Glutathione is a tripeptide used by cells to protect against oxidative and free radical damage. It may also be involved in biochemical mechanisms that cause some tumors to become resistant to anticancer drugs. γ-Glutamyl transpeptidase (GGTP) is a membrane-bound enzyme that cleaves extracellular glutathione, providing cells with amino acids necessary for intracellular synthesis of this compound. Increased expression of GGTP has been found in a number of human tumors; however, few studies have examined the contribution of GGTP to tumor glutathione metabolism in vivo. Our goals were to study the utilization of host glutathione by 3-methylcholanthrene (MCA)-induced sarcomas grown in rats and to evaluate the involvement of tumor GGTP in this process. The left ovaries of 21 female Fischer 344 rats were isolated by laparotomy and placed in subcutaneous positions through stab wounds in the abdominal wall. A 3-mm cube of MCA sarcoma was then sutured to each of the isolated ovaries. The MCA implants obliterated the ovarian tissue, yielding isolated tumors with one arterial supply (the ovarian artery) and one draining vein (the ovarian vein, referred to as the tumor vein). After 2 weeks of tumor growth, blood was drawn from the tumor vein, the inferior vena cava (IVC), and the aorta of 16 animals. Glutathione and cysteine concentrations in plasma samples from this blood were determined by high-performance liquid chromatography and used to calculate glutathione and cysteine utilization ratios for the tumor and the systemic circulations ([{concentration aorta − concentration tumor vein }/concentration aorta ] × 100 and [{concentration aorta −concentration IVC }/ concentration aorta ] × 100, respectively). The utilization ratios from these control animals were compared with those from acivicin (AT-125; an irreversible GGTP inhibitor)-treated rats (the remaining five animals). Data are presented as mean ± standard deviation; reported P values are from two-tailed tests of statistical significance. In the control animals, glutathione and cysteine concentrations were significantly lower in the tumor vein (3.55 ± 1.9 and 5.69 ± 2.8 μM, respectively) and in the IVC (5.65 ± 2.3 and 7.17 ± 2.4 μM, respectively) than in the artery (12.48 ± 5.7 and 12.33 ± 5.9 μM, respectively; all P values <.05). In addition, the glutathione utilization ratio was significantly higher for the tumor circulation than for the systemic circulation (69% ± 14% versus 52% ± 14%; P <.003). The combined glutathione and cysteine utilization ratio was also significantly higher for the tumor circulation than for the systemic circulation (116% ± 35% versus 88% ±28%; P <.02). Treatment with AT-125 lowered the tumor glutathione utilization ratio significantly (45% ±12% for treated animals versus 69% ± 14% for control animals; P <.005). Our results show that glutathione and cysteine in the host circulation are used by MCA sarcomas. The significant reduction in tumor utilization of serum glutathione after treatment with AT-125, a GGTP inhibitor, indicates that GGTP is important in tumor glutathione metabolism. [J Natl Cancer Inst 1996;88:193–7]