slice of kidney tissues was homogenized in tissue protein extraction reagent (T-PER solution, Thermo Scientific) without protease inhibitors. Under the nonreducing condition, the enzyme activity of MMP-2 and MMP-9 in 30 µg of kidney protein was evaluated using the gelatin zymography kit (Cosmo, Tokyo, Japan) according to the manufacturer’s instructions. Gelatinolytic activity was visualized as clear areas in the stained gel.

Double-layer fluorescent zymography.

Serine protease activity was evaluated using double-layer fluorescent zymography, as described previously (20). Briefly, under the nonreducing condition, 40 µg of kidney protein were loaded onto 12% SDS-PAGE. Plasmin from human plasma (Sigma-Aldrich) was applied as the positive control. Following electrophoresis at 4°C, the gel was washed with 2.5% Triton X-100 for 30 min and incubated in 50 mM Tris HCl (pH 8.2). Then, a cellulose acetate membrane was soaked in 100 mM Tris HCl solution containing 0.2 mM N-t-Boc-Gln-Ala-Arg-7-amido-4-methylcoumarin (QAR-MCA) (Peptide Institute Inc., Osaka, Japan), a substrate for most serine proteases. The gel covered with the membrane was incubated at 37°C for 1 h, and the fluorescent proteolytic activity bands were visualized using an ultraviolet transilluminator.

Statistical analyses.

Data are expressed as means ± SE. Statistical analysis was performed using the two-tailed unpaired Student’s t-test or analysis of variance (ANOVA) followed by the Tukey method. P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad PRISM6 (GraphPad Software, La Jolla, CA).

RESULTS

Dox attenuated CDDP-induced renal dysfunction and tubular damage.

There was no difference in body weight, water ingestion, and urine volume between control and Dox groups (27.5 ± 0.8 g and 27.6 ± 0.5 g body wt, respectively), or between CDDP and CDDP + Dox groups (20.9 ± 0.3 g and 21.9 ± 1.1 g body wt, respectively) at 72 h following CDDP administration (Fig. 1, A–C). Twenty-four-hour urine volume of CDDP-treated mice could not be measured due to the small amount of urine and evaporation. The degree of body weight loss by CDDP was compatible to that in a previous report (35a). Dox did not have any effect on renal function and histology at basal conditions. CDDP injection significantly elevated BUN and serum Cr levels, which were both significantly suppressed by Dox (Fig. 2A). CDDP caused severe acute tubular injuries characterized by desquamation of the tubular epithelium, dilation of proximal tubules, and debris in the tubular lumen at 72 h in the cortex without glomerular injuries (Fig. 2, B and C). These pathological changes were markedly attenuated by treatment with Dox (Fig. 2, B and C). On the contrary, CDDP did not induce histological tubular damages except for tubular casts in the medulla (Fig. 2D).

Fig. 1.

Effects of doxycycline (Dox) and cisplatin (CDDP) on body weight and water balance. A: time course of body weight change after CDDP injection. B: water ingestion at 72 h after CDDP injection. C: urine volume at 72 h after CDDP injection. Urine volume of CDDP-treated mice could not be measured at 72 h due to the small amount of urine and evaporation. Control (Cont., n = 4), Dox-treated (n = 4), CDDP-injected (n = 6), and CDDP+Dox (n = 6). Values are means ± SE, *P < 0.05 vs. control.

Fig. 2.

Effects of doxycycline (Dox) on renal dysfunction and tubular damage induced by cisplatin (CDDP). A: blood urea nitrogen (BUN) and serum creatinine (Cr) levels at 72 h after CDDP injection were measured. Cont.; control, Dox; doxycycline-treated, CDDP; cisplatin-injected. B: renal cortex histopathology. Representative photomicrographs of periodic acid-Schiff (PAS)-stained kidney at 72 h after CDDP injection. C: semiquantitative analysis of tubular damage. D: renal medulla histopathology. Representative photomicrographs of PAS-stained kidney at 72 h after CDDP injection. Scale bars = 50 µm, control (n = 4), Dox-treated (n = 4), CDDP; cisplatin-injected (n = 6), and CDDP+Dox (n = 6). Values are means ± SE, *P < 0.05 vs. control, ^#P < 0.05 vs. CDDP.

