Different antifolate-resistant L1210 cell variants with either increased or decreased folylpolyglutamate synthetase gene expression at the level of mRNA transcription.

L1210 cell variants selected in the presence of the lipophilic dihydrofolate reductase inhibitor, metoprine, expressed increased levels of one-carbon, reduced folate transport inward (Sirotnak, F. M., Moccio, D. M., and Yang, C.-H.(1984) J. Biol. Chem. 259, 13139-13144). Growth of one of these variants (L1210/R69), with metoprine in the presence of decreasing concentrations of l,L5-CHO-folateH (natural diastereoisomer of 5-formyltetrahydrofolate), resulted in the selection of other variants (L1210/R82, R83, and R84) with further reduction in one-carbon, reduced folate transport and in two cases (L1210/R83 and R84) with 3-8-fold increased folylpolyglutamate synthetase (FPGS) activity and folate compound polyglutamate formation in situ. Metoprine resistance was further increased, and the requirement for exogenous folate during growth was decreased as well in these variants. The increase in FPGS activity observed in L1210/R83 and R84 was characterized by 3- and 8-fold increases in value for Vmax with no change in K and the same increase in a 60-61-kDa protein as shown by immunoblotting. Northern blotting revealed the same increases in these two variants in the level of a 2.3-kilobase FPGS mRNA when compared with control, while Southern blotting of genomic DNA did not reveal any increase in FPGS gene-copy number or restriction polymorphisms. Also, no difference in stability of FPGS mRNA was found between parental and variant cells. In contrast, nuclear run-on assays revealed differences among these cell types in the rate of FPGS mRNA transcription that correlated with increased FPGS activity, protein, and mRNA level in the variants. Similar studies with a transport-defective, methotrexate-resistant L1210 cell variant (L1210/R25) documented a 2-3-fold decrease in FPGS activity, protein, and mRNA levels that was accounted for by a decrease in FPGS mRNA transcription. These results provide the first examples of constituitively altered transcriptional regulation of FPGS activity associated with acquired resistance to antifolates.

(1-5) of varying chain length. Their metabolism to polyglutamates and that of folate analogues are mediated (1)(2)(3)(4)(5) by the enzyme, folylpolyglutamate synthetase (FPGS), 1 and metabolic turnover of these anabolites appears to be modulated by folylpolyglutamate hydrolase after their mediated entry into lysosomes (for review, see Ref. 6). In tumors and normal proliferative tissues of animals and man, the process of polyglutamylation has pharmacologic relevance with respect to the cytotoxicity (8,9) and therapeutic utility (7)(8)(9)(10)(11)(12)(13) of classical folate analogues. Also, both decreased levels of FPGS activity (14,15) and increased levels of folylpolyglutamate hydrolase activity (16) have been associated with acquired resistance to these analogues. The mechanistic basis for these alterations remain to be elucidated.
The process of folylpolyglutamylation in normal proliferative and neoplastic mammalian tissues is important (1)(2)(3)(4)(5) to the conservation and efficient utility of folate coenzymes that are required for one-carbon transfer reactions during macromolecular biosynthesis. Consequently, levels of FPGS activity appear to be highest in the proliferative fraction of normal differentiating tissues (7,(17)(18)(19). It has been suggested in the context of earlier reports (for review, see Refs. 20 -23) that normal proliferative and tumor cells might control their macromolecular synthesis through regulation of intracellular folate homeostasis. In addition to the biosynthesis and metabolic interconversion of these compounds (20,21), folate homeostasis could also be regulated at the level of mediated entry of exogenous folate (23) and/or through biosynthesis of folylpolyglutamates (1)(2)(3)(4)(5). With the recent derivation (24) of the cDNA for human FPGS, studies addressing this issue at the level of FPGS gene expression and its regulation are now possible. Toward this objective, we now describe studies with a novel group of metoprineresistant variants of the L1210 cell that were further characterized and found to constitutively overproduce FPGS to varying extent. These variant cell lines, which also overproduce the reduced folate transporter (25,26), were selected during growth in the presence of this lipophilic antifolate and decreasing amounts of l,L5-CHO-folateH 4 that was increasingly growth limiting. For comparison, we also describe a methotrexate-resistant L1210 cell variant, which in addition to markedly reduced transport inward (26,27) of folate compounds exhibits lower levels of FPGS activity compared with parental cells. Our results document, as the molecular basis for the altered level of FPGS activity in all of these variants, a constitutive increase or decrease in the rate of FPGS mRNA transcription depending * This work was supported in part by Center Core Grant CA08748 and Grant CA56517 from the National Cancer Institute and the Elsa U. Pardee Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EMBL Data Bank with accession number(s) U33557.

