The clinical efficacy of anticancer nucleoside drugs depends on a complex interplay of transporters mediating entry of nucleoside drugs into cells, efflux mechanisms that remove drugs from intracellular compartments and cellular metabolism to active metabolites. Nucleoside transporters (NTs) are important determinants for salvage of preformed nucleosides and mediated uptake of antimetabolite nucleoside drugs into target cells. The focus of this review is the two families of human nucleoside transporters (hENTs, hCNTs) and their role in transport of cytotoxic chemotherapeutic nucleoside drugs. Resistance to anticancer nucleoside drugs is a major clinical problem in which NTs have been implicated. Single nucleotide polymorphisms (SNPs) in drug transporters may contribute to interindividual variation in response to nucleoside drugs. In this review, we give an overview of the functional and molecular characteristics of human NTs and their potential role in resistance to nucleoside drugs and discuss the potential use of genetic polymorphism analyses for NTs to address drug resistance.
Human nucleoside transporters
Nucleoside transporters (NTs) mediate the uptake of physiologic nucleosides as well as both anticancer and antiviral nucleoside drugs (Baldwin et al., 1999). Specialized cells, including enterocytes, bone marrow cells and certain brain cells (Murray, 1971) lack de novo synthesis pathways and salvage nucleosides from the extracellular milieu for the production of nucleotides for use in RNA and DNA synthesis. In addition to their role as intermediates for many essential cellular biosynthetic pathways, nucleotides play key roles in neurotransmission (Baldwin et al., 1999) and regulation of cardiovascular activity (Shryock and Belardinelli, 1997) and as signaling molecules (Schachter et al., 1995). NTs play a role in the maintenance of extracellular concentrations of adenosine available to bind to receptors to modulate a variety of physiological processes. Plasma membrane transport of nucleosides is a critical determinant in the salvage and release of preformed nucleosides in many cell types (Cass, 1995; Griffith and Jarvis, 1996; Baldwin et al., 1999; Young et al., 2001). Polar hydrophilic nucleoside analogs used in cancer chemotherapy require functional NT proteins to enter cells. Phosphorylation of nucleosides to nucleotides is mediated by nucleoside kinases present in cells. The interaction between nucleoside drugs and intracellular enzymes (e.g., kinases, deaminases and nucleotidases) may regulate the rate and extent of metabolism of nucleoside drugs to cytotoxic derivatives. This review gives an overview of the functional and molecular characteristics of NTs present in human tissues with a focus on localization and regulatory mechanisms. Important structural determinants for recognition of nucleosides by membrane transporters and clinically used nucleoside drugs and resulting resistance mechanisms are reviewed. The reader is referred to recently published review articles on the molecular structure and function of NT proteins and their role in drug resistance (Mackey et al., 1998a; Baldwin et al., 1999; Vickers et al., 2000; Cabrita et al., 2002; Clarke et al., 2002).
The two major classes of NTs in human cells and tissues consist of equilibrative nucleoside transporters (ENTs) and concentrative nucleoside transporters (CNTs). In the human ENT family, two proteins (hENT1 and hENT2) are known to mediate the transport of purine and pyrimidine nucleosides down their concentration gradients. hENTs exhibit broad permeant selectivities and are subdivided based on their sensitivities (hENT1) or resistance (hENT2) to inhibition by nanomolar concentrations of nitrobenzylmercaptopurine ribonucleoside (NBMPR). hENT1 and hENT2 mediate the well-characterized equilibrative-sensitive (es) and equilibrative-insensitive (ei) processes. Although both hENT1 and hENT2 have similar selectivities for nucleosides, hENT2 also recognizes and transports hypoxanthine and other purine nucleobases (Osses et al., 1996; Baldwin et al., 1999; Yao et al., 2002). The role of hENT3 and hENT4 in nucleoside transport processes has not yet been established (Baldwin et al., 2003).
Six functional subtypes of CNT processes (Gutierrez et al., 1992; Belt et al., 1993; Gutierrez and Giacomini, 1993, 1994; Paterson et al., 1993; Cass, 1995; Griffith and Jarvis, 1996; Flanagan and Meckling-Gill, 1997; Ritzel et al., 1997, 1998, 2001a 2001b; Cass et al., 1999) have been described in human cells or tissues. The three best characterized CNTs are Na+-nucleoside cotransporters and are termed cit, cif and cib to indicate that they are concentrative, insensitive to NBMPR and accept as permeants, respectively, pyrimidine nucleosides, purine nucleosides or both pyrimidine and purine nucleosides. The transporter proteins that are responsible for these activities are cit (hCNT1), cif (hCNT2) and cib (hCNT3). However, the proteins responsible for cs (concentrative and sensitive to NBMPR) (Paterson et al., 1993), csg (concentrative, sensitive to NBMPR and accepts guanosine as permeant) (Flanagan and Meckling-Gill, 1997) and cit with specificity towards guanosine have not been identified (Gutierrez et al., 1992; Gutierrez and Giacomini, 1993, 1994) and are not discussed further in this review. The mammalian transporter responsible for the cit process (now known to be CNT1) was first described in freshly isolated mouse intestinal cells (Vijayalakshmi and Belt, 1988) and exhibited a preference for pyrimidine nucleosides. Although inhibited by low concentrations of adenosine, CNT1 does not transport adenosine well. The cif process (i.e., CNT2) first described in mouse enterocytes (Vijayalakshmi and Belt, 1988) transports purine nucleosides including formycin-B (C-nucleoside analog of inosine) in addition to uridine. Uridine appears to be a universal permeant for all NTs. The hCNT1 and hCNT2 proteins exhibit 1 : 1 Na+–uridine coupling ratios.
The cib process (i.e., hCNT3) is broadly selective and transports both purine and pyrimidine nucleosides in a concentrative manner into cells. This process was first described in freshly isolated leukemic blasts (Belt et al., 1992) and human colon cancer Caco-2 cells (Belt et al., 1993). The hCNT3 protein exhibits a 2 : 1 stiochiometry for cotransport of Na+ and uridine.
The isolation of cDNAs encoding hENT1 (Griffiths et al., 1997a) and hENT2 (Griffiths et al., 1997b; Crawford et al., 1998) permitted detailed molecular and functional characterization of these transporters. The hENT1 cDNA encodes a protein with 11 transmembrane domains and a single extracellular site for glycosylation (Griffiths et al., 1997a, 1997b; Yao et al., 1997; Kiss et al., 2000; Handa et al., 2001). hENT1 is known to be a heterogeneously glycosylated protein (45–65 kDa) (Kwong et al., 1993) and the predicted 11 transmembrane topology has been experimentally verified by glycosylation scanning mutagenesis and use of antibodies (Sundaram et al., 2001a). hENT1 and hENT2 share 49% amino-acid identity and 69% similarity and transport purine and pyrimidine nucleosides with different affinities (Griffiths et al., 1997a, 1997b; Ward et al., 2000).
hCNT1, which is responsible for cit activity in humans, was identified in the human kidney by hybridization/RT–PCR cloning and functional expression in Xenopus laevis oocytes of two closely related cDNAs (Ritzel et al., 1997). Minor differences between these two cDNAs are felt to be due to genetic polymorphisms and/or errors during PCR amplification and both encode a functional hCNT1 protein. hCNT1 is predicted to have 13 transmembrane domains. Human cDNAs encoding hCNT2 (also termed hSPNT1), which has cif activity, were isolated from the kidney and intestine (Wang et al., 1997; Ritzel et al., 1998) by hybridization/RT–PCR cloning. The human homologs were almost identical (658 amino acids), with the exception of a single amino-acid substitution at position 75, which is arginine in hCNT2 (Ritzel et al., 1998) and serine in hCNT2/hSPNT1 (Wang et al., 1997). Molecular cloning of cDNAs from differentiated HL-60 cells and human mammary gland resulted in identification of sequences encoding a third concentrative transporter (hCNT3, 691 amino acids) that exhibits a cib-like activity (Ritzel et al., 2001b). There is an 85% amino-acid identity between the CNT in transmembrane domains 7–9 (Loewen et al., 1999). hCNT2 is 72% identical to hCNT1 (Ritzel et al., 1997) and hCNT3 is 48 and 47% identical to hCNT1 and hCNT2, respectively (Ritzel et al., 2001a).
Chromosomal localization of transporter genes
The genes for hENT1 and hENT2, respectively, have been localized to chromosome 6p21.1–p21.2 (Coe et al., 1997) and to chromosome 11 (GenBank™ accession NT_009379; International Human Genome Project). The genes for hCNT1 (GenBank™ accession AF036109) and hCNT2 (GenBank™ accession U62966) map, respectively, to chromosome 15q25–26 (Ritzel et al., 1997) and chromosome 15q15 (15q13–14) (Wang et al., 1997; Ritzel et al., 1998). The hCNT3 gene (GenBank™ accession AF305210) has an upstream phorbol-responsive element and is localized on chromosome 9q22.2 (Ritzel et al., 2001b) (Table 1).
