Protective effect of the ethanol extract of the roots of Brassica rapa on cisplatin-induced nephrotoxicity in LLC-PK1 cells and rats

A study was conducted to determine whether the ethanol extract of the roots of Brassica rapa (EBR) ameliorates cisplatin-induced nephrotoxicity in terms of oxidative stress, as characterized by lipid peroxidation, reactive oxygen species (ROS) production, and glutathione (GSH) depletion in LLC-PK1 cells. Pretreatment of cells with EBR prevented cisplatin-induced decreases in cell viability and cellular GSH content. The effect of EBR was then investigated in rats given EBR for 14 d before cisplatin administration. A single dose of cisplatin (7 mg/kg, i.p.) caused kidney damage manifested by an elevation in blood urea nitrogen (BUN), serum creatinine, and urine lactate dehydrogenase (LDH) levels. Also, renal tissue from cisplatin-treated rats showed a significant increase in malondialdehyde (MDA) production, and in the activities of aldehyde oxidase (AO) and xanthine oxidase (XO). Moreover, a significant decrease in the activities of antioxidant enzymes, such as, glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT) was observed in cisplatin-treated rats versus saline-treated normal group. In contrast, rats given EBR showed lower blood levels of BUN and creatinine, and of urinary LDH. Moreover, EBR prevented the rise of MDA production and the induction of AO and XO activities. This extract also recovered the reduced activities of GPx, SOD and CAT. Taken together, our data indicate that the ethanol extract of the roots of Brassica rapa (EBR) has a protective effect against cisplatin-induced nephrotoxicity because it attenuates oxidative stress.     

Dopamine synthesis and release in LLC-PK1 cells

A porcine-derived renal epithelial cell line, LLC-PK1, was used to examine dopamine (DA) synthesis from L-dopa and DA release into the media. DA synthesis and release were elevated by increasing either L-dopa concentration or incubation time. DA synthesis was shown to be entirely due to conversion of L-dopa to DA by aromatic amino acid decarboxylase. DA release from LLC-PK1 cells was stimulated by NaCl and other Na+ or Cl- salts. DA release by LLC-PK1 cells was not dependent on extracellular Ca2+ or significantly stimulated by the depolarizing actions of KCl. LLC-PK1 cells which are devoid of any neural contributions to renal DA production can synthesize DA from L-dopa and release DA in response to stimulation by either Na+ or Cl-.     

Pharmacological antagonism of fumonisin B1 cytotoxicity in porcine renal epithelial cells (LLC-PK1): a model for reducing fumonisin-induced nephrotoxicity in vivo

Fumonisin B1 is a mycotoxin commonly found on corn. It is hepatotoxic and nephrotoxic in domestic and experimental animals, and causes equine leukoencephalomalacia and porcine pulmonary oedema. It is a potent inhibitor of ceramide synthase. Inhibition leads to accumulation of free sphingoid bases in cells and tissues. In pig kidney epithelial cells (LLC-PK1), fumonisin B1 induces increased tumour necrosis factor alpha (TNFalpha) expression independent of the accumulation of sphingoid bases. The objective of this study was to investigate pharmacological approaches for intervening in fumonisin B1 toxicity using the LLC-PK1 cell model. The toxicity of fumonisin B1 was assayed using cell viability and lactate dehydrogenase (lactate dehydrogenase) release. Pretreatment of cells with myriocin, preventing sphinganine accumulates, prevented the fumonisin B1-induced decrease in cell viability and increased lactate dehydrogenase release. Modulation of adenosine receptor activity did not reduce the fumonisin B1 cytotoxicity. As with myriocin, silymarin pretreatment prevented the fumonisin B1-induced effects on cell viability and lactate dehydrogenase release. When added 6 or 24 hr after treatment of cells with fumonisin B1, both myriocin and silymarin reversed the decreased cell viability and suppressed the increased lactate dehydrogenase release. Myriocin, but not silymarin, blocked the accumulation of sphinganine in fumonisin B1-treated cells. Silymarin, unlike myriocin, induced expression of TNFalpha to an extent similar to fumonisin B1, but pretreatment with silymarin decreased the fumonisin B1-induced TNFalpha expression in LLC-PK1 cells. Results suggest that the mechanisms by which myriocin and silymarin protect renal cells are different, and silymarin potentially prevents fumonisin B1-induced toxicity by modulating TNFalpha expression or signals downstream of the inhibition of ceramide synthase.     

