Accumulation of 5-methyltetrahydrofolic acid and folylpolyglutamate synthetase expression by mitogen stimulated human lymphocytes

British Joirrnal of Haematology, 1993. 84, 595-601

Accumulation of 5-methyltetrahydrofolic acid and folylpolyglutamate synthetase expression by mitogen stimulated human lymphocytes

DANIEL W. FORT,' RICHARD H. LARK,^ ANGELA K. SMITH,' MARGARET MARLING-CASON,' STEVEN D. WE IT MAN,^ BARRY S H A N E ~ A N D BARTON A. K A M E N ' . ~ Department of 'Pediatrics and 2Pharrnacology, Universitg of Texas Southwestern Medical Center, Dallas, Texas, and 'Department of Nutritional Sciences, University of California, Berkeley, California

Received 22 December 1992; accepted for publication 22 February 1993

Summary. The accumulation of 5-methyl[3H]tetrahydrofolic acid ( 5CH3[jH]FH4) by phytohaemagglutinin stimulated lymphocytes (PHA-L) cultured in folate free media was investi- gated to determine the mechanism of uptake of 5CH3FH4 and the requirement of the cells for this vitamin as assessed by monitoring dv novo thymidine synthesis. When grown in 20 n u 5CH%['H]FH4 PHA-L accumulate radiolabel at a rate of 0.04 pmol/h/lOh cells. This doubles the endogenous folate pool of unstimulated cells (0 .6 f0 .16 pmol/lOb cells) in about 15 h. Uptake proceeded via a saturable process, independent of a high affinity folate receptor as assessed by ligand binding and by Northern and Western blot analysis. However, transport was blocked by probenecid. which is consistent with an anion carrier mechanism. Unstimulated

cells lacked folylpolyglutarnate synthetase (FPGS) activity and did not express significant amounts of FPGS mRNA. After 48 h ofmitogen stimulation there was a 4-10-fold increase in FPGS mRNA and folylpolyglutamate formation (Glu, 5) was essentially simultaneous with ~JCH~[~H]FH~ transport. Increasing extracelldar folate to 2pM only increased intracellular folate %fold. but the length of the folylpolyglutamates decreased. The increased folate did not increase de novo thymidine synthesis compared to cells grown in physiological folate. We conclude that mitogen stimulation activates the process(es) for folate accumulation, especially FPGS, and that physiological uptake (0.04 pmol/h/106 cells) is adequate for meeting the cells' need for the vitamin.

Folates are essential cofactors in the biosynthesis of purines and thymidine. Antifolates such as methotrexate (MTX) have been used as a mainstay of cancer chemotherapy for over 40 years and new folate analogues are still being developed as both antimicrobial and cytotoxic agents and for the treat- ment of cancer and autoimmune disorders (Farber et al, 1948; Weinstein e t a / , 1990; Giannini e t a / , 1992; reviewed in Kamen, 1987). Our laboratory has been concerned with folate homeostasis in malignant and nonmalignant cell lines as well as in fresh tissue from surgical and autopsy specimens (Weitman et (11. 1992a). Recently, we defined a receptor coupled transmembrane transport (RCT) mechanism termed potocytosis (Anderson et al, 1992). This process enables cells to rapidly accumulate folate when cultured in low physiolo- gic concentrations ( < 2 nM) of the vitamin. Cells with

Correspondence: Dr Daniel Fort, UT Southwestern Medical Center. Department of Pediatrics. 5 323 Harry Hines Blvd, Dallas, TX 752 3 5. U.S.A.

functioning receptor accumulate folate 30-50 times faster than cells in which the receptor has been blocked or is absent (Kamen et a/ . 1991).

Ligand binding, Northern blotting, immunoblotting, and immunocytochemical techniques have shown that the receptor is not present in all tissues. In fact, distribution of the receptor is limited primarily to epithelial cells (Weitman et aI, 1992b) and is overexpressed in a number of carcinoma cell lines in vitro (Weitman et al, 1992a).

In preliminary studies we did not detect a high afEnity folate receptor in lymphocytes using [3H]folic acid. Mitogen stimulated human lymphocytes were originally studied as a model for megaloblastosis (Das & Hoffbrand, 1970; Hayman & van der Weyden, 1980; van der Weyden et al. 1991a. b). We used this system to study accumulation of high specific activity (6s) SCH3r3H]FH4, the natural serum folate. with specific regard to the folate receptor and folylpolyglutarnate synthetase (FPGS).

