HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 40, 30855-30863, October 6, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/40/30855    most recent M003988200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Drori, S. Articles by Assaraf, Y. G. Search for Related Content PubMed PubMed Citation Articles by Drori, S. Articles by Assaraf, Y. G. Social Bookmarking                      What's this?

Clustering of Mutations in the First Transmembrane Domain of the Human Reduced Folate Carrier in GW1843U89-resistant Leukemia Cells with Impaired Antifolate Transport and Augmented Folate Uptake^*

Stavit Drori^§, Gerrit Jansen^¶, Robert Mauritz^¶, Godefridus J. Peters^¶, and Yehuda G. Assaraf

From the  Department of Biology, The Technion, Haifa 32000, Israel and the ^¶ Department of Oncology, University Hospital Vrije Universteit, 1081 HV Amsterdam, The Netherlands

Received for publication, May 10, 2000, and in revised form, July 17, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have studied the molecular basis for the resistance of human CEM leukemia cells to GW1843, a thymidylate synthase inhibitor. GW1843-resistant cells displayed a ~100-fold resistance to GW1843 and methotrexate but were collaterally sensitive to the lipophilic antifolates trimetrexate and AG337, which enter cells by diffusion. These cells exhibited a 12-fold decreased methotrexate influx but surprisingly had a 2-fold decreased folic acid growth requirement. This was associated with a 4-fold increased influx of folic acid, a 3.5-fold increased steady-state level of folic acid, and a 2.3-fold expansion of the cellular folate pool. Characterization of the transport kinetic properties revealed that GW1843-resistant cells had the following alterations: (a) 11-fold decreased transport K_m for folic acid; (b) 6-fold increased transport K_m for GW1843; and (c) a slightly increased transport V_max for folic acid. Sequence analysis showed that GW1843-resistant cells contained the mutations Val-29  Leu, Glu-45  Lys, and Ser-46  Ile in the first transmembrane domain of the reduced folate carrier. Transfection of the mutant-reduced folate carrier cDNA into methotrexate transport null cells conferred resistance to GW1843. This is the first demonstration of multiple mutations in a confined region of the human reduced folate carrier in an antifolate-resistant mutant. We conclude that certain amino acid residues in the first transmembrane domain play a key role in (anti)folate binding and in the conferring of drug resistance.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Reduced folates are essential cofactors that function as one-carbon donors necessary for the biosynthesis of purines, thymidine, and glycine (1). Unlike prokaryotes, animal cells are devoid of de novo biosynthesis of folic acid and therefore meet their folate requirements by folate uptake from exogenous sources (1). Several transport systems have been described in various mammalian model cell lines that can accommodate transport of folates and their folate-based chemotherapeutic agents including methotrexate (MTX)¹ (2-4). (a) The reduced folate carrier (RFC) is the major uptake route that functions as a bidirectional anion exchanger (5, 6) with a high affinity (K_m = 0.3-5 µM) for reduced folates and MTX but low affinity (K_m = 200-400 µM) for folic acid (4, 5, 7). (b) Folate receptors, glycosylphosphatidylinositol membrane-anchored proteins that mediate the unidirectional uptake of folates, display a high affinity for folic acid and 5-methyltetrahydrofolate (K_D = 1-10 nM) but lower affinity (K_D = 10-300 nM) for other reduced folates and MTX (8-11). (c) An apparently independent transport system with optimal uptake activity at low pH, which recognizes folic acid, reduced folates and MTX with comparable affinities (K_m = 1-5 µM) (12-15).

The molecular cloning and the primary structures of the human, mouse, and hamster RFC genes have been described (16-21). Human RFC is an integral plasma membrane protein with 591 amino acids, is predicted to contain 12 transmembrane domains (TMDs), has a short N terminus and a long C terminus both of which are hydrophilic and presumed to reside within the cytoplasm (19, 22). The human RFC contains a single consensus site for N-linked glycosylation and undergoes an extensive glycosylation resulting in a broadly migrating protein with a molecular mass of 70-120 kDa (23, 24). Transfection of the human, mouse, and hamster RFC cDNAs into transport-deficient cells restored MTX transport and antifolate sensitivity (16-21).

Defective transport as a result of qualitative and/or quantitative alterations in the mammalian RFC has long been recognized as a frequent mechanism of antifolate resistance (25-30). Recently, several studies with mouse (31-36) and human leukemia cell lines (37, 38) displaying impaired antifolate transport have shown that the predominant mechanism underlying these antifolate resistance phenotypes is mutations in the RFC gene (31-38). Thus, single nucleotide changes and/or single amino acid substitutions in the RFC were identified that disrupt RFC expression or abolish antifolate transport (31-38). Importantly, some mutations primarily occurring in certain predicted TMDs of the rodent and human RFC brought about a major loss of antifolate transport while differentially preserving sufficient reduced folate and folic acid uptake to support cellular growth (32-36, 38).