EXPERIMENTAL PROCEDURES
Cells and Culture Conditions-Methods for the isolation of variant L1210 cells with elevated one-carbon reduced folate transport and FPGS activity were similar to those reported earlier (25)(26)(27)(28)(29) from this laboratory. L1210 cell variants were first selected during growth in the presence of increasing amounts of metoprine (IC 50 ϭ 77 nM) in folatefree RPMI medium supplemented with dialyzed fetal calf serum and 20 nM l,L5-CHO-folateH 4 , which allowed maximum growth in drug-free medium. Later selection steps utilized growth in 600 nM metoprine and decreasing amounts of l,LCHO-folateH 4 as the sole folate source in order to avoid the selection of variants with elevated levels of dihydrofolate reductase. At each step in the selection, the variant was cloned by limiting dilution and was examined for [ 3 H]MTX transport inward (25)(26)(27)(28)(29) and FPGS activity. Subsequent selection steps were carried out with the cloned subline. Requirements for l,L5CHO-folateH 4 (EC 50 ) and inhibition by folate antagonists (IC 50 ) were determined by methods described previously (8,27).
Assay for FPGS activity-After processing of cells (2,14) grown in drug-free medium, the protein concentration of the resulting cell-free extract was determined (30,31), and the assay for FPGS activity was carried out as described previously (2,13,14) using aminopterin as substrate. Product formation (aminopterin ϩ GI) under these conditions was linear for at least 2 h at 37°C, and these cell-free preparations under the assay conditions exhibited (12) no detectable folylpolyglutamate hydrolase activity.
Derivation of a Murine FPGS cDNA Probe-A murine FPGS cDNA (ZAP-L1210/R83-1) was obtained by hybridization screening (32) of an L1210 cell cDNA library in gt11 (Stratagene, La Jolla, CA) using a human cDNA, pTZ 18U (24), as a probe. This gt11 cDNA construct incorporates a 2.296-kilobase insert ligated at the XhoI polylinker site including 3Ј-and 5Ј-untranslated region sequences of 461 and 74 base pairs, respectively, and an open reading frame of 1761 base pairs, which codes for a putative mitochondrial leader peptide as well as the enzyme protein. The homology exhibited (data not shown) between the murine and human cDNAs was 79% at the level of the nucleotide sequence. Sequencing of the murine cDNA was by the method of Sanger et al. (33).
Northern Blot Analysis of FPGS mRNA-A method (34) utilizing rapid isolation of poly(A) ϩ RNA directly from the cell lysates by means of an oligo(dT) column was used. The integrity of the RNA was assessed by electrophoresis in 1.1% agarose containing 1 M glyoxal and staining with ethidium bromide. An aliquot (2-10 g) of the same RNA was analyzed (35,36) by Northern blotting and radioautography using a murine FPGS cDNA, ZAP L1210/R83-1 (see above), as a probe and normalized to mRNA content with a human ␥-actin probe, PCD-␥-actin (37). Labeling of each probe was by random priming (Random Primers DNA labeling kit, Boehringer Mannheim) using [␣-32 P]dCTP (3000 Ci/mmol and 10 ng of insert). Direct measurement of total radioactivity in each blot minus background was obtained with a Betagen 603 blot analyzer.