Using radiolabelled 5CHJ3H]FH4, the predominate form of

59s

596 Daniel W. Fort et a1

folate found in serum, with high specific activity, our studies expand and confirm the pioneering work from several laboratories over a decade ago. Das & Hoffbrand (1970) showed that phytohaemagglutinin stimulated lymphocytes (PHA-L) accumulated more [IHIfolate than did nonstimu- lated lymphocytes. Matthews & Wickramsinghe (1 986) showed that PHA-I, exhibit increased [ 'Hlthymidine incorporation, raised deoxyuridine suppression values, had an increased cell volume and exhibited abnormal DNA synthesis when the intracellular folate content fell below 1 prnol/lOh cells. They concluded that there is a critical concentration of folate required for cell division. van der Weyden et a1 (1 991b) confirmed this observation by demonstrating reduced de novo purine and pyrimidine synthesis in PHA-L grown in folate deplete media. These studies were done using predominantly j3H]folic acid at nonphysiological concentrations.

Our results show that when PHA-L are labelled with 20 nM 5CH3[ 'H]FH4 accumulation is limited by folate transport not FPGS activity. Transport is independent of the high affinity folate receptor as judged by negative Northern and Western analysis as well as functional ligand binding analysis using

Metabolism of folylmonoglutamate to higher polyglutamate forms results in trapping offolate. We found a 4-10-fold increase in FPGS mRNA in cells cultured 48 h with mitogen compared to unstimulated cells. When PHA-L were grown with physiological 5CHJ3H]FH4 intracellular [3H]folate was predominantly Glu,,. Although at higher concentrations of extracellular folate (e.g. 2 pM) synthesis of polyghtamates increased, a significant amount of Glul was detected and the predominant folylpolyglutamate derivatives were G I U ~ . ~ and little Glu,, detected.

5CH3[jH]FH4.

METHODS

Chemicals and radiochemicals. RPMI 1640 without folic acid was made from RPMI salt solution (R7528) and RPMI amino acids (R7131) from Sigma Chemical (St Louis, Mo.) with a vitamin mixture (lacking folic acid) prepared in our laboratory. Probenecid (P-8761) and phytohaemagglutinin (PHA) (L4138) were from Sigma. Enzymatically prepared (6Sj-5- methyltetrahydro-[3', 5', 7,9-3H]folate (0.74-1.48 TBq/mM) (MT816) was from Moravek (Brea, Calif.). Thus, only the bioactive isomer was used. Monensin was from Calbiochem (San Diego. Calif.). Ficoll was from Pharmacia (Piscataway. N.J.). Hanks Balanced Salt Solution (HBSS) (10-527B) with calcium and magnesium and without phenol red was from Bioproducts Inc. (Walkersville, Md.). Scintillation fluid (Ultima Gold [6013324]) was from Packard (Meriden, Conn.). Folic acid polyglutamate standards were synthesized by Dr G. Nair, University of South Alabama. Proteinase K was from Boehringer Mannheim (Indianapolis, Ind.). Spectral grade formamide and phenol were from International Bio- technologies, Inc. (New Haven, Conn.). The nitrocellulose and Nytran blotting material were from Schleicher and Schuell (Keene. N.H.). MOV-I 9, a monoclonal antibody directed against the folate receptor, was from Centocor (Malvern, Pa.). All other reagents, buffers, and materials were from Sigma except where otherwise specified.

Preparation of lymphocytesfrom normal volunteers. 300 ml of blood were collected in heparinized syringes from each healthy adult volunteer. The blood was mixed 1: l with sterile HBSS and 40 ml aliquots were layered over 10 ml of Ficoll at room temperature in 50 ml tubes and spun for 30 min at 1500 rpm (500 g) in a table-top centrifuge (Beckman model TJ-6). The middle layer of peripheral blood mononuclear cells (PBMCs) was harvested with a pasteur pipette. The PBMCs were washed three times in HBSS. After the final wash the cells were suspended in 10 ml of HBSS and counted using a Coulter counter (Coulter Electronics, Inc., Hialeah, Fla.). All incubations were carried out at 3 7°C.