Over the past decade several rationally designed novel antifolates were introduced (39). For example, Tomudex (Raltritrexed, ZD1694), which has been recently approved against colorectal cancer (40), and GW1843, which is undergoing clinical evaluation as an anticancer drug (41), are potent thymidylate synthase (TS) inhibitors (39). Like other folate analogues, these antifolates enter cells via RFC, which serves as the major route of entry for folates and antifolates (39). Unlike various antifolates including MTX and ZD1694, GW1843 has an unusual and bulky methyl-oxobenzo-quinazoline structure (Fig. 1). Nevertheless, GW1843 is yet the best RFC transport substrate with a K_i at the submicromolar range (8, 39). Hence, we have here studied the alterations that occur in the RFC in human leukemia cells selected for resistance to GW1843. We show here for the first time the clustering of three mutations in TMD1 of the human RFC. We further demonstrate that these mutations selectively impair antifolate transport via decreased substrate affinity, while markedly augmenting folate cofactor uptake. In contrast, these mutations had little effect on antifolate and folate transport V_max. Consequently, GW1843-resistant cells display expanded folate pools and exhibit a high level resistance to various antifolates. This establishes TMD1 of the human RFC as a key determinant for (anti)folate binding and as a major locus for antifolate drug resistance.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- [3',5',7'-³H]MTX (23.4 Ci/mmol) and [3',5',7',9-³H]folic acid (25.5 Ci/mmol) obtained from Moravek Biochemicals (Brea, CA) were purified prior to use by thin layer chromatography (42) and kept at 80 °C. MTX was from Teva Pharmaceuticals Ltd., and folic acid dl-leucovorin (calcium salt) and were purchased from Sigma. G418 (700-750 µg of active drug/mg) was obtained from Life Technologies, Inc. The non-radioactive antifolate drugs were generous gifts from the following sources: GW1843U89 (43), Dr. G. K. Smith (Glaxo Wellcome); ZD1694 (44), Dr. A. Jackman (Institute for Cancer Research, Sutton, UK); PT523 (45), Dr. W. T. McCulloch (Sparta Pharmaceuticals); DDATHF (46) and LY231514 (MTA) (47) (Lilly); AG2034 (48), Dr. T. J. Boritzki (Agouron Pharmaceuticals); trimetrexate (49), Dr. D. Fry, Parke-Davis.

Cell Cultures and GW1843U89 Selection-- Human CCRF-CEM leukemia cells and their MTX transport-defective CEM/MTX subline (50) were grown in RPMI 1640 medium (containing 2.3 µM folic acid; Biological Industries, Beth Haemek, Israel) supplemented with 10% fetal calf serum (Life Technologies, Inc.), 2 mM glutamine, 100 units/ml penicillin G (Sigma), and 100 µg/ml streptomycin sulfate (Sigma). The growth medium of CEM/MTX cells contained 1 µM MTX (50). The GW70 subline was established by exposure of parental CEM cells to gradually increasing concentrations of GW1843 in normal growth medium. This selection was initiated at 1.5 nM and terminated at 70 nM GW1843, and the cells were therefore termed GW70. GW70/LF cells were established by gradual deprivation of folic acid from the growth medium (2.3 µM) and terminated at 2 nM folic acid. Mouse L1210 leukemia cells and their MTX-resistant L1210/MTX^rA subline with defective MTX transport (29-31) were grown in RPMI 1640 as detailed above. The medium of MTX^rA cells contained 1 µM MTX.

Growth Assays-- We have determined the leucovorin and folic acid growth requirements as well as the sensitivity to MTX of human leukemia CEM and GW70 cells and their clonal transfectants in transport null mouse L1210/MTX^rA cells (29-31). Exponentially growing cells in folic acid-free growth medium supplemented with 10% dialyzed fetal calf serum were first grown for 1-2 weeks in folate-free RPMI 1640 medium supplemented with 10% dialyzed fetal calf serum (Beth Hemeek, Israel) containing 200 µM glycine, 100 µM adenosine, and 10 µM thymidine (GAT) and G418 (750 µg active drug/ml, for transfectants). Thereafter, cells were washed twice with serum-free RPMI 1640 medium lacking folic acid in order to eliminate GAT. Cells were then seeded into 96-well plates (1 × 10⁵ -2 × 10⁵ cells/ml; 0.15 ml/well) at the following densities: L1210 and L1210/MTX^rA at 1.5 × 10⁴ and 2.3 × 10⁴, respectively, and CEM, GW70, and GW70/LF cells were seeded at 3 × 10⁴ cells/well. Thereafter, cells were exposed continuously to various concentrations of leucovorin or folic acid for 72 h, following which cell numbers were determined by hemocytometer count of viable cells using trypan blue exclusion. In the case of MTX, growth inhibition assays were performed in folic acid-containing medium supplemented with 10% fetal calf serum. The 50% inhibitory concentration (IC₅₀) is defined as the drug dose at which cell growth was inhibited by 50% relative to untreated controls. EC₅₀ is defined as the folate cofactor concentration necessary to produce 50% of maximal cell growth.

Southern and Northern Analyses-- High molecular weight genomic DNA was extracted from parental CEM cells and its various sublines. DNA was digested with EcoRI, fractionated by electrophoresis on 0.8% agarose gels, transferred to a Zetaprobe (Bio-Rad) nylon membrane, and UV cross-linked. RFC gene copy number was determined by Southern blot analysis using a ³²P-labeled (51) human RFC cDNA probe isolated from a cDNA library from CEM-7A cells (52). Total RNA was size-fractionated on 1% agarose gels containing formaldehyde, blotted onto a Zetaprobe nylon membrane, and UV cross-linked. Human RFC cDNA and -actin cDNA (53) were labeled by random hexamer priming (51). Southern and Northern blot hybridizations and post-hybridization washes were carried out under high stringency conditions as described previously (38). Southern and Northern blots were quantitated with a BAS 1000 Bio-Imaging Analyzer (Fujix).

Recombinant Human RFC Antiserum and Western Analysis-- The recombinant human RFC antiserum kindly provided by Dr. L. H. Matherly (Barbara Ann Karmanos Cancer Institute, Detroit, MI) was generated by subcloning the complete coding sequence of the human RFC into a glutathione S-transferase (GST) fusion vector. After transformation of Escherichia coli cells, the GST-RFC fusion proteins were purified by glutathione-Sepharose affinity chromatography. Anti-GST-RFC antiserum was raised in rabbits using the purified GST-RFC fusion protein as antigen.

To quantitate the levels of RFC expression, plasma membrane vesicles were isolated in the presence of seven protease inhibitors as described previously (24). Plasma membrane proteins were resolved by electrophoresis on 10% polyacrylamide gels containing SDS and electroblotted onto a PVDF nylon membrane (Millipore). The blots were then blocked overnight at 4 °C in TBST buffer (150 mM NaCl, 0.5% Tween 20, 10 mM Tris-Cl, pH 8.0) containing 5% dry milk. The blots were then reacted with an antiserum (1:1300) to human RFC for 3 h at room temperature in TBST buffer. Blots were then rinsed briefly with TBST, after which three washes each of 15 min were performed in the same buffer. The blots were then reacted with horseradish peroxidase-linked anti-rabbit antiserum (1:2500; Amersham Pharmacia Biotech) in TBST for 1 h at room temperature, after which enhanced chemiluminescence detection was performed according to the manufacturers' instructions (Amersham Pharmacia Biotech).

