A series of 6-substituted classical pyrrolo [2,3-d]pyrimidine antifolates with a 3-to 6-carbon bridge between the heterocycle and the benzoyl-L-glutamate (compounds 2, 3, 4 and 5, respectively) was synthesized starting from methyl 4-formylbenzoate and a Wittig reaction with the appropriate triphenylphosphonium bromide, followed by reduction and conversion to the α-bromomethylketones. Cyclocondensation of 2,4-diamino-4-oxopyrimidine with the α-bromoketones, coupling with diethyl-L-glutamate and saponification afforded 2-5. Compounds 2-5 had negligible substrate activity for RFC but showed variably potent (nanomolar) and selective inhibitory activities toward Chinese hamster ovary cells that expressed FRα or FRβ, and toward FRα-expressing KB and IGROV1 human tumor cells. Inhibition of KB cell colony formation was also observed. Glycinamide ribonucleotide formyl transferase (GARFTase) was identified as the primary intracellular target of the pyrrolo [2,3-d]pyrimidines. The combined properties of selective FR targeting, lack of RFC transport, and GARFTase inhibition resulting in potent antitumor activity are unprecedented and warrant development of these analogs as antitumor agents.

Folates, the generic term for the family of B vitamins, are derived entirely from dietary sources, and are key one-carbon donors required for de novo nucleotide and methionine synthesis. These highly hydrophilic molecules utilize genetically distinct and functionally diverse transport systems to enter cells: the reduced folate carrier (RFC), the proton-coupled folate transporter (PCFT), and the folate receptors. Each plays a unique role in mediating folate transport across epithelia and into systemic tissues. With the recent discovery of the mechanism of intestinal folate absorption, and the clarification of the genetic basis for the autosomal recessive disorder, hereditary folate malabsorption, involving loss-of-function mutations in PCFT protein, it is now possible to piece together how these folate transporters contribute, both individually and collectively, to folate homeostasis in humans. This review focuses on the physiological roles of these major folate transporters with a brief consideration of their impact on the pharmacological activities of antifolates.

No abstract

This review attempts to provide a comprehensive overview of the biology of the physiologically and pharmacologically important transport system termed the "reduced folate carrier" (RFC). The ubiquitously expressed RFC has unequivocally established itself as the major transport system in mammalian cells and tissues for a group of compounds including folate cofactors and classical antifolate therapeutics. Loss of RFC expression or function may have potentially profound pathophysiologic consequences including cancer. For chemotherapeutic antifolates used for cancer such as methotrexate or pemetrexed, synthesis of mutant RFCs or loss of RFC transcripts and proteins results in antifolate resistance due to incomplete inhibition of cellular enzyme targets and insufficient substrate for polyglutamate synthesis. Since RFC was first cloned in 1994, tremendous advances have been made in understanding the complex transcriptional and posttranscriptional regulation of RFC, in identifying structurally and functionally important domains and amino acids in the RFC molecule as a prelude to establishing the mechanism of transport, and in characterizing the molecular defects in RFC associated with loss of transport in antifolate resistant cell line models. Many of the insights gained from laboratory models of RFC portend opportunities for modulating carrier expression in drug resistant tumors, and for designing a new generation of agents with improved transport by RFC or substantially enhanced transport by other folate transporters over RFC. Many of the advances in the basic biology of RFC in cell line models are now being directly applied to human cancers in the clinical setting, most notably pediatric acute lymphoblastic leukemia and osteogenic sarcoma.

2-Amino-4-oxo-6-substituted pyrrolo[2,3-d]pyrimidine antifolates with a thienoyl side chain (compounds 1–3, respectively) were synthesized for comparison with compound 4, the previous lead compound of this series. Conversion of hydroxyl acetylen-thiophene carboxylic esters to thiophenyl-α-bromomethylketones and condensation with 2,4-diamino-6-hydroxypyrimidine afforded the 6-substituted pyrrolo[2,3-d]pyrimidine compounds of type 18 and 19. Coupling with L-glutamate diethyl ester, followed by saponification, afforded 1–3. Compound 3 selectively inhibited proliferation of cells expressing folate receptors (FRs) α or β, or the proton-coupled folate transporter (PCFT), including human tumor cells KB and IGROV1 much more potently than 4. Compound 3 was more inhibitory than 4 toward β-glycinamide ribonucleotide formyltransferase (GARFTase). Both 3 and 4 depleted cellular ATP pools. In SCID mice with IGROV1 tumors, 3 was more efficacious than 4. Collectively, our results show potent antitumor activity for 3 in vitro and in vivo, associated with its selective membrane transport by FRs and PCFT over RFC and inhibition of GARFTase, clearly establishing the 3-atom bridge as superior to the 1, 2 and 4-atom bridge lengths for the activity of this series.