High pressure liquid chromatography of folutes. Folylpolyglu- tamates were separated on a reverse phase CIS column as previously described (Kamen et a!, 1988 j. As noted, methyl or formyl derivatives do not significantly change the elution time of the polyglutamates. Radioactivity was quantitated by a Flo-One/beta, on-line liquid scintillation counter (Radio- matic Instruments). Authentic folylpolyglutamates were used as internal standards.

Determination of iritracellulur folate content. Intracellular folate content was measured by a competitive radioligand binding assay using sequential analysis and extraction techniques described previously (Kamen et al, 1988).

Measurement of 5CH3L3HJFH4 arctimulation. Lymphocytes were suspended in folic acid free RPMI media containing 5% autologous serum or serum from one of the volunteers (blood type Ot) to a final concentration of 3-5 x 10, cells/2 ml/ dish. The folate content of the complete media without radiolabelled folate was < I n M . Nonspecific uptake was determincd by adding unlabelled 5CH3FH4 In 100-fold excess just prior to the addition of the labelled folate. After incubation with 5CHJ3H]FH4 as indicated for each experiment, the cells were harvested by gentle aspiration. washed twice in 10 volumes of ice-cold HBSS, collected by centrifugation, suspended in 5 ml of BBSS and counted using the Coulter counter. After a final spin the cells were suspended in 1 ml of ice-cold freeze-thaw buffer (10 mM Tris, pH 8, with 0.02 pg/ml aprotinin and leupeptin) containing 1% 8- mercaptoethanol. Half the sample was stored at - 8 0 T until analysis of folate polyglutamates was done. The radioactivity of the remainder was determined using a Packard (1900 CA- Tri Carb) liquid scintillation analyser. Counting efficiency was -55%.

Membrane prtparation for Western blot anulgsis. Cells obtained and processed as above were harvested at four time points (0 ,24 ,48 and 72 h) during incubation with PHA. The cells were suspended in 1.5 ml of freeze-thaw buffer and the suspension was frozen at -80°C for at least 20 rnin. The membrane fraction was separated from the cytoplasmic portion by centrifugation at 100 000 g for 20 min at 4°C. The cell membrane fraction was dissolved in 500 p1 of solubiliza- tion buffer (50 mM Tris. pH 7.4. I50 mM NaCI. 25 mi% 8- octylglucoside, 5 mM EDTA. 0.02% sodium azide), rocked for 3 h at 4% and spun at 100000 g for 20 min to remove insoluble debris. 20 pg of protein were subjected to SDS- polyacrylamide gel electrophoresis, transferred to nitrocellulose, and visualized by immunoblot using MOV-19 as the primary antibody and goat anti-mouse IgG-alkaline phos-

Folate Uptake in Human Lymphocytes 597

- d

aJ u CI 0.3- 5 \ L 2 0.2- \ aJ CI m

Y- a 0.1- - 2 a 0.0

phatase conjugate (1 70-6520, Bio-Rad) for antibody detection according to manufacturer's instructions.

Northprri analysis. All solutions used for RNA isolation were made in water treated with diethyl pyrocarbonate (DEPC). The hybridization cDNA probes (for folate receptor, FPGS. or glyceraldehyde phosphate dehydrogenase DNA) were made by using the Multiprime DNA labelling system according to the manufacturer's instructions. The folate receptor rDNA was the EcoR I fragment from the pGEM4Z plasmid (Lacey rt al. 1989). The FPGS cDNA probe was the EcoR I fragment from the pTZ18U-25 plasmid (Garrow et al. 1992). Lympho- cytes were isolated and stored at - 80°C until the RNA was isolated. Total RNA was isolated according to conventional techniques (Sambrook at a[, 1989). Briefly, each pellet was mixed with the following sequence of solutions: 2 ml of 10 mM EDTA (pH 8 ) with 0.5% SDS and 2 ml of 0 . 2 M Na acetate (pH 5.2) with 10 mM EDTA were mixed with 4 nil acidic phenol to remove proteins and DNA: the two phases were separated by centrifugation at 3000 g for 10 min at 4°C. The RNA from the upper (aqueous) layer was precipitated twice at - 20°C after the addition of Tris. NaCl and ethanol. After the final centrifugation the RNA samples were suspended in 20 pl of DEPC-treated water.