Tissue distribution of transporters
hENT1 appears to be a ubiquitous protein, since its mRNA has been found in a diverse array of human tissues (Pennycooke et al., 2001). The broad tissue distribution of hENT1 suggests that this NT is likely to be involved in the uptake of chemotherapeutic nucleoside analogs as well as in the uptake of the physiologic nucleosides. High-level expression of hENT1 mRNA was observed in adrenal tissue (Pennycooke et al., 2001) although its significance is not yet clear. Expression of hENT1 mRNA in other endocrine tissues suggests the possible regulation of hENT1 by steroid hormones (Fideu and Miras-Portugal, 1992, 1993). hENT2 mRNA has been observed in a variety of tissues, including the brain, heart, lung, thymus, prostate, pancreas, with the highest levels occurring in skeletal muscle (Crawford et al., 1998). The predominant distribution of hENT2 mRNA in skeletal muscle (Pennycooke et al., 2001) suggests its role as a transporter in skeletal muscle (Cass et al., 1998; Crawford et al., 1998). The hENT2 protein has also been shown to be widely present in the brain (Jennings et al., 2001).
Mammalian CNT families of transporters are found in highly differentiated tissues such as the epithelial lining of the intestine and kidney (Cass et al., 1999), although they appear to be present at low levels in other tissues. The high abundance of CNTs in liver and renal tissues may contribute to the renal and hepatic toxicities observed when patients are treated with nucleoside analog drugs (Loewen et al., 1999). In the intestine, CNTs were localized by functional studies to brush border membranes (Chandrasena et al., 1997; Patil and Unadkat, 1997; Ngo et al., 2001), whereas ENTs were localized to basolateral membranes (Chandrasena et al., 1997; Lum et al., 2000). There is a wide variation in the expression of hCNT1 mRNA in normal kidney in different individuals (Pennycooke et al., 2001) and relatively low levels of expression of hCNT1 mRNA in tumor samples (Pennycooke et al., 2001). The expression of hCNT2 mRNA was detected in a variety of human tissues, with highest expression in tissues of the digestive system and kidney, followed by the prostate, cervix and lung (Pennycooke et al., 2001). High expression of hCNT2 mRNA was seen in samples of tumor tissue derived from the lung, ovary, uterus and prostate. hCNT3 transcripts have been observed in differentiated HL-60 cells, mammary gland, pancreas, bone marrow, trachea, liver, prostate, brain, heart and intestine (Ritzel et al., 2001b).
Uridine transport in hENT1-producing X. laevis oocytes is inhibited by physiological purine and pyrimidine nucleosides and by the nucleoside drugs cladribine, cytarabine, fludarabine and gemcitabine as well as by NBMPR, dipyridamole, dilazep and draflazine (Griffiths et al., 1997b). Studies of the properties of chimeras between human and rat ENTs have established that transmembrane domains 3–6 contain residues responsible for sensitivity or resistance to coronary vasodilators (Sundaram et al., 1998) and to NBMPR (Sundaram et al., 2001b). Transmembrane domains 1–6 of ENT2 appear to be responsible for transport of 3′-deoxynucleosides (Yao et al., 2001) while transmembrane domains 5–6 appear to be involved in its transport of nucleobases (Yao et al., 2002). Using site-directed mutagenesis, binding of NBMPR to recombinant hENT1 produced in Saccharomyces cerevisiae has been investigated. A mutant in which the methionine at position 33 of hENT1 was modified to isoleucine (generated using a random mutagenesis strategy) exhibited reduced inhibition of uridine transport by dipyridamole (Visser et al., 2002). Modification of glycine to serine at position 154 of hENT1 resulted in loss of NBMPR binding (Hyde et al., 2001). Uridine transport and inhibition by NBMPR were diminished when the highly conserved glycine at position 179 was changed to alanine, and uridine transport was abolished when glycine was changed to leucine, cysteine or valine (SenGupta et al., 2002). Glycine at position 184 was shown to be important in either targeting of hENT1 to the plasma membrane or in correct folding (SenGupta et al., 2002) using GFP-tagged hENT1 protein in yeast expression system. Interaction of hENT1 with analogs of uridine and cytidine containing base or ribose modifications has been described using recombinant transporters produced in S. cerevisiae (Vickers et al., 2002). Decreased recognition of cytidine analogs by hENT1 was demonstrated using 2′-deoxycytidine and cytarabine (araC). Both the 2′- and 3′-hydroxyls are important determinants for interaction of cytidine with hENT1. Addition of bulky substituents at position 5 of the base resulted in lowered affinity of the analogs for hENT1, whereas the 3′-hydroxyl group appeared to be absolutely essential for interaction with the transporter (Zimmerman et al., 1987; Vickers et al., 2002). Antiviral nucleoside drugs that lack the 3′-hydroxyl group on the ribose ring are not or are very weakly transported by hENT1 (Yao et al., 2001).
Studies with recombinant hENT2 produced in oocytes (Griffiths et al., 1997b) revealed complete inhibition of uridine transport by adenosine, inosine or thymidine, partial inhibition by guanosine, cytidine or hypoxanthine and no inhibition by adenine or uracil. hENT2 mediated significant transport of antiviral drugs, zidovudine, zalcitabine and didanosine (Yao et al., 2001). Site-directed mutagenesis of isoleucine at position 33 to methionine in hENT2 resulted in the expression of a protein with higher sensitivity to dipyridamole (Visser et al., 2002). A notable difference between hENT1 and hENT2 is the lower affinity of hENT2 to cytidine; kinetic studies of fluxes of cytidine in X. laevis oocytes gave Km values of 0.56 and >5 mM for hENT1 and hENT2, respectively (Young et al., 2001). Modifications in the 5 position of the pyrimidine moiety of uridine were better tolerated by hENT2 than by hENT1 and modification in the 3′-hydroxyl position of the ribose moiety of uridine abolished interaction with hENT2 (Vickers et al., 2002). A higher affinity for inosine by hENT2 than by hENT1 has been shown and is consistent with a role for hENT2 in the efflux and reuptake of inosine and hypoxanthine generated from adenosine metabolism during and after strenuous physical exercise (Crawford et al., 1998; Ward et al., 2000).
Recombinant hCNT1 produced in X. laevis oocytes transports uridine, zidovudine, zalcitabine, gemcitabine and is inhibited by adenosine, deoxycytidine, thymidine, cytidine and uridine, but not by guanosine or inosine (Ritzel et al., 1997; Mackey et al., 1999; Graham et al., 2000). Although adenosine binds with high affinity to hCNT1, it is transported very poorly compared to uridine. Recombinant hCNT1, when expressed in X. laevis oocytes, mediates the transport of uridine (Km 42 μ M) and gemcitabine (Km 24 μ M) with high affinity. Amino acids 319 (serine) and 320 (glutamine) in transmembrane domain 7 of hCNT1 were shown to be necessary for the pyrimidine nucleoside specificity of hCNT1-mediated transport (Loewen et al., 1999).
Permeant selectivity of hCNT2 was examined using X. laevis oocyte expression system (Ritzel et al., 1998) and hCNT2 accepted inosine, adenosine, uridine, 2′-deoxyadenosine, guanosine and didanosine as permeants. The importance of 3′- and 5′-hydroxyl groups of the ribosyl moiety of uridine for permeant recognition by hCNT2 was demonstrated in the stably transfected CEM-ARAC (transport deficient clone) leukemic cell line (Lang et al., 2001). The purine nucleoside selectivity of hCNT2 is demonstrated by its lower Km for adenosine (8 μ M) than for uridine (40 μ M). hCNT1 and hCNT2 differ in permeant selectivity in that hCNT1 transports zidovudine and zalcitabine whereas hCNT2 transports didanosine. The importance of 3′- and 5′-hydroxyl groups of the ribosyl moiety of uridine for permeant recognition by hCNT2 was demonstrated in a stably transfected CEM-ARAC (transport-deficient clone) leukemic cell line (Lang et al., 2001). Removal of the hydroxyl group at the 2′-position did not affect the interaction of uridine with hCNT2. 5-Fluorouridine and 5-fluoro-2′-deoxyuridine were transported by hCNT2 with high affinity, suggesting that hCNT2 may also be involved in the uptake of these drugs into cells (Lang et al., 2001). Transport activity of halogenated analogs of adenosine by hCNT2 was lower compared with halogenated uridine analogs (Lang et al., 2001). Loewen et al. (1999) identified amino-acid residues in transmembrane domains 7 (Ser 319/Gln 320) and 8 (Ser 353/Leu 354) that, when converted to the corresponding residues in hCNT2, converted the permeant selectivity of CNT1 to a CNT2-like selectivity.
Functional studies of recombinant hCNT3 in X. laevis oocytes resulted in the expression of a transport activity (cib) with broad permeant selectivity. Permeants include the physiological purine and pyrimidine nucleosides, as well as anticancer and antiviral nucleoside analogs including 5-fluorouridine>5-fluoro-2′-deoxyuridine>cladribine>zebularine>gemcitabine>fludarabine>zidovudine>2′3′-dideoxyinosine>2′3′-dideoxycytosine (Ritzel et al., 2001b). Apparent Km values for pyrimidine and purine nucleosides were between 15 and 53 μ M and hCNT3 does not transport nucleobases (Ritzel et al., 2001b). Using site-directed mutagenesis, Loewen et al. (1999) demonstrated cib-like transport activity by mutation of the two residues in transmembrane domain 7 of hCNT1 (Ser319 to Gly and mutation of Gln320 to Met).