Southern Blot Analysis-Genomic DNA from parental and resistant cells as a frozen pellet was prepared (38) and digested with EcoRI and HindIII. The DNA was separated by electrophoresis through a 0.82% agarose gel and transferred (38) to Nytran ϩ (Schleicher and Schuell). DNA probes, hybridization and labeling procedures were the same as that used above. Immunoblotting Procedure-Samples of a 0 -30% ammonium sulfate fraction (13) of cytosol were solubilized in SDS sample buffer and electrophoresed (39) on a 7.5% polyacrylamide slab gel and transferred to nitrocellulose using a Bio-Rad Trans-Blot cell (40). Western blotting was performed (40,41) using anti-human FPGS peptide polyclonal antibody (0.5 g/ml) prepared in rabbits and purified on a peptide-Sepharose column (41,42). The human FPGS peptide utilized was deduced from the cDNA nucleotide sequence (24) beginning at Val 175 and ending at Trp 199 in the order Val-Val-Cys-Gly-Val-Ser-Leu-Gly-Ile-Asp-His-Thr-Ser-Ser-Leu-Leu-Gly-Asp-Thr-Val-Glu-Lys-Ile-Ala-Trp. The peptide was linked by EDC-mediated coupling (43) to keyhole lympet hemacyanin for monthly injections into a rabbit. During blotting, the anti-rabbit horseradish peroxidase-IgG conjugate (Sigma) at a 1:3000 dilution was used as the secondary antibody. The blots were used to expose hyperfilm after incubation with ECL reagents (Amersham Corp.) and quantitated by scanning densitometry (Stratagene).
Nuclear Run-on Assay-These assays were performed with nuclear extracts essentially as described in earlier reports (44,45). The cDNA probes used in the blot for relative transcript level were murine FPGS (see above) and murine dihydrofolate reductase (46). cDNA probes for ␤-(47) and ␥-actin (37) were used as a positive control. A cDNA probe for plasmid-derived neomycin resistance (48) (8). Folyl-polyglutamate hydrolase activity was measured as described previously (12).

Isolation and Biochemical Characterization of the Variant
Cell Lines-All of the variant L1210 cell lines used in these studies were derived from a metoprine-resistant variant (L1210/R69), which was selected (25) for growth in the presence of 600 nM metoprine and 20 nM l,L5CHO-folateH 4 as the sole folate source. These additional variants were selected from L1210/R69 in a stepwise manner following growth in 600 nM metoprine and 4 nM (L1210/R82), then 2 nM (L1210/R83), and finally 1 nM (L1210/R84) l,L5CHO-folateH 4 with cloning of the variant after each selection step. The properties of L1210/R69 and its antecedents were described earlier (25). The requirement for l,L5-CHO-folateH 4 for growth by these newly cloned variants in the presence of 600 nM metoprine was determined  ( Table I) and found to be progressively lower than for parental cells with each additional selection step. Furthermore, resistance to metoprine increased so that the relative requirement for this folate among these variants (Table I) was inversely related to their resistance to metoprine. Two biochemical alterations were documented in these variants, which appeared to allow their growth in the presence of increasing concentrations of metoprine and/or decreasing concentrations of folate when compared with parental L1210 cells. At the initial stages of selection (Ref. 25 and Table I), one of the variants derived exhibited elevated transport inward of folate compounds that was directly related to resistance and inversely related to the folate requirement. At later stages of selection, in addition to a further increase in folate transport inward, two of these variants exhibited (Table I) a 3-and 8-fold increase in FPGS activity that was also directly related to resistance and inversely related to the folate requirement. All three variants were collaterally sensitive to methotrexate (data not shown). In every case (Table I), the folate requirement in the presence of metoprine and the relative resistance to this lipophilic antifolate of the variants when compared with parental cells