Dox did not affect mRNA expression of CDDP transporters or inhibit CDDP accumulation.

Because Dox reduced CDDP-induced AKI, we investigated whether Dox could inhibit the accumulation of CDDP in the renal tissue. Mice were killed as early as at 24 h after CDDP injection when tubular structures were still preserved. In the proximal tubular cells OCT1, OCT2, and Ctr1 are known to be CDDP transporters expressed on the basolateral membrane, while MATE1 is found on the apical region. Dox did not alter mRNA expression of these transporters (OCT1, 1.02 ± 0.04-fold; OCT2, 1.04 ± 0.03-fold; Ctr1, 1.00 ± 0.02-fold; and MATE1, 1.05 ± 0.12-fold; nonsignificant vs. control; Fig. 3A). Furthermore, ICP-MS revealed that Dox had no effect on the renal CDDP content (Fig. 3B). These results suggest that the renoprotective effects of Dox were not derived from the change in renal CDDP accumulation, but rather from reduced detrimental reactions caused by CDDP.

Fig. 3.

Effects of doxycycline (Dox) on mRNA expression of cisplatin (CDDP) transporters and CDDP accumulation in the kidneys. A: gene expression of influx and efflux transporters of CDDP in the kidney. mRNA expression of organic cation transporter 1 (OCT1), OCT2, copper transporter 1 (Ctr1), multidrug and toxin extrusion 1 (MATE1), and GAPDH was determined by real-time PCR. The abundance of each mRNA was normalized to GAPDH, and data are expressed as fold increase over control. B: the amount of ¹⁹⁵Platinum (¹⁹⁵Pt) in the kidney tissue was measured using inductively coupled plasma-mass spectrometry. Control (Cont., n = 4), Dox-treated (n = 4). Values are means ± SE; n.s., nonsignificant.

Dox attenuated CDDP-induced renal ROS production.

Because the enhanced oxidative stress contributes to the development of CDDP-induced AKI, we explored the effects of Dox on ROS production and the gene expression of oxidative stress markers in the kidneys. CDDP increased renal ROS production as revealed by DHE fluorescence analysis, together with the elevated expression of NADPH oxidase components (Nox2, p47 phox, Rac1, and p67 phox) (Fig. 4, A–C). These changes were significantly alleviated with Dox administration. Among these genes, p47phox and Rac1 mRNAs were markedly increased in the CDDP group and strongly suppressed in the CDDP+Dox group (Fig. 4C). Alternatively, the gene expression of SOD2 and SOD3, which function as scavengers of superoxide anion, was reduced following CDDP injection, suggesting impaired antioxidant capability. The downregulation of SOD2 and SOD3 was significantly prevented by Dox treatment (Fig. 4C). These results indicate that Dox attenuated CDDP-induced AKI through its antioxidative effects.

Fig. 4.

Effects of doxycycline (Dox) on oxidative stress induced by cisplatin (CDDP) in the kidneys. A: dihydroethidium (DHE) staining of kidney sections. Frozen sections were incubated with 5 μM DHE. Images were obtained with an Olympus BX50 with BH2-RFL-T3 (Olympus, Tokyo, Japan). Scale bars = 50 µm. B: quantification of fluorescence intensity was performed. Values are summarized in the bar graph. C: gene expression of NADPH oxidase components and SODs. mRNA expression of Nox2, p47 phox, Rac1, p67 phox, SOD1, SOD2, SOD3, and GAPDH was determined by real-time PCR. Abundance of each mRNA was normalized for GAPDH. Data are expressed as fold increase over control. Control (Cont., n = 4); Dox-treated (n = 4); CDDP-injected (n = 6); and CDDP+Dox (n = 6). Values are means ± SE, *P < 0.05 vs. control, ^#P < 0.05 vs. CDDP.