Regulation of NTs
The abundance of NT proteins in cells and tissues depends in part on the pathways that stimulate proliferation and differentiation of cells. A variety of mechanisms are thought to contribute to the regulation of NTs in different cell types. When HL-60 cells were induced to differentiate using chemical inducing agents, changes in es and ei transport activities, now known to be mediated by the hENT1 and hENT2 proteins, were observed (Chen et al., 1986; Delicado et al., 1991; Lee et al., 1991; Sokoloski et al., 1991; Lee, 1994). NBMPR-binding sites decreased while the dissociation constant (Kd) remained unchanged and hENT1 transport activity decreased upon induction of differentiation in HL-60 cells (Chen et al., 1986). hENT2 transport activity was observed in differentiated human neuroblastoma cells but not in undifferentiated cells (Jones et al., 1994).
Lipopolysaccharides and phorbol esters, which activate B cells, upregulate CNTs and downregulate ENTs in lymphocytes from the bone marrow and in a human B lymphoblast cell line (Soler et al., 1998; Soler et al., 2000; Soler et al., 2001a, 2001b). These effects were shown to result from differential regulation of NTs by agents like protein kinase C (PKC) and tumor necrosis factor-α. Nitric oxide may also be involved in the regulation of NTs in activated human B lymphocytes (Soler et al., 2000). Acute stimulation of δ and ɛ isoforms of PKC resulted in an increase in transport activity in cultured MCF-7 breast cancer cells and HeLa cells, whereas downregulation resulted in a decrease in transport activity (Coe et al., 2002). Transcriptional regulation of hENT1 could be due to the presence of regulatory consensus sites in the promoter region of hENT1 and a possible link to PKC signaling pathways (Sankar et al., 2002).
Protein and mRNA levels for hENT1 were reduced and transport activity was inhibited in human endothelial cells with exposure to D-glucose, a condition that leads to activation of purinoceptors (Parodi et al., 2002). Selective regulation of NTs was shown to be critical for proliferation and activation of murine bone marrow-derived macrophages (Soler et al., 2001a). In addition, upregulation of NBMPR-binding sites was described in certain tumors and rapidly dividing cells (Goh et al., 1995). Inhibitors of protein kinase inhibited ENT1-mediated uptake of uridine and thymidine in human erythroleukemia K562 cells, independent of kinase inhibition. These results suggest the involvement of ENTs as possible targets for protein kinase inhibitors in humans (Huang et al., 2003).
Regulation of CNT processes by lipopolysaccharide, phorbol esters and tumor necrosis factor-α was described in cultured human B and T lymphocytes (Soler et al., 1998; Kichenin et al., 2000; Ritzel et al., 2001b). Correlation between extracellular metabolism of nucleotides/nucleosides and cADPR-mediated regulation of intracellular calcium homeostasis was explored to identify specific equilibrative and concentrative NTs responsible for cADPR translocation into cells (Guida et al., 2002).
Nucleoside analogs currently used in cancer chemotherapy
NTs are important for cellular uptake of anticancer pyrimidine and purine nucleoside analog drugs. NTs are expected to play a role in antimetabolite drug sensitivity, by mediating the uptake of anticancer nucleoside drugs, and in drug resistance, by mediating the salvage of normal nucleosides to overcome toxicity. Both pyrimidine and purine nucleoside analogs are currently used clinically as antimetabolite drugs. One of the first pyrimidine nucleoside analogs to be developed was cytarabine, an analog of deoxycytidine (1-β-D-arabinofuranosylcytosine, araC, Cytosar-U®) (Ellison et al., 1967). Intracellular accumulation of cytarabine depends on the plasma concentrations of the drug and enters cells primarily by hENT1-mediated processes (Wiley et al., 1983; Galmarini et al., 2002). Once cytarabine has entered cells, it is phosphorylated by deoxycytidine kinase (dCK) to the 5′-derivative araCMP and subsequently by nucleotide kinases to araCTP (Hande and Chabner, 1978; Owens et al., 1992). Loss of pharmacologically active drug occurs through deamination of cytarabine by cytidine deaminase, which results in the production of a nontoxic metabolite arabinosyluridine and through dephosphorylation of araCMP by cytoplasmic 5′-nucleotidase. Cytotoxicity of cytarabine results from inhibition of DNA polymerase α (Furth and Cohen, 1968; Heinemann et al., 1988) and from incorporation of araCTP into DNA in place of deoxycytidine triphosphate (dCTP). Furthermore when araCMP is incorporated into DNA, its incorporation prevents DNA chain elongation (Kufe et al., 1980; Fram et al., 1983) resulting in a blockade of DNA synthesis. Cytarabine is used predominately in hematological malignancies, such as acute myelogenous leukemia and non-Hodgkin's lymphoma. When used as a single agent, cytarabine produced remissions in about 30% of patients (Ellison et al., 1968). Combination chemotherapy that includes cytarabine with other chemotherapy agents is used in the treatment of chronic myelogenous leukemia, multiple myeloma, Hodgkin's lymphoma and non-Hodgkin's lymphomas.
Gemcitabine (dFdC, 2′,2′-difluorodeoxycytidine, Gemzar®) is an analog of cytarabine, which was modified at the 2′-position of the ribose ring by substitution of two fluorine atoms (Baker et al., 1991) to give gemcitabine. hENT1, hENT2, hCNT1 and hCNT3 mediate uptake of gemcitabine into cells (Mackey et al., 1999) (Zhang J, personal communication). hENT1 and hCNT1 appear to be most efficient in the transport of gemcitabine into cells. After entry into cells, gemcitabine is phosphorylated to its monophosphate (dFd-CMP) by dCK (Abbruzzese et al., 1991) and to its active triphosphate (dFd-CTP) by pyrimidine nucleotide kinases (Plunkett et al., 1995). Inactivation of gemcitabine or its 5′-monophosphate can occur, respectively, by deamination by cytidine deaminase or dephosphorylation by 5′-nucleotidase (Plunkett et al., 1995; Neff and Blau, 1996). Incorporation of dFd-CTP into the growing DNA strand is followed by addition of a natural nucleoside, thereby preventing DNA repair and causing masked chain termination (Huang and Plunkett, 1995). Prolonged accumulation of gemcitabine compared to cytarabine is believed to be due to inhibition of ribonucleotide reductase and deoxycytidine monophosphate deaminase (Heinemann et al., 1992). Gemcitabine diphosphate inhibits ribonucleotide reductase (Huang and Plunkett, 1995) thereby decreasing deoxynucleotide triphosphate levels; decreased levels of dCTP inhibit deoxycytidine monophosphate deaminase, which increases gemcitabine monophosphate. The higher retention of active metabolites of gemcitabine, compared to cytarabine, in cancer cells is due to a combination of different factors: gemcitabine is a better substrate for NTs, is phosphorylated more rapidly and is eliminated from cells more slowly. Gemcitabine's unique self-potentiating mechanisms contribute to increased accumulation of gemcitabine and its metabolites and are believed to be the basis of its unusual broad spectrum of activity among nucleoside analogs. Gemcitabine was originally approved to treat symptoms of pancreatic cancer (Burris et al., 1997). Subsequent studies showed that gemcitabine has activity against metastatic bladder cancer (Stadler et al., 1997) and gemcitabine combined with cisplatin is now the standard treatment for metastatic bladder cancer (von der Maase et al., 2000). Gemcitabine is also active against non-small-lung cancer (Schiller et al., 2002) and breast, ovarian and head and neck cancers (Possinger, 1995; Kaye, 1998).
Capecitabine (5′-deoxy-5-N-[(pentoxy) carbonyl]-cytidine, Xeloda®) is the most recent pyrimidine nucleoside to be introduced into clinical practice. Capecitabine was developed to overcome the low bioavailability of 5-fluorouracil (Lamont and Schilsky, 1999) and, in addition, offers a convenient oral administration option. Capecitabine is a prodrug that is metabolized by carboxylesterase to 5′-deoxy-5-fluorocytidine after oral administration. 5′-Deoxy-5-fluorocytidine is deaminated by cytidine deaminase to 5′-deoxy-5-fluorouridine. hENT1 mediates the uptake of 5′-deoxy-5-fluorouridine (Mackey et al., 2002). A metabolite of capecitabine, 5′-deoxy-5-fluorouridine monophosphate, inhibits thymidylate synthase and the 5′-deoxy-5-fluorouridine triphosphate is incorporated into DNA (Schmoll et al., 1999). The last activation step is catalysed by thymidine phosphorylase, which converts 5′-deoxy-5-fluoruridine into 5-fluorouracil (Lamont and Schilsky, 1999). Thymidine phosphorylase is highly expressed in tumor tissues and is associated with resistance to conventional 5-fluorouracil treatment in several gastrointestinal tumors, in particular colon cancer (Ishikawa et al., 1998). Capecitabine has shown activity in metastatic colorectal cancer that is comparable to that of 5-fluorouracil combined with leucovorin (Hoff et al., 2001). Capecitabine has activity against metastatic breast cancer that has progressed after docetaxel or anthracyclines (Blum et al., 1999; Talbot et al., 2002).