Influence of uranium(VI) speciation for the evaluation of in vitro uranium cytotoxicity on LLC-PK1 cells

Very few data are available concerning the in vitro toxicity of uranium. In this work, we have determined the experimental chemical conditions permitting the observation of uranium(VI) cytotoxicity on LLC-PK1 cells. Uranium solutions made either by dissolving uranyl acetate or nitrate crystals, or by complexing uranium with bicarbonate, phosphate or citrate ligands, were prepared and tested. Experiments demonstrated that only uranium solutions containing citrate and bicarbonate ligands concentrations tenfold higher than the metal, were soluble in the cell culture medium. Cytotoxicity studies of all these uranium compounds were performed on LLC-PK1 cells and compared using LDH release, neutral red uptake and MTT assays. Dose dependent cytotoxicity curves were only obtained with uranium-bicarbonate medium. This study has revealed a toxicity of uranium-bicarbonate complexes for 24 h expositions and for concentrations ranging from 7 x 10(-4)-10(-3) M, under these conditions, the CI50 (cytotoxicity index) was evaluated between 8.5 and 9 x 10(-4) M. In contrast, we noticed a lack of cytotoxicity response for uranium(VI)-citrate complexes. Electron transmission microscopy studies revealed, when LLC-PK1 cells were exposed to the uranium-bicarbonate system, that uranium penetrated and precipitated within the cytoplasmic compartment. Morphological studies conducted with citrate complexes did not show any cellular intake of uranium.     

Influence of cadmium speciation for the evaluation of in vitro cadmium toxicity on LLC-PK(1) cells.

The main objective of the present work was to assess the potentiality of in vitro models to improve our understanding of cadmium-induced toxicity, especially on epithelial renal cells. Indeed cadmium, a potent toxic metal, poses a serious environmental threat and the mechanisms of its renal toxicity need to be clarified. Cytotoxicity studies presented here were performed in a tubular proximal original established porcine kidney cell line (LLC-PK(1)). We have compared cytotoxicity induced by different chemical cadmium forms in LLC-PK(1) cells as a function of media cell culture pH and protein content. Cadmium stock solutions were prepared either by dissolving cadmium chloride or cadmium sulphate with increasing protein concentrations in the media cell culture. Its pH was monitored during experiments. Cytotoxicity was measured by neutral red uptake after 24 h of exposure. Dose-dependent cytotoxicity curves, calculated with REGTOX, were systematically correlated with pH and protein content. Experiments in vitro revealed that cadmium was dose-dependently toxic for LLC-PK(1) for concentrations ranging from 10(-4) to 10(-6) M. We have noticed a lack of influence of the media cell culture pH on the cadmium cytotoxicity. REGTOX determines closely the EC(50) values but EC(50)CdCl(2)>EC(50)CdSO(4) and cadmium have been assayed with an inductively coupled atomic emission spectrometer (ICP/AES) directly in the media cell culture and the cellular pellet.     