Genomic PCR-SSCP Assay-- Genomic PCR-single strand conformational polymorphism (SSCP) analysis of the six exons of the human RFC gene was undertaken as described elsewhere.²

Site-directed Mutagenesis and RFC cDNA Transfections-- An RFC cDNA clone containing the entire coding region (hRFC1) was directionally cloned into pCDNA3(+) at the BamHI-XhoI site (Invitrogen). The various point mutations and the single nucleotide polymorphic variations were inserted in the native RFC cDNA by site-directed mutagenesis (QuickChange, Stratagene). Mouse L1210/MTX^rA leukemia cells that lack functional RFC transport activity (29-31) were electroporated (300 V, 250-330 microfarads) with 40 µg of nonlinearized pCDNA3(+) harboring the native or mutant RFC cDNAs in serum-free RPMI 1640 medium in a final volume of 800 µl. Cells were then immediately diluted in prewarmed serum-containing growth medium (10 ml), allowed to recover for 32 h, adjusted to 10⁵-2 × 10⁵ cells/ml in medium containing G418 (750 µg of active drug/ml), and then distributed into 96-well plates at approximately 2 × 10⁴-4 × 10⁴ cells/well. After 10-20 days of incubation at 37 °C, individual G418-resistant clones were verified microscopically, picked, and expanded for further analyses.

Transport Studies-- (Anti)folate transport measurements were performed essentially as previously described (38). Briefly, cells were harvested in the mid-log phase, washed twice with HEPES-buffered saline solution (HBSS, pH 7.4), and suspended to a density of 2 × 10⁷ cells/ml in HBSS, pH 7.4, at 37 °C. Uptake of [³H]MTX (specific activity 0.5 Ci/mmol) and [³H]folic acid (specific activity 1 Ci/mmol) was measured at an extracellular concentration of 2 µM in 1 ml of cell suspension at 37 °C. For transport studies with [³H]folic acid, HBSS buffer was supplemented with 5 µM trimetrexate in order to block folic acid reduction (54). At selected transport time intervals from 0.5 to 30 min, uptake was terminated by the addition of 10 ml of ice-cold HBSS, after which centrifugation and cell wash with another 10 ml ice-cold HBSS were performed. The final cell pellet was then counted for radioactivity.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

GW1843U89 is an extremely potent folate-based inhibitor of TS (43) with a methyl-oxobenzo-quinazoline structure (Fig. 1). Despite the bulky structure of GW1843U89, the human RFC displays the highest affinity for this compound as a transport substrate (K_i = 0.8 µM), even when compared with the high affinity reduced folate substrates (8). Thus, to explore whether structural alterations in the human RFC could impair GW1843 uptake and result in resistance to this novel antifolate, a human GW70 leukemia cell line was established by stepwise selection of CCRF-CEM cells in gradually increasing GW1843 concentrations (see "Experimental Procedures"). GW70 cells displayed a ~100-fold resistance to GW1843 and MTX (Fig. 2, A and B, and Table I). Furthermore, GW70 cells also exhibited up to 150-fold cross-resistance to various antifolates that depend on RFC for their cellular uptake including ZD1694, ZD9331, LY231514 (MTA), DDATHF, and PT523 (Table II). In contrast, GW70 cells were 2-fold collaterally sensitive to AG337, a lipophilic TS inhibitor which presumably enters cells by diffusion. Similarly, GW70 cells retained parental sensitivity to trimetrexate, a lipid-soluble dihydrofolate reductase (DHFR) inhibitor that also enters cells by diffusion (Table II). GW70 cells showed neither a change in the activities of the target enzyme, TS, nor in folylpoly--glutamate synthetase (FPGS), folylpoly--glutamate hydrolase (FPGH, Table III), or DHFR (55). Surprisingly, however, GW70 cells had a 2-fold decreased folic acid growth requirement as compared with their parental cells (Fig. 2C and Table I). Thus, since these results were consistent with an altered RFC-mediated (anti)folate transport, we measured the transport of MTX and folic acid. Indeed, GW70 cells had a 12-fold decrease in both the influx of MTX (Table III) and the steady-state levels (Fig. 3A). In contrast, and consistent with their decreased folic acid growth requirement, GW70 cells had a 4-fold increased influx of folic acid (Fig. 3B and Table III) and displayed a 3.5-fold increase in the steady-state transport levels of folic acid (Fig. 3B); the latter experiments were performed under conditions in which folic acid reduction was blocked by trimetrexate (54).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Structural formulas of folic acid and the antifolates methotrexate and GW1843U89.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Antifolate growth inhibition and folic acid growth requirement in parental CEM cells and their GW70 subline. Exponentially growing parental CEM (circles) and GW70 cells (squares) in folic acid-containing medium were exposed for 3 days to various concentrations of GW1843 (A) or MTX (B), after which viable cell counting was performed as detailed under "Experimental Procedures." To determine the growth requirements of these cell lines for folic acid (C), cells were grown first for 1-2 weeks in GAT-containing folate-free RPMI 1640 medium (see "Experimental Procedures") supplemented with dialyzed fetal calf serum. Thereafter, cells were exposed for 3 days to various folic acid concentrations followed by viable cell counting. Values depicted are means ± S.D. of 4-6 independent experiments.

View this table: [in this window] [in a new window]   Table III Activities of TS, FPGS, FPGH, and folate transport activity in CEM/WT and their GW1843-resistant subline Results are the means ± S.D. of 3-5 experiments. FPGH, folylpoly--glutamate hydrolase.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Transport of [3H]MTX and [3H]folic acid into parental CEM cells and their GW70 subline. Exponentially growing parental CEM (open circles) and GW70 cells (solid circles) were washed twice with HBSS, adjusted to 2 × 107 cells/ml, and the transport of [3H]MTX (A) and [3H]folic acid (B) was determined at an extracellular radiolabel concentration of 2 µM as detailed under "Experimental Procedures." Transport experiments with [3H]folic acid were performed in the presence of 5 µM trimetrexate in order to block folic acid reduction (54).