Two purine nucleoside antimetabolite drugs, fludarabine (9-β-D-arabinofuranosyl-2-fluoroadenine), which is administered as the 5′-monophosphate (F-araAMP, Fludara®), and cladribine (2-chloro-2′-deoxyadenosine, CdA, Leustatin®) are cytotoxic to both dividing and resting cells. One of the first purine nucleoside analogs, 9-β-D-arabinofuranosyladenine (araA), was abandoned due to its poor solubility and rapid deamination by adenosine deaminase. Addition of a fluorine atom to the adenine moiety to create fludarabine increased resistance to adenosine deaminase and addition of a phosphate group improved the analog's solubility and gave rise to Fludara (F-araAMP) (Brockman et al., 1977; Danhauser et al., 1986). Before entering cells, fludara is dephosphorylated by plasma phosphatases and ecto-5′-nucleotidase to fludarabine (F-araA), which is transported into cells by NTs present on plasma membranes. Fludarabine enters cells mainly by hENT1 and hCNT3 (Gati et al., 1998; Ritzel et al., 2001b). Like other nucleoside analogs, fludarabine is initially phosphorylated by dCK to its monophosphate (fludara, F-araAMP) form after which it is further phosphorylated to F-araATP (Sirotnak et al., 1983). F-AraATP inhibits several enzymes involved in nucleoside synthesis and DNA replication: (i) DNA polymerase (Parker et al., 1988; Huang et al., 1990, ii) DNA primase (Catapano et al., 1991, iii) ribonucleotide reductase (Parker et al., 1988) and (iv) DNA ligase I (Yang et al., 1992). In nonreplicating cells, cytotoxicity of fludarabine (F-araA) is due to inhibition of cellular DNA repair processes (Sandoval et al., 1996). Other mechanisms of action of F-araATP in noncycling cells include incorporation into RNA leading to premature chain termination of RNA transcript and impairing cellular protein synthesis (Huang and Plunkett, 1991). Clinically, Fludara is used to treat low-grade lymphomas and chronic lymphocytic leukemia (Chun et al., 1991) (Table 2).
Cladribine (2-CdA, 2-chloro-2′-deoxyadenosine, Leustatin®) is closely related to fludarabine. Cladribine differs from fludarabine in that it has a chlorine substitution instead of fluorine at the 2 position of the adenine moiety. Cladribine, like fludarabine, is resistant to deamination by adenosine deaminase. Cladribine enters cells via hENT1, hENT2 and hCNT3 (King, 1994; Ritzel et al., 2001b) and is converted into the active form 2-CdATP by the combined action of dCK and cellular nucleotide kinases. The affinity of dCK for cladribine is 10-fold higher than that for fludarabine. Cladribine has a similar mechanism of action to fludarabine in that the 5′-triphosphate inhibits DNA replication and repair as well as ribonucleotide reductase thereby reducing deoxyribonucleotide synthesis. Exposure to cladribine is cytotoxic to dividing cells because of the inhibition of replicative DNA synthesis (Lassota et al., 1994) and to resting cells because of the inhibition of DNA repair processes (Pettitt et al., 1999a, 1999b) and alteration of mitochondrial function or integrity (Genini et al., 2000). Cladribine has been shown to be active in low-grade lymphomas (Dimopoulos et al., 1994), chronic lymphocytic leukemia and hairy cell leukemia (Seymour et al., 1994).
Nucleoside analogs in development
Troxacitabine (Troxatyl; BCH-4556; (−)-2′-deoxy-3′-oxacytidine) has unique properties in terms of its structure. An analog of deoxycytidine, troxacitabine exists as an L-enantiomer in contrast to the physiologic nucleosides and most nucleoside drugs, which are D-enantiomers. Cellular uptake of troxacitabine is mainly due to passive diffusion (Gourdeau et al., 2001) and the presence or absence of NTs in tumor cells is unlikely to influence sensitivity to troxacitabine. Troxacitabine is phosphorylated by dCK (Grove et al., 1995). In contrast to most other cytidine analog drugs, troxacitabine is resistant to deoxycytidine deaminase (Grove and Cheng, 1996). The 5′-triphosphate of troxacitabine is a good substrate for DNA polymerases but lacks the necessary hydroxyl group to allow chain elongation (Grove et al., 1995). Troxacitabine metabolites do not inhibit ribonucleotide reductase, in contrast to gemcitabine metabolites (Grove and Cheng, 1996). In phase I studies, troxacitabine had activity against renal cell cancer and primary unknown cancer (Belanger et al., 2002; Townsley et al., 2003). Giles et al. (2001), (2003)) observed the activity of troxacitabine in patients with refractory leukemia. Troxacitabine is being tested as a potential treatment for acute myeloid leukemia, pancreatic cancer and solid tumors (Ecker, 2002).
Clofarabine (Cl-FaraA, 2-chloro-9-(2′-deoxy-2′-fluoro-β-D-arabinofuranosyl) adenine, Clofarex) has activity against both epithelial and hematologic malignancies. Like cladribine and fludarabine, substitution at position 2 of the base with a halogen confers resistance to deamination (Xie and Plunkett, 1996). Substitution of fluorine on the arabinosyl sugar prevents degradation by bacterial phosphorylase and allows oral administration (Carson et al., 1992). Clofarabine enters cells via hENT1, hENT2, hCNT2 and possibly also hCNT3 (King et al., 2002, 2003). Clofarabine is metabolized to its mono-, di- and triphosphates by dCK and nucleotide kinases (Parker et al., 1999). Cl-F-araATP is incorporated into replicating DNA, which terminates chain elongation (Parker et al., 1991), and inhibits ribonucleotide reductase, which decreases intracellular deoxynucleotide pools (Parker et al., 1991). Clofarabine is undergoing phase II clinical trials for the treatment of pediatric refractory/relapsed acute myeloid and lymphocytic leukemia (Lindemalm et al., 2003).
Distribution of NTs in tumor tissues
Pennycooke et al. (2001) examined the expression of hENT1 mRNA in tumors and found expression in the kidney, breast, prostate, uterus, ovary, cervix, colon, lung, stomach and rectum. Although many tumors expressed lower levels of hENT1 mRNA than the corresponding normal tissues, breast, lung, stomach and rectal cancers generally expressed higher levels of hENT1 mRNA than normal tissues. hENT2 mRNA was expressed in the kidney, breast, prostate, colon and stomach cancers at higher levels than in normal tissues. hCNT1 mRNA expression in tumors was more limited with expression observed only in tumors of kidney, uterine, lung and small intestine. Ovarian cancer expressed more hCNT1 mRNA than normal ovarian tissue, whereas all other tumors had less hCNT1 mRNA expression than the corresponding normal tissues. hCNT2 mRNA was expressed in kidney, breast, prostate, uterine, ovarian, colon, lung, stomach and rectal cancers and the majority of breast, prostate, uterine, ovarian and lung cancers expressed more hCNT2 mRNA than the corresponding normal tissues.
NTs and sensitivity to nucleoside analogs
In vitro studies have demonstrated that nucleoside transport is necessary for many nucleoside analogs to enter cells. Mackey et al. (1998b) showed that gemcitabine required nucleoside transport to cause cytotoxicity in a study that compared gemcitabine cytotoxicity in cell lines with or without NT activity. Gemcitabine's IC50 values (concentrations that inhibited proliferation by 50%) were 118–3000-fold greater in cell lines that lacked nucleoside transport activity. A major role for hENT1 in gemcitabine toxicity was demonstrated by addition of dipyridamole, a potent inhibitor of hENT1 activity, to culture media, which conferred resistance to gemcitabine by reducing the cellular uptake of gemcitabine. Mata et al. (2001) showed that hCNT1 was important for cytotoxicity to capecitabine. 5′-Deoxy-5-fluorouridine, the active metabolite of capecitabine, is a permeant for hCNT1 and the presence of hCNT1 conferred sensitivity of CHO-K1 cells to 5′-deoxy-5-fluoruridine. Similarly, Lang et al. (2001) showed that the presence of hCNT2 activity conferred sensitivity to fluoropyrimidine nucleosides in CEM cells transfected with cDNA encoding hCNT2.
Lu et al. (2002) performed a systematic analysis of NT mRNA expression and cytotoxicity to common antimetabolites such as gemcitabine, cytarabine, cladribine and fludarabine and did not find any significant relationship between cytotoxicity and NT mRNA expression. They did find relationships between NT mRNA expression and cytotoxicity of the analogs O6-methylguanine and hCNT1, tiazofurin and hCNT2, cyclopentenylcytosine and hydroxyurea and hENT2. Several factors may have complicated their analysis of the effects of NT mRNA expression on cytotoxicity: (i) other enzymes such as 5-nucleotidase and cytidine deaminase are known to affect sensitivity to nucleoside analogs and were not measured, (ii) many nucleoside analogs are transported by more than one NT and (iii) there may not be a direct relationship between mRNA expression and protein levels.