could be accounted for by the alterations both at the level of folate compound transport inward and of FPGS activity that were observed when smaller changes (Table I) in the relative dihydrofolate reductase activity were also taken into account. We also provide data in this table (Table I) on a methotrexate resistant variant (L1210/R25) for comparison. This cell line exhibited markedly lower folate compound influx and almost 3-fold lower FPGS activity. Moreover, it will not grow in l,L5-CHO-folateH 4 as the sole folate source and is highly collaterally sensitive to metoprine. The 3-and 8-fold increase in FPGS activity observed in the metoprine-resistant variants when compared with the parental cells was accounted for by ( Fig. 1) a commensurate increase in values for V max for variant FPGS activity with no change in value for apparent K m . Western blotting was carried out after SDS-polyacrylamide gel electrophoresis of partially purified FPGS from cell-free extract from variant and parental L1210 cells ( Fig. 2A) with anti-FPGS peptide antibody. Densitometry (data not shown) revealed the same relative increases (3-8-fold) among these variants in the amount of a 60 -61-kDa protein. Both results taken together suggested that the increase in FPGS activity observed in these variants resulted from an elevation in level of FPGS enzyme protein. In contrast, the V max for FPGS activity in L1210/R25 cells was reduced almost 3-fold (Fig. 1), and SDS-polyacrylamide gel electrophoresis and Western blotting with densitometry of partially purified cell-free extract detected (Fig. 2B)  Variant and Parental L1210 Cells-Northern blotting of FPGS mRNA from variant and parental cells revealed (Fig. 3) substantial differences among these cell types. FPGS mRNA levels, shown in the figure, when compared with parental cells, were increased (Fig. 3A) 3-fold in L1210/R83 cells and 7-8-fold in L1210/R84 cells when blots were normalized during Betagen blot analysis (data not shown) with respect to ␥-actin mRNA level used as a control. In contrast, the level of FPGS mRNA in L1210/R25 cells was decreased 2-3-fold (Fig. 3B) when related in the same way to the same ␥-actin mRNA control.
Since differences in stability of FPGS mRNA may be the explanation for the differences in its level among these variants, we employed the same methodology to determine the stability of FPGS mRNA in these variants compared with parental L1210 cells. In the experiment shown (Fig. 4), L1210/ R84 and parental cells were exposed to actinomycin D during growth in culture, and aliquots of cells were removed after varying periods of time for Northern blotting with FPGS and ␥-actin cDNA. The results of a typical experiment (Fig. 4A) show that the rate of decay of FPGS mRNA with time was essentially the same for each cell line. The radioactivity in each blot was also determined for replicate experiments by a Betagen blot analyzer, and the average results for FPGS mRNA normalized against ␥-actin mRNA are given in Fig. 4B. These data allowed the quantitation of the half-time for FPGS mRNA decay from each cell type, which was found to be the same (half-time ϭ 5.8 Ϯ 0.8 h). Stability of L1210/R25 FPGS mRNA was also determined (data not shown) in the same way and found to be the same as parental cell and L1210/R84 cell FPGS mRNA.
FPGS Gene Copy Number in Parental and Metoprine-resistant L1210 Cells-Increased levels of FPGS mRNA in the metoprine-resistant variants (L1210/R83 and L1210/R84 cells) compared with parental L1210 cells might also be accounted for by an increase in gene copy number (45). This possibility was evaluated by quantitative dot-blotting and Southern blotting (38) of restriction enzyme digested genomic DNA from these various cell lines using the murine FPGS cDNA probe. The results (data not shown) revealed no unique restriction polymorphisms in the DNA and no evidence for increased FPGS gene copy number in these variants compared with parental cells.