The RFC-mediated impaired uptake of MTX in GW70 cells made transport kinetic measurements difficult. To promote RFC overproduction as recently shown (38), GW70 cells were gradually deprived of folic acid in their growth medium. This resulted in the establishment of GW70-LF (LF indicates low folate) cells which required only nanomolar folic acid concentrations for normal growth. GW70-LF cells displayed 5-fold RFC gene amplification (Fig. 4A) and ~50-fold overexpression of both the 3.1-kilobase pair RFC mRNA (Fig. 4B) and the ~80-kDa RFC protein (Fig. 4C), relative to their parent GW70 cells. The truncated ~43-kDa RFC protein overproduced in GW70/LF cells was encoded by the 2-kilobase pair RFC mRNA which resulted from an alternative splicing as we have recently shown in CEM-7A cells (52). Consequently, GW70/LF cells had a 29-fold increase in MTX influx and 10-fold increase in folic acid influx, relative to GW70 cells (Table III). In contrast, GW70 cells had a 3-fold decrease in the principal 3.1-kilobase pair RFC mRNA as compared with parental CEM cells (Fig. 4B). Consistently, Western blot analysis with purified plasma membrane fraction showed that GW70 cells had a 3-fold decrease in the ~80-kDa RFC protein (Fig. 4C). Thus, since the modest decrease in RFC protein expression in GW70 cells could not account for the 100-fold resistance to GW1843 and the prominently decreased transport of MTX, we determined the transport kinetic parameters for both MTX and folic acid. GW70 and GW70/LF cells had a 3-fold increase in the transport K_m for MTX (Tabe IV), as well as a 6- and 3-fold increase in the transport K_i values for GW1843 and ZD1694, respectively (Table V). In contrast, GW70 and GW70/LF cells displayed an 11-fold decreased transport K_m for folic acid (Table IV) and a 3-fold decreased transport K_i for the reduced folate cofactors, 5-formyltetrahydrofolate (leucovorin) and 5-methyltetrahydrofolate (Table V). Strikingly, GW70 cells had a slightly increased transport V_max for folic acid, while retaining 40% of parental MTX transport V_max (Table IV). Furthermore, relative to GW70 cells, GW70/LF cells showed a 52- and 12-fold increased transport V_max for MTX and folic acid, respectively (Table IV). Consequent to their prominently increased folic acid transport, GW70 and GW70/LF cells (growing in normal growth medium containing 2.3 µM folic acid) had a 2.3- and 4.3-fold expansion of the intracellular folate pools, respectively, relative to parental CEM cells (Table VI). Interestingly however, GW70/LF cells growing in medium containing 2 nM folic acid could normally grow with an intracellular folate pool that constituted only 11% of that present in parental CEM cells (Table VI).

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Southern (A), Northern (B), and Western (C) blot analyses of the RFC gene and its expression in CEM, GW70, and GW70/LF cells. A, genomic DNA (10 µg) was digested with EcoRI, fractionated on 0.8% agarose gels, transferred to a nylon membrane, and hybridized with a 32P-labeled human RFC cDNA probe. B, total RNA (20 µg) was fractionated on a denaturing 1% agarose gel containing formaldehyde, blotted onto a Zetaprobe nylon membrane, and hybridized with a 32P-labeled human RFC (upper panel) and -actin (lower panel) cDNA probes. C, proteins (50 µg) of plasma membrane-enriched fraction were resolved by electrophoresis on 10% polyacrylamide gels, electroblotted onto a PVDF nylon membrane, and reacted with a polyclonal serum anti-recombinant human RFC as described under "Experimental Procedures." A-C, lanes a-c were wild type CEM, GW70, and GW70/LF cells, respectively. Note that the RFC antiserum identifies in GW70/LF cells two overexpressed RFC proteins, native RFC with a molecular mass of ~85 kDa as well as a 40-45-kDa truncated RFC that is the product of an alternative splicing (52). kb, kilobase pair.

As these results were strongly indicative of a structurally altered RFC, we developed a genomic polymerase chain reaction-single strand conformational polymorphism assay and examined DNAs from GW70, GW70/LF cells, and parental CEM cells (Fig. 5). GW70 (Fig. 5, A and C, lane b) and GW70/LF cells (Fig. 5, A and C, lane c) displayed an identical electrophoretic mobility pattern in exon 2 of the RFC gene which was drastically different (see arrowheads) from that observed with parental CEM cells (Fig. 5, A and C, lane a). Sequence analysis revealed that exon 2 in both GW70 and GW70/LF cells contained three mutations including a G to C, G to A, and G to T at nucleotide positions 179, 227, and 231, respectively (Table VII; nucleotide numbering was according to Prasad et al. (19)). These mutations resulted in the following amino acid substitutions: Val-29 to Leu, Glu-45 to Lys, and Ser-46 to Ile, respectively (Table VII). Additionally, whereas parental CEM cells (Fig. 5, B and D, lane a) and GW70 cells (Fig. 5, B and D, lane b) had a normal electrophoretic mobility pattern for exon 3, GW70/LF cells (Fig. 5, B and D, lane c) displayed a markedly altered pattern (see arrowheads). Sequence analysis identified an additional G to C mutation at nucleotide 356 in exon 3, resulting in a shift of Asp-88 to His (Table VII). Hence, according to the recently published membrane topology of the human RFC (22), the cluster of the three mutations in GW70 cells Val-29  Leu, Glu-45  Lys, and Ser-46  Ile resides in TMD1, whereas the additional Asp-88  His mutation in GW70/LF maps to intracellular loop 2 (Fig. 6).