Dox suppressed CDDP-induced renal inflammation.

In addition to ROS production, inflammation is also known to play a critical role in CDDP-induced AKI. Expressions of TNF-α, IL-1β, MCP-1, TGF-β, and IL-6 mRNAs were significantly increased by CDDP (TNF-α, 16.4 ± 0.6-fold; IL-1β, 11.7 ± 0.9-fold; MCP-1, 28.6 ± 1.7-fold; TGF-β, 5.4 ± 0.1-fold; and IL-6, 1,416 ± 211-fold; P < 0.05 vs. control; Fig. 5A); these inductions were all markedly suppressed by Dox treatment (TNF-α, 6.0 ± 1.7-fold; IL-1β, 3.8 ± 1.6-fold; MCP-1, 13.5 ± 5.1-fold; TGF-β, 2.3 ± 0.5-fold; and IL-6, 124.9 ± 54.5-fold from control; P < 0.05 vs. CDDP; Fig. 5A). Protein expressions of p-NF-κB p65 and TNF-α were increased by CDDP (p-NF-κB p65/β-actin, 4.8 ± 0.5-fold; TNF-α/β-actin, 1.8 ± 0.3-fold; P < 0.05 vs. control; Fig. 5, B and C) and markedly attenuated by Dox (p-NF-κB p65/β-actin, 1.3 ± 0.1-fold; TNF-α/β-actin, 0.9 ± 0.2-fold from control; P < 0.05 vs. CDDP; Fig. 5, B and C). Reflecting the upregulated inflammatory molecules, the F4/80-positive macrophages infiltrated into interstitium in the CDDP groups, which was substantially suppressed by Dox treatment (Fig. 5, D and E).

Fig. 5.

Effects of doxycycline (Dox) on inflammation and inflammasome induced by cisplatin (CDDP) in the kidneys. A: mRNA expression of TNF-α, IL-1β, MCP-1, TGF-β1, IL-6, and GAPDH was determined by real-time PCR. The abundance of each mRNA was normalized to GAPDH. B: protein levels of p-NF-κB p65 and TNF-α were evaluated by immunoblotting analysis. C: ratios of p-NF-κB p65 and TNF-α to β-actin were calculated. D: representative photomicrographs of immunofluorescence staining for F4/80. White arrows indicate F4/80-positive cells. Scale bars = 50 µm. E: quantitative analysis of F4/80-positive cells. HPF, high power field. F: protein levels of NLRP3, cleaved caspase-1, and cleaved IL-1β were evaluated by immunoblotting analysis. G: ratios of NLRP3, cleaved caspase-3, and cleaved IL-1β to β-actin were calculated. Control (Cont., n = 4); Dox-treated (n = 4); CDDP-injected (n = 6); and CDDP+Dox treatment (n = 6). Values are means ± SE, *P < 0.05 vs. control, ^#P < 0.05 vs. CDDP.

Since the inflammasome system plays an important role in inflammation, and NF-κB regulates the NLRP3 gene expression (2), we further evaluated the NLRP3 inflammasome in this model. Protein expressions of NLRP3, cleaved caspase-1, and cleaved IL-1β were significantly increased by CDDP (NLRP3/β-actin, 2.0 ± 0.2-fold; cleaved caspase-1/β-actin, 3.3 ± 0.7-fold; and cleaved IL-1β/β-actin, 8.9 ± 1.0-fold, P < 0.05 vs. control; Fig. 5, F and G) and were all significantly attenuated by Dox (NLRP3/β-actin, 1.1 ± 0.1-fold; cleaved caspase-1/β-actin, 1.7 ± 0.1-fold; and cleaved IL-1β/β-actin, 1.4 ± 0.2-fold from control; P < 0.05 vs. CDDP; Fig. 5, F and G). Our results demonstrated that the anti-inflammatory effects of Dox resulted in the amelioration of CDDP nephrotoxicity.