Glucuronidation and the transport of the glucuronide metabolites in LLC-PK1 cells

Formation and transport of glucuronide metabolites were studied in LLC-PK1 cells. Glucuronidation of 17beta-estradiol, 1-naphthol, mycophenolic acid, and 4-methylumbelliferone was examined in microsomes prepared from LLC-PK1 cells, human livers, human kidneys, and human intestines. The rate of glucuronide metabolite formation observed with LLC-PK1 microsomes was comparable to rates observed with various human tissue microsomes. The fate of the glucuronide metabolite formed in the LLC-PK1 cells was studied by examining its extracellular transport using mycophenolic acid as a model substrate. After administration of mycophenolic acid, the amount of the glucuronide metabolite exiting to the extracellular compartments significantly decreased in the presence of MK-571, an inhibitor for the multidrug resistance-associated protein (MRP) transporter. However, the intracellular levels of the glucuronide metabolite did not change, suggesting that MK-571 was probably blocking metabolite efflux. In summary, these results suggest that the glucuronidating enzyme(s) expressed in the LLC-PK1 cells are capable of sufficient glucuronidation activity and that endogenous transporter(s) in LLC-PK1 cells are active and determine the distribution of the formed metabolites. Since these cells have been previously used to study drug transport, they may be a useful tool in future studies to explore the effect of drug transporters on glucuronidation.     

Sensitive endpoints for evaluating cadmium-induced acute toxicity in LLC-PK1 cells

Cadmium chloride (CdCl(2)) is a nephrotoxicant that causes damage to the proximal tubular epithelium. In vivo, it increases the permeability of epithelial surfaces, while in vitro, it acts on active trans-epithelial ion transport. The purpose of this study was to investigate CdCl(2) effects on a porcine renal proximal tubular epithelial cell line (LLC-PK1), and, in particular, to identify sensitive endpoints revealing damage both at the epithelial barrier level and at the molecular level. After exposure of the cells to CdCl(2), trans-epithelial resistance (TER) decreased while paracellular permeability (PCP) increased, indicating a structural alteration of the junctional complex. At the molecular level, we observed an increase in protective proteins, such as metallothioneins (MTs) and heat shock proteins (HSP70), starting from 25 microM CdCl(2), together with alterations in cytoskeleton organization. Production of reactive oxygen species (ROS) was also evident, indicating cellular oxidative stress. Our data indicate that CdCl(2) toxicity can be detected at the barrier level and at the molecular level at low concentrations, at which cytotoxicity assays are unable to show any damage. Therefore, these endpoints should prove very useful in studying heavy metal-induced acute toxicity. Exposure of the cells to higher concentrations of CdCl(2) (50 microM) revealed the initiation of apoptosis, mediated by caspase-3.     

Disruption of the endoplasmic reticulum by cytotoxins in LLC-PK1 cells

Prior induction of an endoplasmic reticulum stress response results in protection against reactive cytotoxins in the LLC-PK1 cell line. The purpose of this investigation was to determine therefore if the endoplasmic reticulum was disrupted by iodoacetamide, tert-butylhydroperoxide or sulfamethoxazole hydroxylamine. Toxic concentrations of the three toxins caused a dramatic loss of GRP94 protein within 3-8h of exposure, while induction of GRP78 and calreticulin occurred at 8 and 24h following exposure. There was no evidence of cytosolic elevation of calcium and neither dantrolene nor xestospongin were able to block the cytotoxicity of IDAM and TBHP. Exposure to the toxins led to DNA degradation and cleavage of procaspase-12. There was only evidence of procaspase-3 cleavage after TBHP exposure. These results demonstrate that the ER is disrupted by the reactive cytotoxins examined in LLC-PK1cells and suggest that the cytoprotection against low to moderate concentrations of cytotoxins observed following endoplasmic reticulum stress protein induction is likely due to a mechanism other than maintenance of calcium homeostasis.     