For KNA blotting, 20 p g of RNA from each sample were run on a 1 % agarose gel in the presence of formamide. After transferring RNA onto Nytran, the blot was hybridized with the cDNA probe for folate receptor. FPGS. and marker mRNA (glyceraldehyde phosphate dehydrogenase). Each hybridization was performed overnight at 4 2 T in the presence of 50% forniamide. 5 x SSCPE (1 x SSCPE is 150 mM sodium citrate, 10 mM NaH2P04, 1 m M EDTA, pH to 7.4 with NaOH), 5 x Denhardt's reagent, and 2'5% dextran sulphate. The specific activity of the probe during the hybridization was 3.5 x 10" cpni/nil in 1 5 ml. After each hybridization the blot was rinsed in 2 x SSC/O.l% SDS at 55°C and at 65°C and treated with Proteinase K (0.2 mg/ml final concentration) to remove background noise. The probe was stripped from the blot at 65OC with 50% formamide/5 x SSCPE. CaCOL (human colon carcinoma) cells were included with the lymphocyte samples as positive controls. The amount of mRNA expressed was determined by phosphorimaging (PhosphorImager Model 400B. Molecular Dynamics), which quantifies the radioactivity present on the Northern blot after hybridization with a probe.

[ jH]TdR incorporation into DNA. PHA stimulated lymphocytes were incubated in 0 .20 nu . or 2 PM 5CH3FH4 for 8 h at which time 37 kBq [3H]TdR (final concentration 1 p ~ ) was added for 1 h. At the end of incubation the cells were washed and harvested as before and the pellet was suspended in 10% TCA. The pellet and supernatant were separated by centrifugation and the radioactivity determined in each fraction.

Deoxyundine suppression test and thymidine incorporation, TO determine the effect of PHA stimulation and intracellular folate concentration on thymidine incorporation, [3H]thymidine uptake was measured in PHA-L incubated with three different concentrations of 5CH3FH4 (0, 20 nM or 2 ~ 1 ~ ) either during the entire period of stimulation or a short pulse after 48 h of PHA stimulation. The effect of incubation with 10 /(M

deoxyuridine on thymidine incorporation was also meas-

v) 0.41

Time of Incubation with PHA (h) Fig 1 . Effect of PHA on uptake of 5CH{[jH]FH4 by lymphocytes. Lymphocytes were incubated for 0. 12. 24. 48 or 72 h with PHA prior to a I2 h incubation with 2 0 n M 5CH3[3H]FH+ At the end of the incubation cells were harvested. washed and counted. Radioactivity was determined as detailed in Methods. Each time point represents the mean of samples done in duplicate and are typical of results obtained with at least three separate experiments with blood from at least two donors.

ured. The cells were harvested, washed, and the radioactivity counted as noted above.

RESULTS

Accumulation of radiohbelled 5 C H 3 [ 3 H ] F H 4 by lgniphocytes is dependent on rnitogen (PHA) stirnulation Uptake of 5CH3[]H]FH4 (20 nM) during a 12 h pulse by lymphocytes stimulated with PHA for 0, 24, 48 or 72 h is shown in Fig 1. Unstimulated cells did not accumulate folate. The maximal rate of uptake of 5CHJ3H]FH4 occurred after cells were stimulated for 48 h and declined by 72 h.

Eflect of extracellular folate concentration on PHA-L uptake of

Cells stimulated with PHA for 48 h were incubated in 20 nM 5CH3[{H]FH4 for up to 24 h. The rate of folate accumulation was 0.04 pmol/h/106 cells and remained constant over the time period studied (Fig 2A). This doubles the endogenous folate content of unstimulated cells (0.6 f 0.1 6 pmol/lOb cells, N = 10) in 15 h. By contrast, cells incubated in 2 PM 5CH3[ 'H]FH4 rapidly increased their intracellular concentration ofradiolabelled ligand to 12.5 pmol/l O6 cells by 6 h, but no further net accumulation occurred over the remainder of the time period studied (Fig 2B).