FPGS mRNA Transcription in Variant and Parental L1210 cells-In light of the results above showing no difference among these variants in either FPGS mRNA stability or FPGS gene copy number, we examined the rate of FPGS mRNA transcription in parental L1210 cells and L1210/R84 and L1210/R25 cells by means of a nuclear run-on assay (44,45). Since dihydrofolate reductase activity was also increased (26) in L1210/ R84 cells, we also probed for increased transcription of its mRNA in L1210/R84 cells as a positive control. Dihydrofolate reductase activity was not increased in L1210/R25 cells. The results presented in Fig. 5 show that the rate of FPGS transcript formation when related to controls with L1210/R84 derived nuclear material (Fig. 5A) was markedly increased compared to wild-type, while the relative rate of transcript formation with L1210/R25 derived nuclear material (Fig. 5B) was decreased. The extent of these differences was quantitated with the Betagen Blot analyzer and found (data not shown) to be in general agreement with the differences shown for FPGS activity and mRNA level. That is, transcription of FPGS mRNA was increased 7-8-fold in L1210/R84 cells and decreased almost 3-fold in L1210/R25 cells compared with parental L1210 cells.
Similar to our studies of one-carbon reduced folate transport (25,26), the data presented here appear to document a contrasting role for FPGS as a determinant of resistance to these two categories of folate antagonists. Elevated levels of FPGS activity are exhibited by metoprine-resistant L1210 cells, while FPGS levels are reduced in the MTX-resistant variant also examined in these studies. Within each category of resistant cells, the basis for these alterations appears to reflect different  32 P-Labeled mRNA transcripts obtained with nuclear extracts of each cell type were used in a RNA/DNA blot with murine FPGS cDNA. cDNAs for murine dihydrofolate reductase, ␤and ␥-actin, and plasmid-related neomycin resistance were used as controls. A, blot obtained with mRNA transcripts from L1210 and L1210/R84 cell nuclear extracts. B, blots obtained with mRNA transcripts from L1210 and L1210/R25 cell nuclear extracts done under different conditions. Additional details are provided in the text. The blot shown is from a typical experiment replicated twice.

FIG. 4. Northern blot analysis of the decay of FPGS poly(A) ؉ mRNA from actinomycin D-treated parental and variant L1210 cells with increased (L1210/R84) FPGS activity.
Cells were grown in the presence of 5 g/ml actinomycin D, and aliquots of the cell suspension were removed at various time intervals for mRNA extraction. Additional experimental details are provided in the text and in the legend of Fig. 3. The data in A represent one of a typical series of blots of FPGS mRNA controlled for ␥-actin mRNA. The data in B are from an analysis of radioactivity of replicate blots carried out with a Betagen blot analyzer. levels of FPGS gene expression, specifically, the rate of FPGS mRNA transcription. We have previously documented (49) lower levels of FPGS activity and of folate analogue polyglutamate formation in another group of L1210 cell variants resistant to a new classical folate analogue, edatrexate. The molecular basis for these alterations has yet to be elucidated. However, in contrast to the current results, no alteration was found (50) in preliminary studies on the level of FPGS mRNA in any of these variants. Downward alterations of FPGS activity in variants (Refs. 14 and 15, and this study) resistant to classical antifolates further substantiate the importance of polyglutamylation to their mechanism of action and the role of FPGS in addition to one-carbon reduced folate transport as determinants of cytotoxicity to these agents.
In the case of the lipophilic antifolate, metoprine, resistance appears to be engendered by an increase in both folate transport inward and FPGS activity as it pertains to natural folate compounds. In contrast to classical antifolates, neither property is involved in the internalization or metabolic disposition of this agent in tumor cells. However, because of their role (1)(2)(3)(4)(5) in maintaining intracellular levels of folate coenzyme polyglutamates, increased expression of these properties apparently renders the cell less sensitive to the folate antagonism mediated by this lipophilic antifolate. In support of this notion, other data also show ( Table I) that altered levels of expression of FPGS in addition to folate transport inward profoundly modulate the requirement for exogenous folate during growth of these cells in culture. Such effects indicate a role for both properties in maintaining folate homeostasis in these tumor cells. Although we sought to select for variants with increased FPGS activity by reducing the folate content of the medium in the presence of a fixed level of metoprine, these variants do exhibit increased resistance to metoprine. Therefore, it would be expected in view of the above considerations that derivation of similar variants with elevated levels of FPGS would also occur by selection in increasing levels of metoprine alone.