Expression and activities of several drug-metabolizing enzymes in LLC-PK1 cells

LLC-PK1 cells are frequently used in toxicology research, but little information is available concerning the capacity of these cells to metabolize xenobiotics. We examined the expression and activities of cytochromes P450 (P450) 1A1/1A2 (CYP 1A1/1A2), 2E1 (CYP 2E1), flavin monooxygenase (FMO), 5-lipoxygenase (5-LO) and prostaglandin H synthase (PHS)-associated cyclooxygenase-1 (COX-1). We prepared S9 fractions from LLC-PK1 cells, rat liver, and rat kidney, and measured enzyme activities using ethoxyresorufin O-deethylation (EROD) for CYP 1A1/1A2 and ethoxycoumarin O-deethylation (ECOD) for CYP 2E1, benzydamine N-oxidation (BNO) for FMO, leukotriene B₄ (LTB₄) formation for 5-LO, and thromboxane B₂ (TXB₂) formation for COX-1 activities. To assure that product formation was due to enzymatic activity, we used the following inhibitors: 1-aminobenzotriazole (ABT) for P450, methimazole for FMO, caffeic acid for 5-LO and acetylsalicylic acid (ASA) for COX-1. We also performed Western blot analysis to confirm our observations. All five enzyme activities were demonstrable in rat liver at much greater levels than in rat kidney S9 fractions. Activities in LLC-PK1 cells were significantly lower than activities in rat liver S9 fraction and generally less than activities in rat kidney S9 fraction. Enzyme inhibitors decreased product formation in all three tissues and Western blot analysis supported our observations of low enzyme activity in LLC-PK1 cells. These results indicate that LLC-PK1 cells have very low content of relevant drug-metabolizing enzyme activities.     

Up-regulation of MDR1 function and expression by cisplatin in LLC-PK1 cells

To examine whether cisplatin affects the multidrug transporter MDR1/P-glycoprotein in the kidneys, the effects of cisplatin on cell sensitivity to an anticancer drug, MDR1 function and expression were examined by assessing the growth inhibition by the MDR1 substrate paclitaxel, the uptake and efflux of the MDR1 substrate Rhodamine123 and the level of MDR1 mRNA, respectively. Porcine kidney epithelial LLC-PK1 cells were used, as they have a structure and function similar to those of renal proximal tubular cells and physiologically express low levels of MDR1. The growth inhibitory curve of LLC-PK1 cells by paclitaxel was shifted to a higher concentration range by pretreatment with 1 micro M cisplatin for 48 h. The uptake and efflux of Rhodamine123 were significantly reduced and enhanced, respectively, by pretreatment with 1 micro M cisplatin for 48 h. This enhanced efflux was suppressed by the representative MDR1 substrate/inhibitor ciclosporin. The expression of MDR1 mRNA was increased by the existence of cisplatin for 48 h. These observations taken together suggested that the transient exposure to cisplatin could cause the up-regulation of MDR1 in LLC-PK1 cells.     

Modulation of membrane cholesterol content in cultured LLC-PK1 cells

This study was designed to develop a cell culture model to evaluate the effect of cell membrane fluidity on the function of drug carrier proteins. Fluidity of the cell membrane has been linked to its cholesterol content. LLC-PK₁ cells were grown as a monolayer and were treated with a BSA-PVP-cholesterol complex in time-dependent studies. The total cholesterol content of the cells was increased 4-fold (from 52 to 200 fmol/mg protein) as a result of this treatment. The cholesterol increase appears to level off at 2 h of treatment. Most of the incorporated cholesterol (77%) was determined to be free cholesterol, the rest being esterified. Membrane fractions of the treated cells were isolated and were determined to carry 88% of the cholesterol content. The studies indicate that it is feasible to deliver cholesterol to cultured epithelial cells and that LLC-PK₁ cell monolayers can be used as a model for such studies.     