Polate accumulation in PHA-L i s not blocked by agents that inhibit receptor coupled uptake and appears to be carrier mediated To more thoroughly characterize folate binding and transport in PHA-L we used agents known to block individual steps in receptor coupled transport (Kamen et al. 1988, 1991). PHA-L were incubated for 2 h in 20 nM folic acid to block membrane binding, 30 min in 25 p M monensin to block release of folate from the receptor, or 30 min in 10 nM probenecid to inhibit anion transport. Probenecid completely blocked accumulation of 5CHJ {HIFH,; however, folic acid and monensin had no effect (Fig 3). These kinetic data are

5 CHjI3H]FH4


Cn

a, 0

\D 0

a, +J (d

0 'c

4 12- 4

- \ 7-

4

B 0

4 2 ' 0 E Q

0 4 8 12 16 20 24 Time (h)

Fig 2. Uptake of 5CH$'H]FH4 by PHA-L at physiological and pharmacological extracellular concentrations. Lymphocytes were stimulated for 48 h with PHA prior to exposure to 5CH3[3H]FH4 at an extracellular concentration of either 20 n M (A) or 2 PM (B) for the time periods indicated. At the time points indicated the cells were harvested, washed and counted. Radioactivity was determined as described in Methods. Each time point represents the mean of duplicate samples and as in Fig 1, a typical result for experiment done at least three t i e s using blood from different donors.

U

Monensin Probenecid Folic Acid Fig 3. Effect of incubation with monensin. folic acid and probenecid on accumulation of 5CH3[3H]FH4 by PHA-L. PHA-L were incubated with monensin (25 pd), probenecid (10 mM), or folic acid (20 nM) as described in Methods, prior to an 8 h incubation with 20 n~ ?JCH~[~H]FH~. Uptake of the ligand was measured after the cells were harvested, washed and counted as described in Methods.

I ", 4 a :: ?I ? -:

Fig 4. Folylpolyglutamate synthesis in PHA-L. Cells were incubated in either physiological (20 nM) or pharmacological (2 PM) concentrations of 5CH3[3H]FH4 for 8 h. At the end of the incubation the cells were harvested, washed and counted as described in Methods. Panel A shows the HPLC pattern for the standards, G1~1.3.4,5,7. Panel B shows the pattern of folylpolyglutamate formation for PHA-L incubated in 20 n M 5CH3[3H]FH4. Panel C shows the pattern for PHA-L incubated in 2 PM 5CH3['H]FH4, and panel D shows the pattern in unstimulated lymphocytes incubated in 2 PM 5CH3[3H]FH.+

suggestive of an anion carrier transport mechanism and not receptor coupled transport. The absence of a receptor coupled system was confirmed more directly by (a) the absence of detectable binding of [3H]folic acid by PHA-L (<0.05 pmol/ lo6 cells), (b) the absence of receptor as assayed by Western blotting, a technique 100-fold more sensitive than ligand binding (Weitman et al, 1992a). using MOV-19, a mono-


cells pulse labelled with 2 p~ 5CH3[jH]FH4 accumulated more total folate polyglutamates (see Fig 2) than the cells grown in physiological folate; however, the median chain length was smaller and significant amounts of GIul were detected (Fig 4C). Of interest, the rate of polyglutamate synthesis in PHA-L incubated in 2 p~ folate is comparable to receptor positive MA104 cells incubated in 2 nM folate (1 pmol/h/lOb cells).

5r

x

v 0 24 36 48 60 72

Time of PHA Stimulation (h) Fig 5 . Expression of FPGS mRNA by lymphocytes as a function of time of stimulation with PHA. Total RNA was isolated from lymphocytes as described in Methods. Unstimulated cells (0 h) were immediately collected and the remaining cells were stimulated for the time periods indicated. The quantity of mRNA for FPGS was determined by phosphorimaging and the data were normalized to the amount of FPGS mRNA present in unstimulated cells.

clonal antibody against the folate receptor (Campbell et al. 1991; Coney et al, 1991). and (c) the lack of a hybridization signal using a full-length cDNA for the folate receptor (Lacey et RI, 1989) (negative data not shown).

Synthesis of folylpolyglutariiates requires mitogen stimulation Unstimulated cells. even when incubated in 2 p~ 5CH3[’H]FH4 for 8 h. did not demonstrate any FPGS activity (Fig 4D). When stimulated with PHA for 48 h the metabolism of 5CHJ3H]FH4 to a folyl-penta. -hexa or -heptaglutamate was essentially simultaneous with 5CH$H]FH4 accumulation in cells labelled with 20 nM folate (Fig 4B). In contrast,

20 2000

5-Methyltetrahydrofolate (nM)

Fig 6. Effect of extracellular folate concentration on [3H]thymidine incorporation into DNA of PHA-L. PHA-L were incubated for 8 h in 0. 20 n M or 2 p~ SCHjFH4 prior to incubation for 1 h with 3 7 kBq of [3H]thymidine. Uptake of the labelled thymidine was determined after the cells were harvested, washed and counted as described in Methods.