View larger version (66K): [in this window] [in a new window]   Fig. 5.   Genomic PCR-SSCP analysis of RFC exons 2 and 3 from parental CEM cells and their transport-altered sublines, GW70, and GW70/LF cells. Genomic DNA (0.1 µg) from CEM cells (lane a in A-D), their sublines GW70 (lane b in A-D), and GW70/LF (lane c in A-D) was PCR-amplified in the presence of [32P]dATP using intronic and exonic oligonucleotide primers targeting exon 2 (A and C) and exon 3 (B and D). The PCR products were resolved by electrophoresis on pre-casted commercial SSCP gels containing 10% polyacrylamide (ExcelGel, Amersham Pharmacia Biotech) (A and B) or laboratory-made glycerol lacking (D) or containing gels (C), as detailed under "Experimental Procedures." Bands with wild type electrophoretic mobility are denoted by thin lines (A) and arrows (B-D), whereas those with aberrant electrophoretic mobility are marked by arrowheads. Note that in exon 2 of RFC four normal bands (arrows) were observed with wild type CEM cells (C, lane a), whereas GW70 (lane b in C) and GW70/LF cells (lane c in C) were so altered that these bands could no longer be detected.

View larger version (28K): [in this window] [in a new window]   Fig. 6.   The clustering of mutations in the first transmembrane domain of the human RFC in GW70 and GW70-LF cells. The predicted secondary structure of the human RFC was based on a hydropathy analysis of Prasad et al. (19) as well as on recent membrane topology studies by Ferguson and Flintoff (22). Mutated amino acid residues are shown as solid circles, and the polymorphic amino acid residue 27 that is Arg in GW70 and GW70/LF is shown as a gray circle.

Recently we identified a polymorphism at amino acid 27.² Thus, whereas parental CEM cells were heterozygous thus containing two RFC species, one with Arg-27 (G at nucleotide 174) and another with His-27 (A at nucleotide 174), GW70 cells lost this heterozygosity and became homozygous for the Arg-27 allele. This loss of heterozygosity was consistent with the genomic rearrangements observed in the Southern analysis with DNA from both GW70 and GW70/LF cells when compared with parental CEM cells (Fig. 4A, compare lanes b and c with lane a).

It is well established that various organic and inorganic anions including chloride inhibit RFC-mediated transport of folates and antifolates (5, 6). Conversely, it was recently shown that structurally altered mouse RFC (33) and human RFC (38), both of which harbor a Glu-45  Lys substitution, resumed MTX transport activity in a chloride-dependent manner. Thus, since the Glu-45  Lys substitution was also present among the mutation triplet in GW70 and GW70/LF cells, we examined the anion dependence of the MTX transport. Like parental CEM cells (38), CEM-7A cells overexpressing the wild type RFC (52) showed a concentration-dependent inhibition of MTX transport as the chloride concentration was increased (Fig. 7). In contrast, in chloride-free buffer, GW70/LF cells lost 90% of their MTX transport activity (Fig. 7). However, MTX transport activity increased in a chloride concentration-dependent manner in GW70/LF cells; 100% of MTX transport activity was recovered at physiological chloride concentrations (Fig. 7). Thus, GW70/LF cells had an obligatory chloride dependence in order to maintain maximal MTX transport activity.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Chloride dependence of [3H]MTX influx in GW70/LF and CEM-7A cells. GW70/LF cells (107 cells/assay) that overexpress an altered RFC and CEM-7A cells (4 × 106 cells/assay) that overproduce native RFC (52) were harvested by centrifugation and washed twice with an anion-free buffer (20 mM HEPES, 225 mM sucrose, pH 7.4, with KOH). Subsequently, cells were resuspended in the anion-deficient buffer or in a buffer in which the anion-free buffer was mixed isotonically with HEPES-buffered saline (20 mM HEPES, 140 mM NaCl, 5 mM KCl, 2 mM MgCl2, 7 mM D-glucose, pH 7.4, with NaOH) to achieve final extracellular chloride concentrations of 0, 10, 25, 50, 100, and 150 mM. Linear rates of [3H]MTX uptake were measured for 1.5 min (CEM-7A) and 3 min (GW70/LF) at an extracellular concentration of 2 µM as described under "Experimental Procedures."

In order to determine the role of the three RFC mutations identified in GW70 cells on GW1843 resistance, we performed human RFC cDNA transfections into transport-null mouse leukemia MTX^rA cells. Prior to these transfection studies, we took into consideration the fact that we recently studied the human RFC mutation Glu-45  Lys (38). This mutation conferred a high level resistance to MTX and cross-resistance to several antifolates including GW1843 (38). Thus, in the present transfection studies we focused on the other RFC mutations identified in GW70 cells. Following transfection, only clonal variants that expressed similar human RFC mRNA levels were selected for further studies (Fig. 8). RFC cDNA harboring the conservative substitution Val-29  Leu did not cause any appreciable resistance to GW1843 relative to native RFC cDNA transfection (Table VIII). Likewise, transfection of the double mutant Glu-45  Lys/Ser-46  Ile did not result in a significant GW1843 resistance (Table VIII). Importantly however, transfection of an RFC cDNA containing the triple mutant Val-29  Leu/Gly-45  Lys/Ser-46  Ile conferred a prominent level of resistance to GW1843 (Table VIII). Moreover, transfection of an RFC cDNA harboring only the single mutation Ser-46Ile also imparted a high level resistance to GW1843 (Table VIII). Thus, it is the combination of the three RFC mutations Val-29  Leu/Glu-45  Lys/Ser-46  Ile or the single Ser-46  Ile mutation that could confer a significant level of resistance to GW1843.

