Nephrotoxicity of chemotherapeutics is a major hindrance in the treatment of various tumors. Therefore, test systems that reflect mechanisms of human kidney toxicity are necessary, and to reduce animal testing cell culture based systems have to be developed. One cell type that is of specific interest in this regard are renal proximal tubular epithelial cells, as they reabsorb substances from human primary urine filtrates and thus are exposed to urinary excreted xenobiotics and are a major target of cisplatin toxicity. While animal studies using gamma glutamyl transferase (GGT) knock-out mice or GGT inhibitors show that GGT activity increases kidney toxicity of cisplatin, the use of various cell models gives contradictory results. We therefore used a cell panel of immortalized human renal proximal tubular epithelial (RPTECs) cell lines differing in GGT activity. Low GGT activity resulted in high cisplatin sensitivity, as observed in RPTEC-SV40 cells or after siRNA mediated knock-down of GGT in RPTEC/TERT1 cells that have high GGT activity. However, the addition of GGT did not rescue, but also increased cisplatin sensitivity and adding GGT inhibitor as well as substrate (glutathione) or product (cysteinyl-glycine) of GGT resulted in decreased sensitivity. While our data suggest that the use of cell panels are of value in toxicology and toxicogenomics, they also emphasize on the complex interplay of toxins with the intracellular and extracellular microenvironment. In addition, we hypothesize that especially epithelial barrier formation and polarity of RPTECs need to be considered in toxicity models to validly predict the in vivo situation.