Galmarini et al. (2002) studied the effects of hENT1 mRNA expression on efficacy of cytarabine treatment of acute myelogenous leukemia. They retrospectively studied the expression of dCK, 5′-nucleotidase, DNA polymerase, topoisomerase I and topoisomerase II, and hENT1 in blast cells of 123 patients who were treated with cytarabine. Decreased expression of hENT1 was associated with an increased risk of early relapse. Stam et al. (2003) studied 18 infants and 24 children with acute lymphoblastic leukemia to determine why infants were sensitive to cytarabine. Leukemic blasts from infants were threefold more sensitive to cytarabine than blasts from children (P=0.007). When mRNA levels of dCK, cytidine deaminase, pyrimidine nucleotidase I, deoxycytidylate deaminase and hENT1 were measured, decreased levels of dCK (P=0.026) but increased levels of hENT1 (P=0.001) were observed in infants whereas there were no differences in any of the other cytarabine-metabolizing enzymes between infants and children. It was concluded that increased hENT1 abundance was responsible for increased sensitivity of infant acute lymphocytic leukemia to cytarabine.
Genetic basis of drug resistance
Much of the current molecular understanding of the role of transport in the efficacy of drugs has come from experiments using cell lines and animal models. However, clinicians and patients are faced with uncertainties that an administered drug may not produce the desired effects. This lack of predictability in the context of therapeutic outcome has largely been attributed to genetic variations in patient population. The human genome across populations is largely identical (up to 99.9%) and much of the observed variability may lie in the remaining genome, prompting large-scale attempts to characterize these variations (Sachidanandam et al., 2001; Strohman, 2002). Over 90% of these variations reside in single nucleotide polymorphism (SNPs) and the remaining may be insertions, deletions and short tandem repeats. Insertion and deletion polymorphisms as well as SNPs in transporter proteins when present in coding or regulatory regions may produce phenotypes that influence drug uptake. The mere abundance of SNPs covering the entire genome over other classes of mutations described above are useful for fine mapping the genome as these occur at a frequency of 500–1000 bp (Brookes, 1999; Cargill et al., 1999; Damaraju et al., 2002).
While somatic mutations in transporters may explain some of the variability in response to drugs between individuals, the genetic basis of drug resistance can only be understood by studying germ line mutations across populations. Unlike somatic mutations, germ line variants are stable and heritable (Brookes, 1999). Huge efforts are underway to identify and characterize germ line mutations (most notably SNPs) in different ethnic populations. We provide here a summary of genetic polymorphism analyses for NTs, a rapidly emerging area, and discuss membrane proteins that are potentially implicated in the transport of nucleosides to address drug resistance in tumors. The drug resistance phenotype may arise as a result of over- or under-representation of transporters, through tissue specific regulation of expression (Goldman, 2002).
Drug resistance through gene redundancy
Multidrug resistance (MDR) proteins confer a resistant phenotype to a broad spectrum of drugs used in cancer chemotherapy (Wijnholds et al., 2000). These transporters are broadly classified as ATP-binding cassette proteins (ABC) and constitute a superfamily of proteins (Goldman, 2002). Methyl transferases, organic anion and cation transporters (OAT and OCT), oxidoreductase flavoproteins and NTs are major players in drug transport and metabolism in cells and may impart drug resistance phenotypes to cancer cells. Resistance to nucleoside analogs in tumors is thought to be a major problem in the clinic and several hypotheses have been proposed to account for altered NT activities. However, regulation through germ line polymorphisms of ENT and CNT genes has not been studied.
Redundancy at the level of substrate selectivity or specificity is common among the families of drug metabolism enzymes and transporters. Extensive gene duplication events may be responsible for the evolution of these gene families conferring a survival advantage to the species. As a result, conventional gene knockout or transgenic approaches have limited value in deciphering the role of these drug transporters in health and disease. For instance, MDR1 knockout mice showed normal development, an indication that this gene product may serve a detoxifying role only when challenged with xenobiotic compounds (Illmer et al., 2002). It is therefore not surprising that transporters that are otherwise well characterized through cloning and expression studies are yet to be defined for their role in disease etiology. The near ubiquitous expression of multiple transporters in various tissues further complicates genotype–phenotype correlations. Such complex and multigene associations are commonly described for polygenic diseases such as cancer, diabetes and cardiac abnormalities (Sachidanandam et al., 2001). The implication of several transporters in conferring drug resistance is analogous to the implication of several candidate genes contributing to a phenotype in polygenic diseases. Therefore, drug resistance mechanisms must be analysed in the context of the broad permeant selectivities of various transporters. For instance, nucleoside analogs are transported not only by the ENT and CNT families but also by the OAT and OCT families (Chen and Nelson, 2000). MRP5, a member of the MDR protein family, has been shown to transport nucleotide analogs. Inhibitors of OAT also inhibited the MDR5 protein function (Wijnholds et al., 2000).
Polymorphism analysis for drug transporters
Extensive sequencing of drug transport genes in healthy Japanese populations identified germ line variations that could potentially explain differences in drug efficacies in the context of this ethnic group (Iida et al., 2001; Hamajima et al., 2002). Detailed experimental evidence linking these polymorphisms to differential drug therapies is awaited. This group has deposited over 350 SNPs in genes encoding OAT 1, 2 and 3 and in the genes of the ABC family of transporters in the databases. Over 90% of these SNPs were novel, compared to existing SNPs in the dbSNP database (NCBI, National Center for Biotechnology Information at http://www.ncbi.nlm.nih.gov/SNP/index.html). SNPs deposited in databases, often from sequence alignments with the expressed sequence tags (EST) database or from sequencing a small sample of chromosomes, may require further validation by extensive sequencing. Also, a significant number of chromosomes across different racial groups need to be sequenced to identify true polymorphisms or to detect and correct sequencing errors inherent in the initial screens for SNPs. These quality control measures are essential prior to routine monitoring of SNPs in affected individuals for comparisons of therapeutic efficacy. The OAT and ABC family SNPs were predominantly in intronic and regulatory regions, and those that were exonic showed amino-acid changes (Iida et al., 2001; Hamajima et al., 2002). The high occurrence of SNPs in intronic and regulatory regions needs to be correlated with alternatively spliced transcripts and/or with the levels of expression in tissues to imply a role in disease. Studies on MDR1 gene polymorphisms on the expression of P-glycoprotein and effects of therapy outcome in acute myeloid leukemia patients were reported recently (Illmer et al., 2002). Coding region SNPs in MDR1 gene in exons 12, 21 and 26 were correlated with decreased or increased levels of protein expression in homozygous wild-type and the heterozygous variants, providing a clear example of genotype–phenotype associations. Evidence was presented for linkage disequilibrium of the SNPs in the 12, 21 and 26 exons, indicating that combined polymorphisms at these sites could affect the regulation of MDR1 expression (Illmer et al., 2002).
Comparative genomic studies on NTs
A recent comprehensive comparative genetic analysis of NT family members has led to the identification of the precise chromosomal locations for the genes encoding hENT1, hENT2 and hENT3 on chromosomes 6, 11 and 10, respectively (Coe et al., 1997; Sankar et al., 2002). The promoter and other regulatory regions have recently been defined for these genes (Sankar et al., 2002). hENT1 and hENT2 are believed to have originated through gene duplications, based on homology and structural organization of the genes, although they are located on two different chromosomes. On the other hand, the hENT3 gene is structurally very distinct from hENT1 and hENT2 (Sankar et al., 2002). Through comparative genomic studies, hENT4 has been identified on chromosome 7 and is believed to be evolutionarily distinct and more ancient compared to hENTs 1–3 (Acimovic and Coe, 2002). However, the OMIM (Online Mendelian Inheritance in Man, http://www.ncbi.nlm.nih.gov/Omim/) database has no entries linking hENT loci to any disease phenotype. The complexity of the observed expression of NTs in several tissues has been ascribed to splice variants of the major transporters, in particular hENT2 (Sankar et al., 2002). It remains to be established if alternative splicing of hENT transcripts is due to exposure of cryptic splice sites as a result of polymorphisms in genomic DNA.
NT gene regulation at the transcriptional and post-transcriptional level is much less studied due to lack of information about promoter and enhancer elements as well as 3′ noncoding regions. Recent work that defined these regions will stimulate further studies on the regulation of NTs (Sankar et al., 2002). NT gene regulation studies would corroborate earlier findings on the differential tissue abundance of NTs in normal and tumor tissues (Clarke et al., 2002). Tissue-specific expression of CNT and ENT proteins may also be modulated by p53 (Acimovic and Coe, 2002). A simple approach to correlate the expressed NT proteins to drug sensitivity or resistance in human cell lines was inconclusive, suggesting that NT expression and activity are influenced by other genes in a tissue-specific manner (Acimovic and Coe, 2002).
Polymorphisms in NT genes
The Human Membrane Transporter Database (HMTD, http://184.108.40.206/transporter/trans.html) has information on polymorphisms and tissue-specific expression of several transporters. Tissue-specific expression of hENT and hCNT proteins was found in this database, whereas data on polymorphisms for these genes were missing. A search for SNPs for these genes in the NCBI database revealed SNPs for hENT1 and hENT2 on chromosomes 6 and 11, respectively; a total of 24 SNPs were found, 12 in the mRNA-untranslated region and the remaining in introns. These putative hENT SNPs need to be validated by sequencing methodology using a large number of chromosomes. Recently, two coding region variants in hENT1 gene were identified in an ethnically diverse DNA samples, and functional characterization of these variants in S. cerevisiae indicated similar kinetics of uptake of nucleosides and nucleoside analogs as compared to the native hENT1 (Osato et al., 2003). Site-directed mutagenesis of the conserved glycine in transmembrane domain 5 of hENT1 abolished transport and NBMPR-binding activities without affecting membrane targeting and assembly of the hENT1 protein (SenGupta et al., 2002). Comparative genomic analysis of ENT proteins across species also indicated that transmembrane domains 3–6 (exons 4–6) are highly conserved and are involved in binding and transport of nucleosides (Sankar et al., 2002). Experimental studies with chimeras from ENTs of rats and humans have demonstrated the involvement of transmembrane domains 3–6 in interactions with inhibitors and/or permeants (Sundaram et al., 1998, 2001b; Yao et al., 2001, 2002).