View larger version (49K): [in this window] [in a new window]   Fig. 8.   Northern blot analysis of MTX transport null (MTXrA) clonal variants transfected with the native and mutated RFC cDNAs. MTX transport-deficient murine MTXrA cells were transfected with a native human RFC cDNA (lane a), or with mutated RFC cDNAs harboring Ile-46 (lane b), Lys-45/Ile-46 (lane c), or Leu-29/Lys-45/Ile-46 (lane d), after which G418-resistant clones were isolated, expanded, and total RNA extracted as described under "Experimental Procedures." RNA (20 µg) was resolved by electrophoresis on denaturing 1.5% agarose gels containing formaldehyde and then blotted onto a Zetaprobe nylon membrane. The blots were then hybridized with 32P-oligolabeled human RFC (upper panel) or human -actin (lower panel) and washed under high stringency conditions that did not allow for the detection of the endogenous mouse RFC mRNA.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The present paper is the first to demonstrate the clustering of three point mutations in the RFC in tumor cells stepwise selected for antifolate resistance. Thus, RFC from GW70 cells contained three single amino acid substitutions Val-29  Leu/Glu-45  Lys/Ser-46  Ile, all of which mapped to TMD1 (Fig. 6). Transfection of this triple mutant RFC cDNA or the Ser-46  Ile mutant conferred upon transport-null cells resistance to GW1843. The targeting of single amino acid residues in TMD1 has been observed in several independent studies with MTX-resistant human and mouse leukemia cells: (a) Gly-44  Glu substitution of the mouse RFC has been observed in two independent MTX-resistant murine L1210 leukemia clones that were obtained by chemical mutagenesis (36). These Gly-44  Glu mutants lost their MTX transport ability (i.e. 1% that of parental cells). (b) The very same Glu-45  Lys substitution (38) present in GW70 cells has been detected in human CEM-MTX cells that were isolated by stepwise selection to MTX (50). This human Glu-45  Lys mutation disrupted MTX influx. Furthermore, Glu-45  Lys mutation has also been identified in two independent MTX-resistant murine L1210 clones (C8 and D13) that were obtained by chemical mutagenesis (36). This Glu-45  Lys replacement was associated with up to 96% loss of parental MTX transport V_max (33, 36). Furthermore, this Glu-45  Lys substitution decreased the transport K_i for folic acid by 7-fold, while conversely increasing the transport K_m for MTX by 7-fold (33). (c) Ser-46  Asn has been identified in a murine MTX-resistant clone (G1a) obtained by chemical mutagenesis (32). In this mutant, the transport V_max values for MTX and 5-methyltetrahydrofolate were decreased by 44- and 6.6-fold, respectively (32). This led to the selective conservation of sufficient folate cofactor uptake to support growth while losing MTX transport (32). (d) Ile-48  Phe mutation has been found in murine RFC in L1210/D3 cells that were established by stepwise selection to the glycinamide ribonucleotide transformylase (GARTFase) inhibitor, 5,10-dideaza-5,6,7,8-tetrahydrofolic acid (DDATHF) (34, 35). L1210/D3 cells that also contained an additional mutation, Trp-105  Gly, displayed a highly enhanced transport of folic acid due to a ~20-fold decrease in the transport K_m for folic acid (34, 35). Thus, the cluster of the three mutations in TMD1 of RFC in GW70 cells and the various mutations identified in TMD1 of human and mouse RFC strongly suggest that the region near amino acids 44-48 takes active part in the formation of the (anti)folate-binding site. This region can therefore be regarded as an important mutation determinant for the acquisition of transport-based antifolate anticancer drug resistance.

We have previously found that polymorphism exists at amino acid residue 27 of the human RFC ². Thus, using genomic PCR-SSCP analysis, we previously identified three equally distributed polymorphic groups as follows: (a) RFC homozygous for Arg-27, (b) RFC homozygous for His-27, and (c) heterozygous RFC with both Arg-27 and His-27. Hence, parental CEM cells were found heterozygous and thus contained the two RFC variants, both the Arg-27 and the His-27. In contrast, both GW70 and GW70/LF cells lost their His-27 allele and became homozygous for Arg-27. This genomic loss of RFC heterozygosity is supported by the Southern blot analysis that showed that the RFC gene in GW70 and GW70/LF cells had undergone some genomic rearrangements (Fig. 4A). Thus, it is likely that during the multiple step selection to GW1843, a growth advantage was acquired by a clone that lost the wild type His-27 allele which promotes the uptake of GW1843, while preserving the advantageous triple mutant RFC allele that imparted both antifolate resistance and allowed for high levels of folic acid uptake (Fig. 3). This likely scenario is supported by a recent report in which MTX-resistant mouse leukemia L1210-G1a cells were found homozygous for Ser-46  Asn mutation that conferred a high level resistance to MTX but at the same time allowed for sufficient folic acid to be taken up that could support cell growth (32).

Several common functional features are shared by GW70 and human CEM-MTX cells (38). 1) Like GW70 cells, CEM-MTX cells contained a Glu-45  Lys mutation in TMD1 (38). 2) Like GW70 cells, CEM-MTX cells displayed a markedly increased uptake of folic acid (38). 3) Like GW70 cells, CEM-MTX cells were absolutely dependent on physiological chloride concentrations in order to maintain maximal MTX transport (38). However, when the Glu-45  Lys mutation was present in the context of Val-29  Leu and Ser-46  Ile as observed in TMD1 of GW70 cells, this altered RFC acquired different kinetic features from those exhibited by the Glu-45  Lys mutant. This is supported by several lines of evidence. (a) While the transport of MTX was essentially abolished in CEM-MTX cells (with the Glu-45  Lys mutant RFC), GW70 cells (with the triple RFC mutations) retained 40% of parental MTX transport V_max and over 117% of parental folic acid V_max. (b) While CEM-MTX cells retained the parental transport K_m for MTX, GW70 cells had 3- and 6-fold increases in the transport K_m for MTX and for GW1843, respectively. (c) While CEM-MTX cells had a 9-fold decreased transport K_m for 5-formyltetrahydrofolate, that in GW70 cells was only modestly decreased (3-fold). (d) While CEM-MTX cells had a 31-fold decreased transport K_m for folic acid, that in GW70 cells was decreased by 11-fold. Taken collectively, the RFC from GW70 cells with the triple TMD1-clustered mutations was prominently decreased in its affinity toward the selecting antifolate GW1843, while largely retaining carrier mobility. This is in complete contradistinction to the Glu-45  Lys mutation in CEM-MTX cells which disrupted carrier mobility without altering the transport K_m for MTX. This further enforces the conclusion that TMD1 plays an important role in the formation of the (anti)folate-binding site.

