Metabolism of Cisplatin to a nephrotoxin in proximal tubule cells

Abstract

Cisplatin, a commonly used chemotherapeutic agent, is nephrotoxic. The mechanism by which cisplatin selectively kills the proximal tubule cells was heretofore unknown. Recent studies in mice and rats have shown that the nephrotoxicity of cisplatin can be blocked by acivicin or (aminooxy)acetic acid, the same enzyme inhibitors that block the metabolic activation of a series of nephrotoxic halogenated alkenes. In this study, it was hypothesized that cisplatin is activated in the kidney to a toxic metabolite through the same pathway that has been shown to activate the halogenated alkenes. This activation begins with the formation of a glutathione-conjugate that is metabolized to a cysteinyl-glycine-conjugate, to a cysteine-conjugate, and finally to a reactive thiol. In this study, a protocol was developed in which confluent monolayers of LLC-PK(1) cells were exposed to clinically relevant concentrations of cisplatin or cisplatin-conjugate for 3 h. Cell viability was assayed at 72 h. The role of gamma-glutamyl transpeptidase (GGT) and cysteine-S-conjugate beta-lyase in the metabolism of each of the cisplatin-conjugates was investigated. Pre-incubation of cisplatin with glutathione, cysteinyl-glycine, or N-acetyl-cysteine to allow for the spontaneous formation of cisplatin-conjugates increased the toxicity of cisplatin toward LLC-PK(1) cells. Inhibition of GGT activity showed that GGT was necessary only for the toxicity of the cisplatin-glutathione-conjugate. Inhibition of cysteine-S-conjugate beta-lyase reduced the toxicity of each of the cisplatin-conjugates. These data demonstrate that metabolism of cisplatin in proximal tubule cells is required for its nephrotoxicity. The elucidation of this pathway provides new targets for the inhibition of cisplatin nephrotoxicity.

Figure 1.

Metabolic activation of glutathione-conjugates to reactive thiols. Halogenated alkenes (X represents the alkene and Y a halogen molecule: fluorine, chlorine, or bromine) have been shown to be metabolized to nephrotoxins via this pathway (9,12,60). The haloge-nated alkenes form glutathione-S-conjugates, which are cleaved to cysteinyl-glycine-conjugates by GGT. Aminodipeptidase cleaves the cysteinyl-glycine-conjugates to cysteine-conjugates. Cysteine-S-con-jugate beta-lyase catalyzes the production of unstable reactive thiols, which are toxic. We propose that cisplatin is metabolized through this same pathway. In the proposed pathway, Y would represent one of the chlorines in cisplatin and X would represent the remainder of the cisplatin molecule. In the proposed pathway, the sulfur of the gluta-thione molecule binds to the platinum displacing the chlorine.

Figure 2.

Survival of confluent monolayers of LLC-PK₁ cells after continous exposure to cisplatin (A) or a 3 h pulse (B). (A) Confluent monolayers of LLC-PK₁ cells were incubated in DMEM containing 50 mM cisplatin (diamonds) or 100 mM cisplatin (squares). At each time point, cell survival was determined relative to untreated cells. Each point represents the mean of three points ± SD. (B) Confluent monolayers of LLC-PK₁ cells were incubated for 3 h in a balanced salt solution containing 50 mM cisplatin (diamonds), 100 mM cisplatin (squares), or 150 mM cisplatin (triangles). The cisplatin was removed at the end of the 3 h exposure and fresh medium added to the cells. At each time point, cell survival was determined relative to the number of viable cells in control wells. Each point represents the average of three means SEM.

Figure 3.

Toxicity of a 3 h treatment with cisplatin and cisplatin-conjugates toward confluent monolayers of LLC-PK₁ cells. LLC-PK₁ cells were exposed for 3 h to 50 mM cisplatin or 50 mM cisplatin preincubated with equimolar glutathione, cysteinyl-glycine, or N-acetyl-cysteine. The cisplatin solutions were removed at the end of the 3 h exposure, and fresh medium was added to the cells. At 72 h, the number of cells killed was determined as a percentage of the number of cells in control wells. Each point represents mean of three points ± SD. * differed significantly from the cisplatin-treated cells (P < 0.05)

Figure 4.

Effect of acivicin on the toxicity of a 3 h treatment of cisplatin-conjugates. LLC-PK₁ cells were exposed for 3 h to 50 mM cisplatin preincubated with equimolar glutathione, cysteinyl-glycine, or N-acetyl-cysteine. Cells were either treated with no inhibitor (black bars) or were treated with acivicin (white bars) to inhibit GGT activity. The cisplatin solutions and acivicin were removed at the end of the 3 h exposure and fresh medium added to the cells. At 72 h, the number of cells killed by the cisplatin-conjugates was determined as a percentage of the number of cells in control wells. Each point represents mean of three points ± SD. * treatment with acivicin significantly reduced the percentage of cells killed (P < 0.001)

Figure 5.

Effect of AOAA on the toxicity of a 3 h treatment of cisplatin-conjugates. LLC-PK₁ cells were exposed for 3 h to 50 µM cisplatin preincubated with equimolar glutathione, cysteinyl-glycine, or N-acetyl-cysteine. Cells were either treated with no inhibitor (black bars) or were treated with AOAA (white bars) to inhibit cysteine-S-conjugate beta-lyase activity. The cisplatin solutions and AOAA were removed at the end of the 3 h exposure and fresh medium added to the cells. At 72 h, the number of cells killed was determined as a percentage of the number of cells in control wells. Each point represents mean of three points ± SD. * treatment with AOAA significantly reduced the percentage of cells killed (P < 0.005)