FOLATE DEPLETION INCREASES SENSITIVITY OF SOLID TUMOR CELL LINES TO 5-FLUOROURACIL AND ANTIFOLATES H.H.J. BACKUS 1 , H.M. PINEDO 1 , D. WOUTERS 1 , J.M. PADR ´ ON 1,2 , N. MOLDERS 1 , C.L. van der WILT 1 , C.J. van GROENINGEN 1 , G. JANSEN 1 and G.J. PETERS 1 * 1 Department of Medical Oncology, University Hospital Vrije Universiteit, Amsterdam, The Netherlands 2 DSM Research, Geleen, The Netherlands Cancer cell lines in standard cell culture medium or in animal models are surrounded by an environment with rel- atively high folate (HF) levels, compared with folate levels in human plasma. In the present study we adapted 4 colon cancer (C26-A, C26-10, C26-G and WiDr) and 3 squamous cell carcinoma of the head and neck (HNSCC) cell lines (11B, 14C and 22B) to culture medium with low folate (LF) levels (2.5, 1.0 and 0.5 nM, respectively) and investigated whether folate depletion had an effect on sensitivity to antifolates and which mechanisms were involved. All LF cell lines showed a higher sensitivity to 5-fluorouracil (5-FU) alone or in combi- nation with leucovorin (LV) (2–5-fold), to the thymidylate synthase (TS) inhibitors, AG337 (2–7-fold), ZD1694 (3– 49- fold), ZD9331 (3– 40-fold), LY231514 (2–21-fold) or GW1843U89 (4 –29-fold) or to the dihydrofolate reductase (DHFR) inhibitor PT523 (2–50-fold) compared with their HF variants cultured in standard medium containing up to 8 M folic acid. LV could only increase sensitivity to 5-FU in HNSCC cell lines 14C and 14C/F. The differences in sensitiv- ity could partially be explained by a 2–7-fold increased trans- port activity of the reduced folate carrier (RFC) in LF cell lines, whereas no significant change in folylpolyglutamate synthetase (FPGS) activity was observed. Furthermore, the protein expression and catalytic activity of the target enzyme TS were up to 7-fold higher in HF colon cancer cells com- pared with the LF variants (p < 0.05). Although the TS protein expression in LF HNSCC cells was also lower than in HF variants, the TS catalytic activity and FdUMP binding sites were up to 3-fold higher (p < 0.05). Thus, changes in TS levels were associated with differences in sensitivity. These results indicate that folate depletion was associated with changes in TS and RFC levels which resulted in an increase in sensitivity to 5-FU and antifolates. The folate levels in LF medium used in this study are more representative for folate levels in human plasma and therefore these data could be more predictive for the activity of 5-FU and antifolates in a clinical setting than results obtained from cell lines cultured in HF medium or in animal models. Int. J. Cancer 87:771–778, 2000. © 2000 Wiley-Liss, Inc. Reduced folate cofactors are essential for the survival and proliferation of normal and malignant mammalian cells. Since mammalian cells lack a de novo folate biosynthesis pathway, they depend on extracellular uptake and utilization of folates for DNA synthesis. A number of folate-based drugs have been developed for cancer treatment (Jackman and Calvert, 1995; Peters and Ackland, 1996; Takemura and Jackman, 1997; Rustum et al., 1997). These folate analogues (antifolates) inhibit 1 or more of 3 key enzymes in folate metabolism; thymidylate synthase (TS), dihydrofolate reductase (DHFR) and glycinamide ribonucleotide transformylase (GARFTase). TS catalyzes the non-reversible methylation of de- oxyuridine-5'-monophosphate (dUMP) to deoxythymidine-5'- monophosphate (dTMP), a precursor for DNA synthesis, for which 5,10-methylene tetrahydrofolate (5,10-CH 2 -THF) is the methyl donor (Peters and Ackland, 1996; Takemura and Jackman, 1997; Calvert, 1999). The synthesis of dTMP also results in oxidization of 5,10-CH 2 -THF to the inactive dihydrofolate (DHF), which can be converted back to tetrahydrofolate (THF) by