Vitamin E protects against the mitochondrial damage caused by cyclosporin A in LLC-PK1 cells

Cyclosporin A (CsA) has nephrotoxic effects known to involve reactive oxygen species (ROS), since antioxidants prevent the kidney damage induced by this drug. Given that mitochondria are among the main sources of intracellular ROS, the aims of our study were to examine the mitochondrial effects of CsA in the porcine renal endothelial cell line LLC-PK1 and the influence of the antioxidant Vitamin E (Vit E).Following the treatment of LLC-PK1 cells with CsA, we assessed the mitochondrial synthesis of superoxide anion, permeability transition pore opening, mitochondrial membrane potential, cardiolipin peroxidation, cytochrome c release and cellular apoptosis, using flow cytometry and confocal microscopy procedures. Similar experiments were done after Vit E preincubation of cells.CsA treatment increased superoxide anion in a dose-dependent way. CsA opened the permeability transition pores, caused Bax migration to mitochondria, and decreased mitochondrial membrane potential and cardiolipin content. Also CsA released cytochrome c into cytosol and provoked cellular apoptosis. Vit E pretreatment inhibited the effects that CsA induced on mitochondrial structure and function in LLC-PK1 cells and avoided apoptosis.CsA modifies mitochondrial LLC-PK1 cell physiology with loss of negative electrochemical gradient across the inner mitochondrial membrane and increased lipid peroxidation. These features are related to apoptosis and can explain the cellular damage that CsA induces. As Vit E inhibited these effects, our results suggest that they were mediated by an increase in ROS production by mitochondria.     

Effect of antioxidants on para-aminophenol-induced toxicity in LLC-PK1 cells

The present studies were designed to investigate the susceptibility of LLC-PK1 cells to cytotoxicity induced by para-aminophenol (PAP) and the ability of antioxidants to prevent PAP-induced cytotoxicity. LLC-PK1 cells were incubated for 4 h with varying concentrations of PAP (0-0.2 mM). Incubation was continued for 20 h and viability was monitored at 24 h after initial exposure to PAP. For coincubation experiments, cells were incubated for 4 h with various antioxidants [including ascorbate, glutathione (GSH), butylated hydroxytoluene (BHT), beta-nicotinamide adenine dinucleotide (NADH), or beta-nicotinamide adenine dinucleotide phosphate (NADPH)] in the absence or presence of 0.1 mM PAP. For preincubation experiments, cells were incubated for 1 h with ascorbate, GSH or NADPH. Antioxidants were removed and cells were exposed to 0 or 0.1 mM PAP for 4 h. Viability was determined 24 h following PAP exposure. LLC-PK1 cells displayed a steep concentration-response relationship for PAP; 0.1 mM PAP caused approximately 50% loss of viability. Coincubation with ascorbate, GSH and NADPH was without effect on cell viability in the absence of PAP and attenuated PAP-induced losses in viability. In contrast, NADH was ineffective in preventing PAP-induced cytotoxicity. BHT alone produced a significant loss of cell viability and was ineffective in preventing PAP cytotoxicity. Inability of NADH to prevent PAP-induced cytotoxicity was related to rapid degradation of NADH in aqueous solution. Preincubation of cells with ascorbate or GSH but not NADPH was associated with attenuation of PAP-induced cytotoxicity. These data suggest that (1) PAP is cytotoxic to LLC-PK1 cells, (2) a portion of PAP cytotoxicity is due to nonenzymatic oxidation that occurs in the incubation medium, and (3) a portion of PAP cytotoxicity is due to enzymatic or nonenzymatic oxidation that occurs within cells.     

Cephaloridine-induced inhibition of cytochrome c oxidase activity in the mitochondria of cultured renal epithelial cells (LLC-PK(1)) as a possible mechanism of its nephrotoxicity