Dox inhibited MMP-2 and MMP-9 activated by CDDP in the kidneys.

Because Dox was reported to mitigate ischemia-reperfusion-induced AKI via the inhibition of the MMP activity, we investigated the pathogenic role of MMPs and the effects of Dox on this pathway in CDDP-induced AKI. Activities of MMP-2 and MMP-9, analyzed using gelatin gel zymography, were increased in the kidneys of the CDDP group, whereas they were suppressed in the CDDP+Dox group (Fig. 6A). mRNA expressions of MMP-2 and MMP-9, and their inhibitor TIMP-1, were significantly increased in the CDDP group (MMP-2, 1.7 ± 0.3-fold; MMP-9, 6.5 ± 0.3-fold; and TIMP-1, 290.1 ± 23.6-fold; P < 0.05 vs. control; Fig. 6B). Such upregulation of MMP-9 and TIMP-1 was markedly decreased in the CDDP+Dox group (MMP-9, 1.9 ± 0.5-fold; TIMP-1, 45.7 ± 18.2-fold from control; P < 0.05 vs. CDDP; Fig. 6B). Interestingly, treatment with Dox inhibited the conversion of pro-MMP-2 to its active form (Fig. 6A), without affecting its gene expression.

Fig. 6.

Effects of doxycycline (Dox) on matrix metalloproteinase (MMP) activity in the kidneys. A: enzyme activity of MMP-2 and MMP-9 in 30 μg of kidney protein was evaluated using the gelatin zymography. Gelatinolytic activity was visualized as clear areas in the stained gel. B: mRNA expression of MMP-2, MMP-9, TIMP-1, and GAPDH was determined by real-time PCR. The abundance of each mRNA was normalized to GAPDH. Data are expressed as fold increase over control (Cont.). Control (n = 4); Dox-treated (n = 4); CDDP-injected (n = 6); CDDP+Dox treatment (n = 6). Values are means ± SE, *P < 0.05 vs. control, ^#P < 0.05 vs. CDDP.

Dox inhibited serine proteases activated by CDDP in the kidneys.

Because MMPs are converted from pro-enzymes to their active forms by serine proteases such as plasmin (50), and because Dox largely blocked the conversion of MMP-2 without altering its gene expression, we examined the effect of Dox administration on the activity of serine proteases including plasmin (~80 kDa molecular wt) in the kidney tissues. Double-layer fluorescent zymography analysis revealed that the serine protease activity was substantially increased in the CDDP group, particularly at ~80 kDa. Furthermore, such activation was markedly suppressed in the CDDP+Dox group (Fig. 7). Collectively, these results indicate that Dox may have attenuated MMP-2 activation via the inhibition of the serine protease activity.

Fig. 7.

Effects of doxycycline (Dox) on serine protease activities in the kidneys. Double-layer fluorescent zymography using N-t-Boc-Gln-Ala-Arg-7-amido-4-methylcoumarin (QAR-MCA) substrate was performed to evaluate serine proteases activity in the kidney. Fluorescent proteolytic activity bands were visualized using an ultraviolet transilluminator. Plasmin from human plasma was loaded as the positive control. Control (Cont, n = 4); Dox-treated (n = 4); cisplatin (CDDP)-injected (n = 6); CDDP+Dox treatment (n = 6).

Dox inhibited CDDP-induced apoptosis in renal tubular cells.