A search for hCNT gene polymorphisms in the NCBI database indicated 61 SNPs in hCNT1, 2, and 3 genes in intronic and untranslated regions. Recently, Gray et al. (2003) reported the first systematic and functional analysis of CNT1 gene polymorphisms. These authors used denaturing HPLC and sequencing to identify coding region SNPs in 247 DNA samples from ethnically diverse populations. Functional analysis (binding and transport) of these variants was carried out in X. laevis oocytes (Gray et al., 2003) and several variants affecting CNT-mediated functions were identified. These observations lend credence to the popularly held belief that genetic variations may explain drug resistance mechanisms and could in principle account for the clinically observed therapeutic efficiency of nucleoside drugs in the affected populations.
SNP discovery combined with functional characterization of hENT and hCNT gene families should help address tissue-specific expression and inhibitor sensitivities in various tumor groups. Technologies are in place now for investigating NT promoter sequence variants using high-throughput reporter gene expression. Application of genomic technologies by way of association studies to otherwise well-characterized NTs should for the first time offer clues on the role of these proteins in health and disease. It would be interesting to see if one or more NT or other drug transporter gene polymorphisms contribute to drug uptake and therapeutic efficiencies in an additive manner, as one would suppose for a polygenic effect in a typical association study.
Abbruzzese JL, Grunewald R, Weeks EA, Gravel D, Adams T, Nowak B, Mineishi S, Tarassoff P, Satterlee W and Raber MN . (1991). J. Clin. Oncol., 9, 491–498.
Acimovic Y and Coe IR . (2002). Mol. Biol. Evol., 19, 2199–2210.
Baker CH, Banzon J, Bollinger JM, Stubbe J, Samano V, Robins MJ, Lippert B, Jarvi E and Resvick R . (1991). J. Med. Chem., 34, 1879–1884.
Baldwin S, Beal P, Yao SYM, King AE, Cass CE and Young JD . (2003). Pflugers Arch., June 28 (Epub ahead of print).
Baldwin SA, Mackey JR, Cass CE and Young JD . (1999). Mol. Med. Today, 5, 216–224.
Belanger K, Moore M, Baker SD, Dionne J, Maclean M, Jolivet J, Siu L, Soulieres D, Wainman N and Seymour L . (2002). J. Clin. Oncol., 20, 2567–2574.
Belt JA, Harper EH, Byl JA and Noel LD . (1992). Proc. Am. Assoc. Cancer Res., 33, 20.
Belt JA, Marina NM, Phelps DA and Crawford CR . (1993). Adv. Enzyme Regul., 33, 235–252.
Blum JL, Jones SE, Buzdar AU, LoRusso PM, Kuter I, Vogel C, Osterwalder B, Burger HU, Brown CS and Griffin T . (1999). J. Clin. Oncol., 17, 485–493.
Brockman RW, Schabel Jr FM and Montgomery JA . (1977). Biochem. Pharmacol., 26, 2193–2196.
Brookes AJ . (1999). Gene, 234, 177–186.
Burris III HA, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, Cripps MC, Portenoy RK, Storniolo AM, Tarassoff P, Nelson R, Dorr FA, Stephens CD and Von Hoff DD . (1997). J. Clin. Oncol., 15, 2403–2413.
Cabrita MA, Baldwin SA, Young JD and Cass CE . (2002). Biochem. Cell Biol., 80, 623–638.
Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, Shaw N, Lane CR, Lim EP, Kalyanaraman N, Nemesh J, Ziaugra L, Friedland L, Rolfe A, Warrington J, Lipshutz R, Daley GQ and Lander ES . (1999). Nat. Genet., 22, 231–238.
Carson DA, Wasson DB, Esparza LM, Carrera CJ, Kipps TJ and Cottam HB . (1992). Proc. Natl. Acad. Sci. USA, 89, 2970–2974.
Cass CE . (1995). Drug transport in antimicrobial and anticancer chemotherapy Georgopapadakou NH (ed). Marcel Dekker: New York, pp. 403–451.
Cass CE, Young JD and Baldwin SA . (1998). Biochem. Cell Biol., 76, 761–770.
Cass CE, Young JD, Baldwin SA, Cabrita MA, Graham KA, Griffiths M, Jennings LL, Mackey JR, Ng AM, Ritzel MW, Vickers MF and Yao SY . (1999). Pharm. Biotechnol., 12, 313–352.
Catapano CV, Chandler KB and Fernandes DJ . (1991). Cancer Res., 51, 1829–1835.
Chandrasena G, Giltay R, Patil SD, Bakken A and Unadkat JD . (1997). Biochem. Pharmacol., 53, 1909–1918.
Chen R and Nelson JA . (2000). Biochem. Pharmacol., 60, 215–219.
Chen SF, Cleaveland JS, Hollmann AB, Wiemann MC, Parks Jr RE and Stoeckler JD . (1986). Cancer Res., 46, 3449–3455.
Chun HG, Leyland-Jones B and Cheson BD . (1991). J. Clin. Oncol., 9, 175–188.
Clarke ML, Mackey JR, Baldwin SA, Young JD and Cass CE . (2002). Cancer Treat. Res., 112, 27–47.
Coe I, Zhang Y, McKenzie T and Naydenova Z . (2002). FEBS Lett., 517, 201–205.
Coe IR, Griffiths M, Young JD, Baldwin SA and Cass CE . (1997). Genomics, 45, 459–460.
Crawford CR, Patel DH, Naeve C and Belt JA . (1998). J. Biol. Chem., 273, 5288–5293.
Damaraju S, Sawyer M and Zanke B (eds) (2002). Genomic Approaches to Clinical Drug Resistance. Kluwer Academic Publisher: MA, USA.
Danhauser L, Plunkett W, Keating M and Cabanillas F . (1986). Cancer Chemother. Pharmacol., 18, 145–152.
Delicado EG, Sen RP and Miras-Portugal MT . (1991). Biochem. J., 279 (Part 3), 651–655.
Dimopoulos MA, Kantarjian H, Weber D, O'Brien S, Estey E, Delasalle K, Rose E, Cabanillas F, Keating M and Alexanian R . (1994). J. Clin. Oncol., 12, 2694–2698.
Ecker G . (2002). Curr. Opin. Invest. Drugs, 3, 1533–1538.
Ellison RR, Carey RW and Holland JF . (1967). Clin. Pharmacol. Ther., 8, 800–809.
Ellison RR, Holland JF, Weil M, Jacquillat C, Boiron M, Bernard J, Sawitsky A, Rosner F, Gussoff B, Silver RT, Karanas A, Cuttner J, Spurr CL, Hayes DM, Blom J, Leone LA, Haurani F, Kyle R, Hutchison JL, Forcier RJ and Moon JH . (1968). Blood, 32, 507–523.
Fideu MD and Miras-Portugal MT . (1992). Neurochem. Res., 17, 1099–1104.
Fideu MD and Miras-Portugal MT . (1993). Cell Mol. Neurobiol., 13, 493–502.
Flanagan SA and Meckling-Gill KA . (1997). J. Biol. Chem., 272, 18026–18032.
Fram RJ, Egan EM and Kufe DW . (1983). Leuk. Res., 7, 243–249.
Furth JJ and Cohen SS . (1968). Cancer Res., 28, 2061–2067.
Galmarini CM, Thomas X, Calvo F, Rousselot P, Rabilloud M, El Jaffari A, Cros E and Dumontet C . (2002). Br. J. Haematol., 117, 860–868.
Gati WP, Paterson AR, Belch AR, Chlumecky V, Larratt LM, Mant MJ and Turner AR . (1998). Leuk. Lymphoma, 32, 45–54.
Genini D, Adachi S, Chao Q, Rose DW, Carrera CJ, Cottam HB, Carson DA and Leoni LM . (2000). Blood, 96, 3537–3543.
Giles FJ, Cortes JE, Baker SD, Thomas DA, O'Brien S, Smith TL, Beran M, Bivins C, Jolivet J and Kantarjian HM . (2001). J. Clin. Oncol., 19, 762–771.
Giles FJ, Faderl S, Thomas DA, Cortes JE, Garcia-Manero G, Douer D, Levine AM, Koller CA, Jeha SS, O'Brien SM, Estey EH and Kantarjian HM . (2003). J. Clin. Oncol., 21, 1050–1056.
Goh LB, Mack P and Lee CW . (1995). Anticancer Res., 15, 2575–2579.
Goldman ID . (2002). Clin. Cancer Res., 8, 4–6.
Gourdeau H, Clarke ML, Ouellet F, Mowles D, Selner M, Richard A, Lee N, Mackey JR, Young JD, Jolivet J, Lafreniere RG and Cass CE . (2001). Cancer Res., 61, 7217–7224.