To clarify the mechanism of cephalosporin nephrotoxicity, the effects of cephaloridine (CLD), a nephrotoxic cephalosporin antibiotic, on the mitochondria of the pig kidney proximal tubular epithelial cell line LLC-PK(1) were studied in culture. The activity of cytochrome c oxidase in the mitochondria of LLC-PK(1) cells was significantly decreased from 9 h after addition of 1.0 mM CLD to the cultured cells. These effects were dose-dependent and accompanied with a significant decrease in the ATP content in the cells, followed by marked morphological changes in the mitochondria. These alterations were observed in the treated cells before the increase in lipid peroxidation. The activities of NADH-cytochrome c reductase and succinate dehydrogenase in the mitochondria and NADPH-cytochrome P450 reductase, NADH-cytochrome b(5) reductase, and 7-ethoxycoumarin O-deethylase in the microsomes of the treated cells were not affected. Superoxide anion production by the mitochondria prepared from LLC-PK(1) cells or NADH-cytochrome c reductase was not affected by addition of CLD (1-10 mM), but adriamycin (0.1 mM) or paraquat (0.1 mM) significantly increased the superoxide anion production. These results suggested that the primary action of CLD is inhibition of cytochrome c oxidase activity in the mitochondrial electron transport chain, which decreases intracellular ATP content in renal tubular epithelial cells and that these effects of CLD are followed by increased lipid peroxidation and cellular injury.     

Gentamicin-induced apoptosis in LLC-PK1 cells: involvement of lysosomes and mitochondria

Gentamicin accumulates in lysosomes and induces apoptosis in kidney proximal tubules and renal cell lines. Using LLC-PK1 cells, we have examined the concentration- and time-dependency of the effects exerted by gentamicin (1-3 mM; 0-3 days) on (i) lysosomal stability; (ii) activation of mitochondrial pathway; (iii) occurrence of apoptosis (concentrations larger than 3 mM caused extensive necrosis as assessed by the measurement of lactate dehydrogenase release). Within 2 h, gentamicin induced a partial relocalization [from lysosomes to cytosol] of the weak organic base acridine orange. We thereafter observed (a) a loss of mitochondrial membrane potential (as from 10 h, based on spectrophotometric and confocal microscopy using JC1 probe) and (b) the release of cytochrome c from granules to cytosol, and the activation of caspase-9 (as from 12 h; evidenced by Western blot analysis). Increase in caspase-3 activity (assayed with Ac-DEVD-AFC in the presence of z-VAD-fmk]) and appearance of fragmented nuclei (DAPI staining) was then detected as from 16 to 24 h together with nuclear fragmentation. Gentamicin produces a fast (within 4 h) release of calcein from negatively-charged liposomes at pH 5.4, which was slowed down by raising the pH to 7.4, or when phosphatidylinositol was replaced by cardiolipin (to mimic the inner mitochondrial membrane). The present data provide temporal evidence that gentamicin causes apoptosis in LLC-PK1 with successive alteration of the permeability of lysosomes, triggering of the mitochondrial pathway, and activation of caspase-3.     

Effects of lipoproteins on cyclosporine A toxicity and uptake in LLC-PK1 pig kidney cells.

Cyclosporine A (CSA) is an effective immunosuppressant, but side effects such as renal toxicity can limit its therapeutic use. The current studies investigate the effects of lipoproteins on CSA-induced renal toxicity in the pig epithelial cell line LLC-PK(1). Protein synthesis and tritiated CSA were used as measures of toxicity and uptake of CSA, respectively, in the LLC-PK(1) cell line. The three main classes of lipoproteins, very low (VLDL), low (LDL), and high density lipoproteins (HDL) at hypo-, normo-, and hyperlipidemic levels were tested for their ability to affect CSA-induced toxicity and uptake. The major component of each lipoprotein was also tested to determine its effects on CSA-induced toxicity and uptake. ApoA-I, the major protein component of HDL, and intact LDL particles showed the most significant effects of CSA uptake and toxicity. The uptake and toxicity of CSA was effectively reduced with elevated LDL concentrations but showed a significant increase (p < 0.05) when incubated with elevated concentrations of apoA-I. Increasing VLDL and HDL concentrations slightly reduced CSA toxicity and uptake, but showed little effect with increased incubation time. Triglyceride and cholesterol, the respective major components of VLDL and LDL, did not alter CSA uptake or toxicity under the conditions tested. LDL and apoA-I are identified as the major effectors of CSA toxicity and uptake in LLC-PK(1) cells. These effects may be mediated through receptors such as the LDL receptor or those involved in protein reabsorption. The data presented here clearly demonstrate a relationship between CSA-induced toxicity and the nature of the associated lipoprotein.     