Our results indicate that the mechanism of resistance to GW1843 and other antifolates is based on a combination of two critical components in GW70 cells as follows: (a) markedly increased uptake of folic acid resulting in expansion of the intracellular folate pools, and (b) decreased antifolate (i.e. GW1843) uptake. The marked increases in both folic acid uptake and steady-state levels were due to the following: 1) 11-fold decrease in the transport K_m for folic acid, and 2) retention of parental folic acid transport V_max. These changes led to an expansion of the intracellular folate pool in GW70 cells. Thus, we (38, 56) as well as others (34) have shown that relatively small increases (2-5-fold) in the folate pool size can bring about a high level of resistance to various antifolates. This is due to the occupation by the abundant folates of different folate-dependent enzymes such as DHFR, TS, FPGS, GARTF, and AICARTF thereby competitively ablating antifolate binding to these enzymes (34, 38, 56). Indeed, GW70/LF cells that were grown on medium containing 2 nM folic acid had (a) a 9-fold reduction in the folate pool size relative to parental CEM cells (Table VI), (b) lost their antifolate resistance phenotype, and thereby (c) became markedly hypersensitive to several antifolates including MTA and TMQ (Table II). In contrast, GW70/LF cells grown on medium containing 2.3 µM folic acid had a 4.3-fold expanded folate pool relative to parental cells and consequently displayed up to 1,000-fold antifolate resistance (e.g. DDATHF, Table II). This resistance was particularly high to antifolates that absolutely depend on polyglutamylation for their cytotoxicity including DDATHF, MTA, and ZD1694 (8, 38, 56). These results demonstrate the dominant role cellular folate pool size plays in conferring antifolate resistance. Another contributing factor to GW1843 resistance was the decreased GW1843 uptake which was due to the following: 1) a 6-fold decrease in the transport affinity for GW1843, and 2) a 2.5-fold decrease in the antifolate (measured as MTX uptake) transport V_max. These resulted in a 12-fold reduction in the steady-state transport levels of MTX. In conclusion, the combination of prominently decreased steady-state transport levels of the antifolate along with an expanded folate pool underlies the high level of antifolate resistance displayed by GW70 cells.

GW1843 has a unique structure when compared with various antifolates and reduced folates that use RFC as their main uptake route (Fig. 1). GW1843 has a bulky methyl-oxobenzoyl side ring attached to the quinazoline structure. Despite this bulky group, GW1843 is still the best transport substrate for the human RFC (8, 38). Thus, parental CEM cells display a ~10-fold lower transport K_m for GW1843 than for MTX (8, 38). A close examination of the GW1843 structure reveals the following. 1) The methyl-oxobenzoyl ring and the quinazoline to which it is attached are one continuous planar structure. 2) This methyl-oxobenzoyl ring and the quinazoline to which it is attached are a hydrophobic structure, unlike the hydrophilic pteridine present in MTX and natural folates (Fig. 1). 3) The indolinyl spacer that connects the quinazoline and the glutamate side chain is a more rigid structure than the more flexible spacer joining the pteridine with the p-aminobenzoic acid in MTX and folates (Fig. 1). Based on the high affinity of RFC for GW1843, it is possible that the binding of GW1843 to the RFC may be enhanced through stabilizing hydrophobic interactions via the planar methyl-oxobenzo-quinazoline structure. In contrast, the charged diaminopteridine structure of MTX is devoid of this potentially stabilizing hydrophobic interaction. This notion is strongly supported by previous antifolate structure-function studies of Westerhof et al. (8) that showed that substitution, in MTX or aminopterin, of the charged 2-amino group by a hydrophobic methyl group, significantly decreased the transport K_i of these antifolates. These results suggest that the planar and hydrophobic methyl-oxobenzo-quinazoline structure of GW1843 may indeed interact via hydrophobic interactions with hydrophobic amino acids from TMD1 and presumably other TMDs thereby increasing the affinity of RFC for this compound.

	ACKNOWLEDGEMENTS

We are indebted to Drs. David G. Priest and Marlene A. Bunni for the folate pool analysis. We thank Dr. Larry H. Matherly for providing the recombinant human RFC antiserum and Dr. Garry K. Smith for the generous gift of GW1843U89. We extend our gratitude to Yarim M. Assaraf for his help with computer analysis.

	FOOTNOTES

^* This study was supported by research grants from the Zimmer Cancer Research Fund (to Y. G. A.), the Hedson Fund for Medical Research (to Y. G. A.), and by the Dutch Cancer Society Grant NKB-VU-96-1260 (to G. J).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Recipient of a Young Investigator Exchange Fellowship of the EORTC-Pharmacology and Molecular Mechanisms Group.

To whom correspondence should be addressed: Dept. of Biology, the Technion-Israel Institute of Technology, Haifa 32 000, Israel. Tel.: 972-4-8293744; Fax: 972-4-8225153; E-mail: assaraf@tx.technion.ac.il.