A series of seven 2-amino-4-oxo-6-substituted thieno [2,3-d]pyrimidines, with bridge length variations (from 2-8 carbon atoms) were synthesized as selective folate receptor (FR) α and β substrates and as antitumor agents. The syntheses were accomplished from appropriate allylalcohols and 4-iodobenzoate to afford the aldehydes which were converted to the appropriate 2-amino-4-carbethoxy-5-substituted thiophenes 23-29. Cyclization with chlorformamidine afforded the thieno [2,3-d]pyrimidines 30-36 which were hydrolyzed and coupled with diethyl-L-glutamate, followed by saponification to give the target compounds 2-8. Compounds 3-6 were potent growth inhibitors (IC 50 4.7 to 334 nM) of human tumor cells (KB and IGROV1) that express FRs. In addition, compounds 3-6 inhibited the growth of Chinese hamster ovary (CHO) cells that expressed FRs but not the reduced folate carrier (RFC) or proton-coupled folate transporter (PCFT). However, the compounds were inactive toward CHO cells that lacked FRs but contained either the RFC or PCFT. By nucleoside and 5-amino-4-imidazole carboxamide (AICA) protection studies, along with in vitro and in situ enzyme activity assays, the mechanism of antitumor activity was identified as the dual inhibition of glycinamide ribonucleotide formyltransferase and, likely, AICA ribonucleotide formyltransferase. The dual inhibitory activity of the active thieno [2,3-d]pyrimidine antifolates and the FR specificity represent unique mechanistic features for these compounds distinct from all other known antifolates. The potent inhibitory effects of compounds 3-6 toward cells expressing FRs but not PCFT provide direct evidence that cellular uptake of this series of compounds by FRs does not depend on the presence of PCFT and argues that direct coupling between these transporters is not obligatory.

Abstract2-Amino-4-oxo-6-substituted pyrrolo [2,3-d]pyrimidines with a thienoyl side chain and 4-6 carbon bridge lengths (compounds 1-3) were synthesized as substrates for folate receptors (FRs) and the proton-coupled folate transporter (PCFT). Conversion of acetylene carboxylic acids to α-bromomethylketones and condensation with 2,4-diamino-6-hydroxypyrimidine afforded the 6-substituted pyrrolo [2,3-d]pyrimidines. Sonogashira coupling with (S)-2-[(5-bromo-thiophene-2-carbonyl)-amino]-pentanedioic acid diethyl ester, followed by hydrogenation and saponification, afforded 1-3. Compounds 1 and 2 potently inhibited KB and IGROV1 human tumor cells that express FRα, reduced folate carrier (RFC), and PCFT. The analogs were selective for FR-and PCFT over RFC. Glycinamide ribonucleotide formyltransferase was the principal cellular target. In SCID mice with KB tumors, 1 was highly active against both early (3.5 log kill, 1/5 cures) and advanced (3.7 log kill, 4/5 complete remissions) stage tumors. Our results demonstrate potent in vitro and in vivo antitumor activity for 1 due to selective transport by FRs and PCFT over RFC.

This report describes the isolation, nucleotide sequencing, and functional expression of human cDNAs that restore reduced folate carrier activity in transport-defective cells. Based on homology to a partial murine cDNA probe, two functional cDNAs were isolated from a lambda gt11 library prepared from methotrexate transport upregulated K562 cells (K562.4CF). A 2.8-kilobase (kb) clone, KS43, contained a 1776-base pair open reading frame. The 2.5-kb clone, KS32, contained an internal deletion (626 base pairs) resulting a shortened open reading frame and 3'-untranslated region. KS43 and KS32 encoded proteins with multiple hydrophobic domains, one consensus N-glycosylation site, and predicted molecular masses of 65 and 58 kDa, respectively. The deduced amino acid sequence of KS43 is 79% and 80% homologous to the mouse and hamster sequences, respectively (Dixon, K. H., Lanpher, B. C., Chiu, J., Kelley, K., and Cowan, K. H. (1994) J. Biol. Chem. 269, 17-20; Williams, F. M. R., Murray R. C., Underhill, T. M., and Flintoff, W. F. (1994) J. Biol. Chem. 269, 5810-5816). Northern blots identified one primary transcript at 3.1 kb in parental K562, K562.4CF, and transport-impaired K500E cells; transcript levels varied by 7-fold. The expression of both KS43 and KS32 in methotrexate transport-defective Chinese hamster ovary cells restored methotrexate sensitivity and transport. Certain transport characteristics of the transfected cells resembled both the wild type human (K562) and hamster "classical" reduced folate carriers, suggesting the expression of a hybrid system. For instance, based on Ki values, up to a 4-fold increased affinity for 1843U89 over wild type hamster cells (typical of human cells), and a 19-fold increased affinity for methotrexate over K562 cells (typical of hamster cells) was observed. Further, a photoaffinity probe with high specificity for the reduced folate carrier labeled 94-kDa proteins in K562 cells and the transfectant containing the full-length KS43, and a 85-kDa protein in the transfectant containing the 3'-truncated KS32. No specifically labeled proteins were detected in wild type or mock-transfected hamster cells. Collectively, our results suggest that the KS43/KS32 cDNAs encode the human reduced folate carrier; however, additional modulatory/regulatory factors may be required to manifest the full spectrum of transport substrate activities typical of this system.