LLC-PK1 cells as a model for renal toxicity caused by arsine exposure

The mechanisms of arsine (AsH3) toxicity are not completely understood. Studies were undertaken to determine AsH3 and arsenite [As(III)] toxicity in a renal tubular epithelial cell line to model kidney dysfunction caused by AsH3 exposure. The hypothesis was that As(III) is the toxic metabolite responsible for the renal toxicity of AsH3. There was a concentration- and time-dependent toxic response after As(III) incubation. As(III) produced significant LDH leakage as early as 1 h and intracellular potassium loss at 5 h. AsH3 produced no changes in these parameters. AsH3 affected neither potassium nor LDH levels over 24 h and up to 1 mM AsH3 concentration. In this system, As(III) induced LDH leakage before K+ loss. Oxidative stress-like toxicity effects were also studied by determining levels of glutathione (GSH), glutathione disulfide (GSSG), and heat-shock protein 32 (Hsp32) levels. GSH levels were not markedly affected by any arsenical over a 6-h period or up to 100 microM concentration of the arsenical. However, 100 microM AsH3 significantly increased GSSG levels as early as 30 min and reached a maximum at 2.5 h. Incubation with 10 microM AsH3 was sufficient to significantly increase GSSG levels. As(III) had no marked effect on GSSG. Both arsenicals (50 microM) produced a slight increase (about threefold) in Hsp32 levels after 4-h incubation. These results showed that unchanged AsH3 produced oxidative stress-like toxic effects without producing cell death. However, similar As(III) concentrations induced the stress response and were toxic to the cells. These data indicated that AsH3 is not directly toxic to LLC-PK1 cells.     

Role of P-glycoprotein in accumulation and cytotoxicity of amrubicin and amrubicinol in MDR1 gene-transfected LLC-PK1 cells and human A549 lung adenocarcinoma cells

Amrubicin is a completely synthetic 9-aminoanthracycline agent for the treatment of lung cancer in Japan. The cytotoxicity of C-13 hydroxy metabolite, amrubicinol, is 10 to 100 times greater than that of amrubicin. The transporters responsible for the intracellular pharmacokinetics of amrubicin and amrubicinol remains unclear. This study was aimed to determine whether P-glycoprotein (P-gp) plays functional and preventive role in cellular accumulation and cytotoxicity of amrubicin and its active metabolite amrubicinol by in vitro transport and toxicity experiments. Cytotoxicity and intracellular accumulation of amrubicin and amrubicinol were evaluated by LLC-PK1 cells, MDR1 gene-transfected LLC-PK1 (L-MDR1) cells overexpressing P-gp, and human A549 lung adenocarcinoma cells. L-MDR1 cells showed 6- and 12-fold greater resistance to amrubicin and amrubicinol, respectively, than the parental LLC-PK1 cells. The intracellular accumulation of both drugs in L-MDR1 cells was significantly reduced compared to the LLC-PK1 cells. The basal-to-apical transepithelial transport of both drugs markedly exceeded, whereas the apical-to-basal transport of both drugs was significantly lower in L-MDR1 cells than LLC-PK1 cells. Cyclosporin A (CyA) restored the sensitivity, intracellular accumulation and transport activity for both drugs in L-MDR1 cells. In A549 cells, CyA significantly increased the intracellular accumulation and cytotoxicity of both drugs. These findings indicated that P-gp is responsible for cellular accumulation and cytotoxicity of both amrubicin and amrubicinol, therefore suggesting that the antitumor effect of amrubicin could be affected by the expression level of P-gp in lung cancer cells in chemotherapeutic treatments.