Expression of FPGS mRNA parallels uptake of 5CH3[3H]FH4 in

Unstimulated cells had no evidence of functional FPGS activity (i.e. the ability to form folylpolyglutamates) even when the cells were pulsed with 2 p~ 5CH3[’H]FH4 to bypass transport (Fig 4D). Therefore, we quantitated mRNA for FPGS from both unstimulated lymphocytes and PHA-L stimulated for 24, 36,48, 60 and 72 h using phosphorimaging detection. mRNA for FPGS was detectable at low levels in unstimulated lymphocytes but increased 4.5-fold over the resting level to peak expression after 48 h of stimulation and diminished thereafter (Fig 5). Cells incubated for 48 h without mitogen have a 5-fold decrease in message compared to the freshly isolated cells, thus there is approximately a 20- fold difference in mRNA in cells incubated for 48 h with or without mitogen.

PHA-L

Folate uptake under physiological conditions is adequate for de novo thymidine synthesis The dU suppression test has been used to assess the adequacy of cellular folate for the de novo synthesis of thymidine nucleotides (Das & Hoffbrand, 1970: Hayman & van der Weyden, 1980). Since folate uptake by PHA-L was only 0.04 pmol/h/lO“ cells compared to 1 pmol/h/106 cells in receptor positive MA104 cells, we assessed the functional adequacy of the folate pool in PHA-L incubated with physiological or pharmacological concentrations of 5CH3[3H]FH4. Cells pulse labelled with [3H]TdR in the absence of any 5CH3FH4 incorporated nearly twice as much radiolabel as those incubated 8 h in 20 nM or 2 p~ 5CH3FH4 (Fig 6). Significantly, even though cells incubated in 2 ~ L M 5CH3FH4 have 30-40-fold more folate than cells incubated for 8 h in 20 nM folate, the degree of dU suppression is equivalent. Further, incubation of cells with either 10 ~ L M dU, or with 10 p~ dU and 20 nM 5CHjFH4 had no significant effect on [3H]TdR. This latter observation provides more evidence that folate, not the dUMP pool, is rate limiting in the synthesis of thymidine.

DISCUSSION

We chose to study the accumulation and metabolism of 5CHJ3H]FH4 in mitogen stimulated human lymphocytes because this model has been previously characterized with respect to uptake of supraphysiological concentrations of folic acid and some analogues (Das & Hoffbrand, 1970; Matthews & Wickramasinghe, 1986). PHA-Lprovide a model system in which studies of 5CH3FH4 and its relationship to total folate accumulation and thymidine nucleotide synthesis can be studied in primary culture. Our results have confirmed prior


work and have extended our understanding of the mechan- isms of uptake, transport and metabolism of the physiological plasma folate, 5CH3FH+ There is a direct relationship between the timing of mitogen stimulation and the expression of FPGS mKNA.

Unlike MA104 cells (monkey kidney epithelial cells), IGROV cells (ovarian carcinoma), KB cells (nasopharyngeal carcinoma). and other cells having a folate receptor, the process of accumulation in lymphocytes appears to be a receptor independent mechanism more akin to the reduced folate transport system well defined in murine leukaemic models as reviewed by Sirotnak (1985). Evidence for this includes the lack of folate receptor by immunoblotting and the absence of mRNA for the folate receptor by Northern blotting. The absence of the folate receptor was demonstrated functionally. In cells expressing the folate receptor accumulation proceeds by the following four steps which have been called potocytosis: (1) binding of folate to the receptor, (2) internalization of the receptor-ligand complex, ( 3 ) release of the ligand from the receptor, and (4) transport of folate via an anion carrier to the cytoplasm. Specific blocking of these steps has been accomplished using folic acid to block binding of 5CH3[3H]FH4. an ionophore, monensin, to block release of the ligand from the receptor and probenecid to block the anion carrier transport mechanism. Folic acid and monensin did not block transport and intracellular accumulation of 5CHJ3H]FH4 whereas probenecid inhibition was nearly complete (Fig 3 ) .