Graham KA, Leithoff J, Coe IR, Mowles D, Mackey JR, Young JD and Cass CE . (2000). Nucleos. Nucleot. Nucl. Acids, 19, 415–434.
Gray JH, Owen RP, Urban TJ and Giacomini KM . (2003). Clin. Phramacol. Therap., 73, P59.
Griffith DA and Jarvis SM . (1996). Biochim. Biophys. Acta, 1286, 153–181.
Griffiths M, Beaumont N, Yao SY, Sundaram M, Boumah CE, Davies A, Kwong FY, Coe I, Cass CE, Young JD and Baldwin SA . (1997a). Nat. Med., 3, 89–93.
Griffiths M, Yao SY, Abidi F, Phillips SE, Cass CE, Young JD and Baldwin SA . (1997b). Biochem. J., 328 (Part 3), 739–743.
Grove KL and Cheng YC . (1996). Cancer Res., 56, 4187–4191.
Grove KL, Guo X, Liu SH, Gao Z, Chu CK and Cheng YC . (1995). Cancer Res., 55, 3008–3011.
Guida L, Bruzzone S, Sturla L, Franco L, Zocchi E and De Flora A . (2002). J. Biol. Chem., 277, 47097–47105.
Gutierrez MM, Brett CM, Ott RJ, Hui AC and Giacomini KM . (1992). Biochim. Biophys. Acta, 1105, 1–9.
Gutierrez MM and Giacomini KM . (1993). Biochim. Biophys. Acta, 1149, 202–208.
Gutierrez MM and Giacomini KM . (1994). Biochem. Pharmacol., 48, 2251–2253.
Hamajima N, Saito T, Matsuo K, Suzuki T, Nakamura T, Matsuura A, Okuma K and Tajima K . (2002). J. Epidemiol., 12, 229–236.
Handa M, Choi DS, Caldeiro RM, Messing RO, Gordon AS and Diamond I . (2001). Gene, 262, 301–307.
Hande KR and Chabner BA . (1978). Cancer Res., 38, 579–585.
Heinemann V, Hertel LW, Grindey GB and Plunkett W . (1988). Cancer Res., 48, 4024–4031.
Heinemann V, Xu YZ, Chubb S, Sen A, Hertel LW, Grindey GB and Plunkett W . (1992). Cancer Res., 52, 533–539.
Hoff PM, Ansari R, Batist G, Cox J, Kocha W, Kuperminc M, Maroun J, Walde D, Weaver C, Harrison E, Burger HU, Osterwalder B, Wong AO and Wong R . (2001). J. Clin. Oncol., 19, 2282–2292.
Huang M, Wang Y, Cogut SB, Mitchell BS and Graves LM . (2003). J. Pharmacol. Exp. Ther., 304, 753–760.
Huang P, Chubb S and Plunkett W . (1990). J. Biol. Chem., 265, 16617–16625.
Huang P and Plunkett W . (1991). Mol. Pharmacol., 39, 449–455.
Huang P and Plunkett W . (1995). Semin. Oncol., 22, 19–25.
Hyde RJ, Cass CE, Young JD and Baldwin SA . (2001). Mol. Membr. Biol., 18, 53–63.
Iida A, Saito S, Sekine A, Harigae S, Osawa S, Mishima C, Kondo K, Kitamura Y and Nakamura Y . (2001). J. Hum. Genet., 46, 590–594.
Illmer T, Schuler US, Thiede C, Schwarz UI, Kim RB, Gotthard S, Freund D, Schakel U, Ehninger G and Schaich M . (2002). Cancer Res., 62, 4955–4962.
Ishikawa T, Sekiguchi F, Fukase Y, Sawada N and Ishitsuka H . (1998). Cancer Res., 58, 685–690.
Jennings LL, Hao C, Cabrita MA, Vickers MF, Baldwin SA, Young JD and Cass CE . (2001). Neuropharmacology, 40, 722–731.
Jones KW, Rylett RJ and Hammond JR . (1994). Brain Res., 660, 104–112.
Kaye SB . (1998). Br. J. Cancer, 78 (Suppl. 3), 1–7.
Kichenin K, Pignede G, Fudalej F and Seman M . (2000). Eur. J. Immunol., 30, 366–370.
King KM and Cass CE . (1994). Proc. Am. Assoc. Cancer Res., 35, A3436.
King KM, Damaraju V, Mowles DA, Lang T, Vickers MF, Young JD and Cass CE . (2003). Proc. Am. Assoc. Cancer Res., 44, 1500.
Kiss A, Farah K, Kim J, Garriock RJ, Drysdale TA and Hammond JR . (2000). Biochem. J., 352 (Part 2), 363–372.
Kufe DW, Major PP, Egan EM and Beardsley GP . (1980). J. Biol. Chem., 255, 8997–8900.
Kwong FY, Wu JS, Shi MM, Fincham HE, Davies A, Henderson PJ, Baldwin SA and Young JD . (1993). J. Biol. Chem., 268, 22127–22134.
Lamont EB and Schilsky RL . (1999). Clin. Cancer Res., 5, 2289–2296.
Lang TT, Selner M, Young JD and Cass CE . (2001). Mol. Pharmacol., 60, 1143–1152.
Lassota P, Kazimierczuk Z and Darzynkiewicz Z . (1994). Arch. Immunol. Ther. Exp. (Warsz), 42, 17–23.
Lee CW . (1994). Biochem. J., 300 (Part 2), 407–412.
Lee CW, Sokoloski JA, Sartorelli AC and Handschumacher RE . (1991). Biochem. J., 274 (Part 1), 85–90.
Lindemalm S, Liliemark J, Gruber A, Eriksson S, Karlsson MO, Wang Y and Albertioni F . (2003). Haematologica, 88, 324–332.
Loewen SK, Ng AM, Yao SY, Cass CE, Baldwin SA and Young JD . (1999). J. Biol. Chem., 274, 24475–24484.
Lu X, Gong S, Monks A, Zaharevitz D and Moscow JA . (2002). J. Exp. Ther. Oncol., 2, 200–212.
Lum PY, Ngo LY, Bakken AH and Unadkat JD . (2000). Cancer Chemother. Pharmacol., 45, 273–278.
Mackey JR, Baldwin SA, Young JD and Cass CE . (1998a). Drug Resist. Update, 310–324.
Mackey JR, Jennings LL, Clarke ML, Santos CL, Dabbagh L, Vsianska M, Koski SL, Coupland RW, Baldwin SA, Young JD and Cass CE . (2002). Clin. Cancer Res., 8, 110–116.
Mackey JR, Mani RS, Selner M, Mowles D, Young JD, Belt JA, Crawford CR and Cass CE . (1998b). Cancer Res., 58, 4349–4357.
Mackey JR, Yao SY, Smith KM, Karpinski E, Baldwin SA, Cass CE and Young JD . (1999). J. Natl. Cancer Inst., 91, 1876–1881.
Mata JF, Garcia-Manteiga JM, Lostao MP, Fernandez-Veledo S, Guillen-Gomez E, Larrayoz IM, Lloberas J, Casado FJ and Pastor-Anglada M . (2001). Mol. Pharmacol., 59, 1542–1548.
Murray AW . (1971). Annu. Rev. Biochem., 40, 811–826.
Neff T and Blau CA . (1996). Exp. Hematol., 24, 1340–1346.
Ngo LY, Patil SD and Unadkat JD . (2001). Am. J. Physiol. Gastrointest. Liver Physiol., 280, G475–G481.
Osato DH, Huang CC, Kawamoto M, Johns SJ, Stryke D, Wang J, Ferrin TE, Herskowitz I and Giacomini KM . (2003). Pharmacogenetics, 13, 297–301.
Osses N, Pearson JD, Yudilevich DL and Jarvis SM . (1996). Biochem. J., 317 (Part 3), 843–848.
Owens JK, Shewach DS, Ullman B and Mitchell BS . (1992). Cancer Res., 52, 2389–2393.
Parker WB, Bapat AR, Shen JX, Townsend AJ and Cheng YC . (1988). Mol. Pharmacol., 34, 485–491.
Parker WB, Shaddix SC, Chang CH, White EL, Rose LM, Brockman RW, Shortnacy AT, Montgomery JA, Secrist III JA and Bennett Jr LL . (1991). Cancer Res., 51, 2386–2394.
Parker WB, Shaddix SC, Rose LM, Shewach DS, Hertel LW, Secrist III JA, Montgomery JA and Bennett Jr LL . (1999). Mol. Pharmacol., 55, 515–520.
Parodi J, Flores C, Aguayo C, Rudolph MI, Casanello P and Sobrevia L . (2002). Circ. Res., 90, 570–577.
Paterson AR, Gati WP, Vijayalakshmi D, Cass CE, Mant MJ, Young JD and Belch AR . (1993). Proc. Am. Assoc. Cancer Res., 34, A84.
Patil SD and Unadkat JD . (1997). Am. J. Physiol., 272, G1314–G1320.
Pennycooke M, Chaudary N, Shuralyova I, Zhang Y and Coe IR . (2001). Biochem. Biophys. Res. Commun., 280, 951–959.
Pettitt AR, Clarke AR, Cawley JC and Griffiths SD . (1999a). Br. J. Haematol., 105, 986–988.