DHFR (Bertino, 1993). GARFTase is involved in the purine de novo synthesis pathway (Baldwin et al., 1991). TS can also be inhibited by 5-fluoro-2'-deoxy-uridine-5'-mono- phosphate (FdUMP), the active metabolite of 5-fluorouracil (5- FU). The combination of 5-FU with leucovorin (LV) forms part of the standard treatment of advanced colorectal carcinoma (Moertel, 1978; Moertel, 1994). The effect of 5-FU is mediated by formation of a ternary complex between FdUMP, TS and 5,10-CH 2 -THF leading to TS inhibition (Berger et al., 1984; Spears et al., 1988; Peters and Jansen, 1996). This complex formation is stimulated by LV (Mini et al., 1990; Peters et al., 1994). Although TS inhibition is an important mechanism of 5-FU cytotoxicity, 5-FU is also incorporated into RNA (Peters and Ko ¨hne, 1999) and DNA (Lo ¨nn and Lo ¨nn, 1988). In contrast to 5-FU, both antifolates and natural folates have 2 fundamental properties, which are of major importance for their activity. First, the requirement for cellular uptake via the reduced folate carrier (RFC); second, (anti)folates can be polyglutamylated by the enzyme folylpolyglutamate synthetase (FPGS). Develop- ment of resistance to antifolates, which limits clinical effective- ness, can therefore occur at the levels of RFC, FPGS and target enzymes DHFR and TS (Bertino, 1993; Peters and Jansen, 1996; Gorlick et al., 1996). In order to circumvent the problem of resistance, several new antifolate compounds [e.g., ZD1694 (To- mudex; Jackman et al., 1991), MTA (multi-targeted antifolate or LY231514; Taylor et al., 1992; Shih et al., 1997), GW1843U89 (Hanlon and Ferone, 1996), ZD9331 (Jackman et al., 1997), AG337 (Thymitaq; Webber et al., 1996; Rafi et al., 1998) and PT523 (Rhee et al., 1994; Rosowsky, 1999)] have been developed in the past decade. The activity of these new antifolates has been extensively investigated in vitro and several of these antifolates are currently in clinical development (Rusthoven et al., 1999; Coccini et al., 1998; Rinaldi, 1999; Calvert and Walling, 1998). In vitro and in vivo models, however, have a disadvantage that may influ- ence the sensitivity to the tested antifolates. In particular, mouse plasma contains folate levels, predominantly 5-CH 3 -THF, up to 10-fold higher than in human plasma (5–20 nM 5-CH 3 -THF) Abbreviations: HEPES, 4-(hydroxyethyl)-1-piperazine-ethanesulfonic acid; RFC, reduced folate carrier; dUMP, deoxyuridine-5'-monophos- phate; THF, tetrahydrofolate; 5-FU, 5-fluorouracil; LV, l-leucovorin (L-5-formyltetrahydrofolate); MTX, methotrexate; TMQ, trimetrexate; PT523, N  -(4-amino-4-deoxypteroyl)-N  -(hemiphthaloyl-L-ornithine); GW1843U89, (S)-2-(5-(((1, 2-dihydro-3-methyl-1-oxobenzo(f)quinazolin- 9-yl)-methyl)-amino)-1-oxo-2-isoindolinyl)-glutaric acid; ZD9331, (2S)-2- {O-fluoro-p-[N-(2,7-dimethyl-4-oxo-3,4-dihydro-quinazolin-6-ylmethyl)- N-(prop-2-ynyl)amino]benzamido}-4-(tetrazol-5-yl)-butyric acid; ZD1694, N-[5-(N-(3,4-dihydro-2-methyl-4-oxoquinazolin-6-yl-methyl)- amino)-2-thenyl)]-L-glutamic acid (Tomudex, Raltitrexed); LY231514, N-(4-(2-(2-amino-4,7-dihydro-4-oxo-3H-pyrrolo[2,3-D]pyrimidin-5-yl)- ethyl)-benzoyl]-L-glutamic acid (ALIMTA, Pemetrexed, MTA, (multi tar- geted antifolate)); AG337, 3,4-dihydro-2-amino-6-methyl-4-oxo-5-(4- pyridylthio)-quinazoline (Thymitaq, Nolatrexed); HNSCC, squamous cell carcinoma of the head and neck. Grant sponsor: Dutch Cancer Society; Grant number: VU 96-1240. *Correspondence to: Department of Medical Oncology, University Hos- pital Vrije Universiteit, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. Fax: 31-20-4443844. E-mail:gj.peters@azvu.nl Received 30 November 1999; Revised 6 March 2000; Accepted 10 March 2000 Int. J. Cancer: 87, 771–778 (2000) © 2000 Wiley-Liss, Inc. Publication of the International Union Against Cancer