Published, JBC Papers in Press, July 17, 2000, DOI 10.1074/jbc.M003988200

² S. Drori, Y. Kaufman, G. Jansen, M. G. Rots, L. H. Matherly, R. G. Gorlick, J. R. Bertino, and Y. G. Assaraf, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: MTX, methotrexate; RFC, reduced folate carrier; TMD, transmembrane domain; DHFR, dihydrofolate reductase; TS, thymidylate synthase; FPGS, folylpoly--glutamate synthase; GARTF, glycinamide ribonucleotide transformylase; GW1843U89, (S)-2-[5-(((1,2-dihydro-3-methyl-1-oxobenzo(f)quinazoline-9-yl)methyl)amino)-1-oxo-2-isoindolinyl]-glutaric acid; MTA (LY231514), N-(4-(2-(-amino-4,7-dihydro-4-oxo-pyrrolo[2,3-D]pyrimidin-5-yl)ethyl)benzoyl]- L-glutamic acid; ZD1694, N-[5(N-(3,4-dihydro-2-methyl-4-oxoquinazolin-6-yl-methyl)amino)2-thenyl)]-L-glutamic acid; ZD9331, (2S)-2-{O-fluoro-p-[N-(2,7-dimethyl-4-oxo-3,4-dihydroquinazolin-6-ylmethyl)-N-(prop-2-ynyl)amino]benzamido}-4-(tetrazol-5-yl)butyric acid; DDATHF, 5,10-dideaza-5,6,7,8-tetrahydrofolic acid; PT523, N-(4-amino-4-deoxypteroyl)-N-(hemiphthaloyl)-L-ornithine; AG337, 3,4-dihydro-2-amino-6-methyl-4-oxo-5-(4-pyridylthiol)quinazoline; PCR, polymerase chain reaction; SSCP, single strand conformational polymorphism; LF, low folate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Deng, Z. Hou, L. Wang, C. Cherian, J. Wu, A. Gangjee, and L. H. Matherly Role of Lysine 411 in Substrate Carboxyl Group Binding to the Human Reduced Folate Carrier, as Determined by Site-Directed Mutagenesis and Affinity Inhibition Mol. Pharmacol., April 1, 2008; 73(4): 1274 - 1281. [Abstract] [Full Text] [PDF]

				Y. Kaufman, I. Ifergan, L. Rothem, G. Jansen, and Y. G. Assaraf Coexistence of multiple mechanisms of PT523 resistance in human leukemia cells harboring 3 reduced folate carrier alleles: transcriptional silencing, inactivating mutations, and allele loss Blood, April 15, 2006; 107(8): 3288 - 3294. [Abstract] [Full Text] [PDF]

				L. Rothem, B. Berman, M. Stark, G. Jansen, and Y. G. Assaraf The Reduced Folate Carrier Gene Is a Novel Selectable Marker for Recombinant Protein Overexpression Mol. Pharmacol., September 1, 2005; 68(3): 616 - 624. [Abstract] [Full Text] [PDF]

				K. Robien Folate During Antifolate Chemotherapy: What We Know... and Do Not Know Nutr Clin Pract, August 1, 2005; 20(4): 411 - 422. [Abstract] [Full Text] [PDF]

				L. Rothem, M. Stark, and Y. G. Assaraf Impaired CREB-1 Phosphorylation in Antifolate-Resistant Cell Lines with Down-Regulation of the Reduced Folate Carrier Gene Mol. Pharmacol., December 1, 2004; 66(6): 1536 - 1543. [Abstract] [Full Text] [PDF]

				C. Kneuer, K. U. Honscha, and W. Honscha Sodium-dependent methotrexate carrier-1 is expressed in rat kidney: cloning and functional characterization Am J Physiol Renal Physiol, March 1, 2004; 286(3): F564 - F571. [Abstract] [Full Text]

				L. Rothem, M. Stark, Y. Kaufman, L. Mayo, and Y. G. Assaraf Reduced Folate Carrier Gene Silencing in Multiple Antifolate-resistant Tumor Cell Lines Is Due to a Simultaneous Loss of Function of Multiple Transcription Factors but Not Promoter Methylation J. Biol. Chem., January 2, 2004; 279(1): 374 - 384. [Abstract] [Full Text] [PDF]

				L. Rothem, A. Aronheim, and Y. G. Assaraf Alterations in the Expression of Transcription Factors and the Reduced Folate Carrier as a Novel Mechanism of Antifolate Resistance in Human Leukemia Cells J. Biol. Chem., March 7, 2003; 278(11): 8935 - 8941. [Abstract] [Full Text] [PDF]

				Y. G. Assaraf, L. Rothem, J. H. Hooijberg, M. Stark, I. Ifergan, I. Kathmann, B. A. C. Dijkmans, G. J. Peters, and G. Jansen Loss of Multidrug Resistance Protein 1 Expression and Folate Efflux Activity Results in a Highly Concentrative Folate Transport in Human Leukemia Cells J. Biol. Chem., February 21, 2003; 278(9): 6680 - 6686. [Abstract] [Full Text] [PDF]

				R. Yang, R. Sowers, B. Mazza, J. H. Healey, A. Huvos, H. Grier, M. Bernstein, G. P. Beardsley, M. D. Krailo, M. Devidas, et al. Sequence Alterations in the Reduced Folate Carrier Are Observed in Osteosarcoma Tumor Samples Clin. Cancer Res., February 1, 2003; 9(2): 837 - 844. [Abstract] [Full Text] [PDF]

				C. Laverdiere, S. Chiasson, I. Costea, A. Moghrabi, and M. Krajinovic Polymorphism G80A in the reduced folate carrier gene and its relationship to methotrexate plasma levels and outcome of childhood acute lymphoblastic leukemia Blood, November 15, 2002; 100(10): 3832 - 3834. [Abstract] [Full Text] [PDF]

				J. R. Whetstine, A. J. Gifford, T. Witt, X. Y. Liu, R. M. Flatley, M. Norris, M. Haber, J. W. Taub, Y. Ravindranath, and L. H. Matherly Single Nucleotide Polymorphisms in the Human Reduced Folate Carrier: Characterization of a High-Frequency G/A Variant at Position 80 and Transport Properties of the His27 and Arg27 Carriers Clin. Cancer Res., November 1, 2001; 7(11): 3416 - 3422. [Abstract] [Full Text] [PDF]

				J. Worm, A. F. Kirkin, K. N. Dzhandzhugazyan, and P. Guldberg Methylation-dependent Silencing of the Reduced Folate Carrier Gene in Inherently Methotrexate-resistant Human Breast Cancer Cells J. Biol. Chem., October 19, 2001; 276(43): 39990 - 40000. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/40/30855    most recent M003988200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Drori, S. Articles by Assaraf, Y. G. Search for Related Content PubMed PubMed Citation Articles by Drori, S. Articles by Assaraf, Y. G. Social Bookmarking                      What's this?