As apoptosis is known to be an important mechanism of CDDP-induced nephropathy, we evaluated apoptosis of renal tubular cells. Apoptotic cells characterized by karyorrhexis and cell shrinkage were observed in the interstitial area in the CDDP group (Fig. 2B). The number of TUNEL-positive cells was increased in the CDDP group and reduced in the CDDP+Dox group (Fig. 8, A and B). Expressions of TNFR1, TNFR2, and Fas mRNAs were increased in the CDDP group (TNFR1, 7.3 ± 0.8-fold; TNFR2, 3.4 ± 0.1-fold; and Fas, 4.7 ± 0.2-fold; P < 0.05 vs. control; Fig. 8C) and were significantly decreased in the CDDP+Dox group (TNFR1, 3.2 ± 0.8-fold; TNFR2, 2.1 ± 0.4-fold; and Fas, 3.1 ± 0.7-fold from control; P < 0.05 vs. CDDP; Fig. 8C). Furthermore, the protein expression levels of Bax and cleaved caspase-3 were significantly increased by CDDP (Bax/β-actin, 1.5 ± 0.1-fold; cleaved caspase-3/β-actin, 3.3 ± 0.6-fold; P < 0.05 vs. control; Fig. 8, D and E), while Bcl-2 expression was decreased by CDDP (Bcl-2/β-actin, 0.3 ± 0.0-fold; P < 0.05 vs. control; Fig. 8, D and E). Accordingly, the ratio of Bax to Bcl-2 proteins was increased by CDDP (5.7 ± 0.9-fold; P < 0.05 vs. control; Fig. 8E). These changes were substantially alleviated by treatment with Dox (Bax/β-actin, 1.2 ± 0.1-fold; cleaved caspase-3/β-actin, 1.4 ± 0.2-fold; Bcl-2/β-actin, 0.5 ± 0.0-fold; and Bax/Bcl-2, 2.7 ± 0.1-fold from control; P < 0.05 vs. CDDP; Fig. 8, D and E). These results showed that Dox significantly suppressed renal tubular cell apoptosis induced by CDDP.

Fig. 8.

Effects of doxycycline (Dox) on apoptosis induced by cisplatin (CDDP) in the kidneys. A: representative image of TUNEL staining in the cortex. Apoptotic cells are visualized as green and nuclei are stained with DAPI (blue). Scale bars = 50 µm. B: quantitative analysis for the number of TUNEL-staining positive cells. C: mRNA expression of TNFR1, TNFR2, Fas, FasL, and GAPDH was determined by real-time PCR. The abundance of each mRNA was normalized to GAPDH. D: protein levels of Bax, Bcl-2, caspase-3, and cleaved caspase-3 were evaluated by immunoblotting analysis. E: ratios of Bax, Bcl-2, Bax, caspase-3, and cleaved caspase-3 to β-actin were calculated. Control (Cont., n = 4); Dox-treated (n = 4); CDDP-injected (n = 6); CDDP+Dox treatment (n = 6). Values are means ± SE, *P < 0.05 vs. control, ^#P < 0.05 vs. CDDP.

DISCUSSION

Although CDDP has been proven to be an effective therapy against many forms of cancer, nephrotoxicity remains to be a serious side effect with no established preventative measures; therefore, searching for strategy to prevent CDDP nephrotoxicity has been an active area of investigation. Meanwhile, Dox is shown to exhibit several pleiotropic effects, including antitumor, anti-inflammatory, antioxidative, and MMP-inhibiting actions (7, 12, 14, 23, 27, 29, 42, 43). The present study has addressed the question of whether Dox exerts renoprotective effects against it and has demonstrated that the treatment with Dox effectively attenuated CDDP-induced AKI in mice. The mechanisms of renoprotection could be multifaceted: i.e., reduction of oxidative stress, inflammation, and apoptosis, together with suppression of the activated MMPs and serine proteases. Although minocycline has been previously reported to mitigate CDDP-induced ototoxicity by suppressing auditory hair cell apoptosis (24), the present study is the first to demonstrate protective effects of the tetracycline antibiotic Dox on CDDP nephrotoxicity.