Pettitt AR, Sherrington PD and Cawley JC . (1999b). Br. J. Haematol., 106, 1049–1051.
Plunkett W, Huang P, Xu YZ, Heinemann V, Grunewald R and Gandhi V . (1995). Semin. Oncol., 22, 3–10.
Possinger K . (1995). Anticancer Drugs, 6 (Suppl. 6), 55–59.
Ritzel MW, Ng AM, Yao SY, Graham K, Loewen SK, Smith KM, Hyde RJ, Karpinski E, Cass CE, Baldwin SA and Young JD . (2001a). Mol. Membr. Biol., 18, 65–72.
Ritzel MW, Ng AM, Yao SY, Graham K, Loewen SK, Smith KM, Ritzel RG, Mowles DA, Carpenter P, Chen XZ, Karpinski E, Hyde RJ, Baldwin SA, Cass CE and Young JD . (2001b). J. Biol. Chem., 276, 2914–2927.
Ritzel MW, Yao SY, Huang MY, Elliott JF, Cass CE and Young JD . (1997). Am. J. Physiol., 272, C707–C714.
Ritzel MW, Yao SY, Ng AM, Mackey JR, Cass CE and Young JD . (1998). Mol. Membr. Biol., 15, 203–211.
Sachidanandam R, Weissman D, Schmidt SC, Kakol JM, Stein LD, Marth G, Sherry S, Mullikin JC, Mortimore BJ, Willey DL, Hunt SE, Cole CG, Coggill PC, Rice CM, Ning Z, Rogers J, Bentley DR, Kwok PY, Mardis ER, Yeh RT, Schultz B, Cook L, Davenport R, Dante M, Fulton L, Hillier L, Waterston RH, McPherson JD, Gilman B, Schaffner S, Van Etten WJ, Reich D, Higgins J, Daly MJ, Blumenstiel B, Baldwin J, Stange-Thomann N, Zody MC, Linton L, Lander ES and Altshuler D . (2001). Nature, 409, 928–933.
Sandoval A, Consoli U and Plunkett W . (1996). Clin. Cancer Res., 2, 1731–1741.
Sankar N, Machado J, Abdulla P, Hilliker AJ and Coe IR . (2002). Nucleic Acids Res., 30, 4339–4350.
Schachter JB, Yasuda RP and Wolfe BB . (1995). Cell Signal, 7, 659–668.
Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, Krook J, Zhu J and Johnson DH . (2002). N. Engl. J. Med., 346, 92–98.
Schmoll HJ, Buchele T, Grothey A and Dempke W . (1999). Semin. Oncol., 26, 589–605.
SenGupta DJ, Lum PY, Lai Y, Shubochkina E, Bakken AH, Schneider G and Unadkat JD . (2002). Biochem., 41, 1512–1519.
Seymour JF, Kurzrock R, Freireich EJ and Estey EH . (1994). Blood, 83, 2906–2911.
Shryock JC and Belardinelli L . (1997). Am. J. Cardiol., 79, 2–10.
Sirotnak FM, Chello PL, Dorick DM and Montgomery JA . (1983). Cancer Res., 43, 104–109.
Sokoloski JA, Lee CW, Handschumacher RE, Nigam A and Sartorelli AC . (1991). Leuk. Res., 15, 1051–1058.
Soler C, Felipe A, Casado FJ, Celada A and Pastor-Anglada M . (2000). J. Leukocyte Biol., 67, 345–349.
Soler C, Felipe A, Mata JF, Casado FJ, Celada A and Pastor-Anglada M . (1998). J. Biol. Chem., 273, 26939–26945.
Soler C, Garcia-Manteiga J, Valdes R, Xaus J, Comalada M, Casado FJ, Pastor-Anglada M, Celada A and Felipe A . (2001a). FASEB J., 15, 1979–1988.
Soler C, Valdes R, Garcia-Manteiga J, Xaus J, Comalada M, Casado FJ, Modolell M, Nicholson B, MacLeod C, Felipe A, Celada A and Pastor-Anglada M . (2001b). J. Biol. Chem., 276, 30043–30049.
Stadler WM, Kuzel T, Roth B, Raghavan D and Dorr FA . (1997). J. Clin. Oncol., 15, 3394–3398.
Stam RW, den Boer ML, Meijerink JP, Ebus ME, Peters GJ, Noordhuis P, Janka-Schaub GE, Armstrong SA, Korsmeyer SJ and Pieters R . (2003). Blood, 101, 1270–1276.
Strohman R . (2002). Science, 296, 701–703.
Sundaram M, Yao SY, Ingram JC, Berry ZA, Abidi F, Cass CE, Baldwin SA and Young JD . (2001a). J. Biol. Chem., 276, 45270–45275.
Sundaram M, Yao SY, Ng AM, Cass CE, Baldwin SA and Young JD . (2001b). Biochemistry, 40, 8146–8151.
Sundaram M, Yao SYM, Ng AML, Griffiths M, Cass CE, Baldwin SA and Young JD . (1998). J. Biol. Chem., 273, 21519–21525.
Talbot DC, Moiseyenko V, Van Belle S, O'Reilly SM, Alba Conejo E, Ackland S, Eisenberg P, Melnychuk D, Pienkowski T, Burger HU, Laws S and Osterwalder B . (2002). Br. J. Cancer, 86, 1367–1372.
Townsley CA, Chi K, Ernst DS, Belanger K, Tannock I, Bjarnason GA, Stewart D, Goel R, Ruether JD, Siu LL, Jolivet J, McIntosh L, Seymour L and Moore MJ . (2003). J. Clin. Oncol., 21, 1524–1529.
Vickers MF, Kumar R, Visser F, Zhang J, Charania J, Raborn RT, Baldwin SA, Young JD and Cass CE . (2002). Biochem. Cell Biol., 80, 639–644.
Vickers MF, Young JD, Baldwin SA, Mackey JR and Cass CE . (2000). Emerging Therap. Targets, 4, 515–539.
Vijayalakshmi D and Belt JA . (1988). J. Biol. Chem., 263, 19419–19423.
Visser F, Vickers MF, Ng AM, Baldwin SA, Young JD and Cass CE . (2002). J. Biol. Chem., 277, 395–401.
von der Maase H, Hansen SW, Roberts JT, Dogliotti L, Oliver T, Moore MJ, Bodrogi I, Albers P, Knuth A, Lippert CM, Kerbrat P, Sanchez Rovira P, Wersall P, Cleall SP, Roychowdhury DF, Tomlin I, Visseren-Grul CM and Conte PF . (2000). J. Clin. Oncol., 18, 3068–3077.
Wang J, Su SF, Dresser MJ, Schaner ME, Washington CB and Giacomini KM . (1997). Am. J. Physiol., 273, F1058–F1065.
Ward JL, Sherali A, Mo ZP and Tse CM . (2000). J. Biol. Chem., 275, 8375–8381.
Wijnholds J, Mol CA, van Deemter L, de Haas M, Scheffer GL, Baas F, Beijnen JH, Scheper RJ, Hatse S, De Clercq E, Balzarini J and Borst P . (2000). Proc. Natl. Acad. Sci. USA, 97, 7476–7481.
Wiley JS, Jones SP and Sawyer WH . (1983). Eur. J. Cancer Clin. Oncol., 19, 1067–1074.
Xie KC and Plunkett W . (1996). Cancer Res., 56, 3030–3037.
Yang SW, Huang P, Plunkett W, Becker FF and Chan JY . (1992). J. Biol. Chem., 267, 2345–2349.
Yao SY, Ng AM, Muzyka WR, Griffiths M, Cass CE, Baldwin SA and Young JD . (1997). J. Biol. Chem., 272, 28423–28430.
Yao SY, Ng AM, Sundaram M, Cass CE, Baldwin SA and Young JD . (2001). Mol. Membr. Biol., 18, 161–167.
Yao SY, Ng AM, Vickers MF, Sundaram M, Cass CE, Baldwin SA and Young JD . (2002). J. Biol. Chem., 277, 24938–24948.
Young JD, Cheeseman CI, Mackey JR, Cass CE and Baldwin SA . (2000). Gastrointestinal Transport, Molecular Physiology (Current Topics in Membranes, Vol. 50), Fambrough D, Benos D, Barrett K and Domowitz M. (eds). Academic Press: San Diego, CA, pp. 329–378.
Zimmerman TP, Mahony WB and Prus KL . (1987). J. Biol. Chem., 262, 5748–5754.
Supported by the Alberta Cancer Foundation, the Canadian Institutes of Health Research and the National Cancer Institute of Canada
About this article
Cite this article
Damaraju, V., Damaraju, S., Young, J. et al. Nucleoside anticancer drugs: the role of nucleoside transporters in resistance to cancer chemotherapy. Oncogene 22, 7524–7536 (2003). https://doi.org/10.1038/sj.onc.1206952
- nucleoside transporters
- anticancer nucleosides
- drug resistance
Synthesis and Antiproliferative Evaluation of 5′‐Arylchalcogenyl‐3‐(phenylselanyl‐triazoyl)‐thymidine
Polymeric modification of gemcitabine via cyclic acetal linkage for enhanced anticancer potency with negligible side effects
Antiproliferative and Antimigration Activities of Fluoro-Neplanocin A via Inhibition of Histone H3 Methylation in Triple-Negative Breast Cancer