CDDP-induced renal injury is mainly confined to the S3 segment of the proximal tubules (3). In this region, CDDP is transported from the basolateral to the apical side in a transepithelial manner and excreted into the urine. Influx transporters, such as OCT1/OCT2/Ctr1, are expressed on the basolateral membrane, while the efflux transporter MATE1 is expressed on the apical side (34, 38, 57). Previous studies have shown that altered expression of these transporters can influence the severity of CDDP-induced AKI. Rodents fed a magnesium-deficient diet displayed severe CDDP nephrotoxicity via upregulation of OCT2 and enhanced accumulation of CDDP in the renal tissue, while oral supplementation with magnesium restored these deleterious changes (46, 56). Therefore, we first evaluated whether Dox could inhibit the accumulation of CDDP in the kidneys by altering expression of its transporters. However, we found that Dox had no effect on mRNA expression of CDDP transporters, and the renal CDDP content was not reduced. These results indicate that Dox could mitigate detrimental reactions directly in the proximal tubules caused by CDDP.

The increase in intracellular calcium by CDDP induces free radical production via NADPH oxidase activation, leading to nephrotoxicity (21, 28). In the present experiment, CDDP resulted in ROS production as reflected by DHE staining, accompanied by increased expression of NADPH oxidase subunits and decreased expression of SODs. We found that Dox substantially reduced ROS production through the suppression of p47phox and Rac1 and the restoration of SOD2 and SOD3. Previous reports have suggested that free radical scavengers or antioxidants (e.g., edaravone and heme oxygenase-1) also protect renal tissues against CDDP-induced nephrotoxicity (25, 45, 48, 53). Because the Dox chemical structure contains phenol rings that can support ROS scavenging (10), Dox might also directly eliminate oxidative stress, in addition to altering gene expression. Overall, the antioxidative pathway may be one important mechanism by which Dox operates to alleviate CDDP-induced AKI.

Inflammation is also recognized to be pivotal in the progression of CDDP-induced AKI (8, 26). One previous study showed that inhibition of TNF-α reduced expression of other cytokines, such as IL-1β and MCP-1, and alleviated CDDP nephrotoxicity (39). Since Dox is known to possess anti-inflammatory effects by reducing the production of TNF-α, IL-1β, IL-6, and MCP-1 (30, 47), we studied changes in inflammatory molecules with CDDP and Dox treatment. CDDP increased mRNA expression of TNF-α, IL-1β, IL-6, MCP-1, and TGF-β as well as the protein expression of TNF-α and p-NF-κB p65, a potent inducer of inflammatory cytokines. These changes were all markedly suppressed by treatment with Dox, indicating its anti-inflammatory effects in CDDP-induced AKI. Increasing evidence suggests that the inflammasomes, innate immune system sensors that activate caspase-1 and inflammatory cytokines, are involved in the progression of many diseases. In this study, the NLRP3 inflammasome system was activated in the CDDP group and was inhibited by Dox treatment. Collectively, Dox exerted prominent anti-inflammatory effects to lessen CDDP-induced renal injuries. Interestingly, minocycline, another tetracycline antibiotic, reportedly attenuates diabetic nephropathy by reducing the NLRP3 inflammasome (44). Therefore, tetracycline antibiotics may become promising agents to alleviate inflammation-related kidney diseases.

MMPs are endopeptidases that degrade the extracellular matrix and are associated with several types of kidney injuries such as ischemia-reperfusion AKI or the obstructive nephropathy model (6, 22, 52). It was demonstrated that Dox inhibited MMPs by binding directly to their proteolytic active sites or reducing their transcription (35). In the present study, CDDP enhanced the MMP-2 and MMP-9 enzymatic activity, together with their mRNA expression in the kidneys. Dox inhibited the activity of both members of MMPs, accompanied by reduced gene expression of MMP-9 but not MMP-2. Because serine proteases, such as plasmin, are essential for the conversion of pro-MMP to its active forms (50), we evaluated the serine protease activity in the kidneys. CDDP markedly increased the activity of different types of serine proteases, with the most robust form observed to be ~80 kDa. This could likely be identified as a plasmin. Since the administration of Dox suppressed the activity of these serine proteases, this effect would reduce MMP-2 activation independent of its gene expression. This hypothesis is supported by the fact that Dox inhibits plasmin activity in both a direct and indirect manner by inhibiting urokinase-type plasminogen activator, which converts plasminogen to active plasmin (4, 11). We recently reported that plasmin was also activated in the kidney of rats treated with aldosterone and a high-salt diet (18). We found that plasmin caused the induction of inflammatory and profibrotic cytokines, augmenting glomerular injury and tubulointerstitial fibrosis; serine protease inhibition by camostat mesylate effectively suppressed both plasmin activity and cytokine production and attenuated renal injuries (18). Therefore, the inhibitory effect of Dox on plasmin could contribute to the alleviation of CDDP nephrotoxicity. MMPs and serine proteases play important roles in cancer invasion (1, 9). Taken together, cotreatment with CDDP and Dox may be a potential chemotherapeutic strategy.

CDDP induces apoptosis in proximal tubular cells through both extrinsic and intrinsic pathways (37). In the extrinsic pathway, stimulation of death receptors, such as Fas and TNFR, leads to the activation of the caspase cascade, while in the intrinsic pathway detrimental stimuli can cause the translocation of Bax from the cytosol to the mitochondria, leading to the release of cytochrome c. Both pathways ultimately induce the activation of caspase-3, which prompts DNA fragmentation and cell death. Previous studies demonstrated that Bax-deficient mice were resistant to CDDP-induced AKI (54) and that overexpression of Bcl-2 to suppress Bax ameliorated cellular damage in cultured renal tubular cells treated with CDDP (17). In the present study, Dox suppressed the increased expression of TNFR1, TNFR2, and Fas, and Bax/Bcl-2 ratio, as well as caspase-3 activation and apoptosis induced by CDDP, suggesting that Dox effectively inhibited both the extrinsic and intrinsic pathways. Dox has been shown to exert antiapoptotic effects and alleviate cerebral injuries in a neonatal hypoxia-ischemia model (15). Interestingly, Dox can trigger apoptosis by activating the intrinsic pathway (Bax-caspase-3) in several types of tumor cells (7, 14, 33, 58). Although we cannot explain the mechanism underlying this discrepancy among nontumor and tumor cells, these differential effects of Dox may be favorable for CDDP-based chemotherapy.

In the present study, we applied the highest dose and longest pretreatment with Dox based on previous protocols used in mice (40, 41) to maximize the effect of Dox. Thus, the efficacy of lower doses or shorter pretreatment durations needs to be investigated further. In our preliminary study, in which Dox was initiated immediately following CDDP injection without pretreatment, no amelioration of nephrotoxicity was observed (data not shown). Therefore, we hypothesize that a certain period of pretreatment with Dox is necessary to achieve tissue concentrations sufficient to protect the kidneys from CDDP. Dox has been in clinical use for several decades without severe adverse effects, and it is typically administered for bacteremia and abscess for two to four weeks, even up to six months (32). Thus, the pretreatment with Dox employed in this study should be clinically applicable, and considering the antitumor properties of Dox, the effect of longer treatment might also be favorable.

In conclusion, Dox prevented CDDP-induced AKI through multiple effects, such as the suppression of oxidative stress, inflammation, the activity of MMPs and serine proteases, and finally, apoptosis in the kidney tissues. Our findings suggest that Dox may become a novel strategy for the prevention of CDDP-induced AKI in humans.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

T.N., Y.K., Y. Iwata, T.M., and N.S. performed experiments; T.N., Y. Iwata, Y.M., T.M., N.S., Y.N., H.J., H.S., and K.K. analyzed data; T.N. prepared figures; T.N. drafted manuscript; Y.K. and M.A. conceived and designed research; Y.K., Y.M., M.A., Y. Izumi, T.K., Y.N., H.J., H.S., K.K., and M.M. interpreted results of experiments; Y.K., Y. Izumi, T.K., and M.M. edited and revised manuscript; Y.K. and M.M. approved final version of manuscript.