Regardless of the extracellular folate concentration, polyglutamation occurs rapidly and can be seen within the first hour of incubation. There is an inverse relationship between the extracellular concentration of folate and the number of glutamyl chains added to the polyglutamate. At an extracellular concentration of 20 n M there is very little G h l detected with a predominance of Glub and greater formed (Fig 4). A 10-fold increase in the extracellular folate concentration (200 nM) leads to the formation of a predominance of Glu5 and Gluo with some Glul and essentially none greater than Glur, (data not shown). When the extracellular folate concentration is increased to 2 PM there is essentially no polyglutamation offolate beyond Glu5 (Fig 4). This pattern of polyglutamation for each condition of extracellular folate concentration holds true over a 24 h time period. We conclude that in PHA-L incubated with physiological concentrations of folate, transport is limiting relative to FPGS. Incubating cells in 2 p~ extracellular folate results in the synthesis of 3-4 pmol of Glu, in 6 h ( - 0.5-1 pmol/h) a value similar to that seen in MA104 cells growing in only 2 nM folate.

Our investigation of the expression of mRNA for FPGS showed that PBMCs isolated directly from blood have some detectable mKNA which declines when cells are cultured in the absence of mitogen. This may be due to contamination with non-T mononuclear cells. However, PHA-L show an increase in the amount of expression. The pattern of appearance and disappearance of FPGS mRNA correlates well with uptake of 5CHj[3H]FH4 and formation of polyglutamates (Figs 4 and 5).

Accumulation of 5CH3FH4 by PHA-L at physiological (20

n M ) extracellular concentrations reduces the incorporation of [3H]TdR into DNA as effectively as pharmacological concentrations (2 p M ) (Fig 6). Deoxyuridine has no effect on the incorporation of [3H]TdR into DNA. These findings confirm that the endogenous folate pool is rate limiting for thymidylate synthesis. Moreover, the relatively slow accumulation of folate when the cells are pulsed with physiological amount of 5CH$H]FH4 was sufficient for normal folate homeostasis.

PHA-L provide a convenient model for the study of receptor independent folate accumulation and metabolism. We have demonstrated a close association between mitogen stimulation of lymphocytes and the activation of cellular machinery necessary for the uptake of folate and its intracellular conversion to folylpolyglutamates by the action of FPGS. This work builds on prior studies and provides a framework from which further investigations of folate and antifolate metabolism can be applied to lymphoid lines that have been either virally transformed or are malignant.

ACKNOWLEDGMENTS

This work was supported in part by the Texas Division of the American Cancer Society Clinical Oncology Fellowship, Daniel Fort: by The King Foundation of Dallas, Richard Lark: by NCI grants, Barry Shane, CA41991, and Barton Kamen, R01-CA52625; and by the Burroughs Wellcome Founda- tion, Barton Kamen.

The authors thank Patricia Ellisor for secretarial support.

REFERENCES

Anderson, R.G.W., Kamen. B.A., Rothberg. K.G. & Lacey. S.W. (1992) Potocytosis: sequestration and transport of small mol- ecules by caveolae. Scienre. 255, 410-411.

Campbell. I.G., Jones, T. & Trowsdale, J. (1991) High-affinity folate binding protein is a marker for ovarian cancer. Cancer Research,

Coney, L.R.. Tomassetti. A.. Carayannopoulos. L.. Frasca. V.. Kamen. B.A., Colnaghi, M. & Zurawski. V.R. (1991) Cloning of a tumor- associated antigen: MOV-18 and MOV-19 antibodies recognize a folate-binding protein. Cancer Research. 51, 6125-6132.

Das, K.C. & Hoffbrand. A.V. (1970) Studies of folate uptake by phytohaemagglutinin-stimulated lymphocytes. British Journal of Haernatology, 19, 203-221.

Farber. S., Diamond, L.K.. Mercer. R.D.. Sylvester, R.F. & Wolff, J.A. ( 1 948) Temporary remissions in acute leukemia in children produced by folic acid antagonist, 4-aminopteroyl-glutamic acid (aminopterin). New England Journal of Medicine, 238 , 787-7533,

Garrow. T.A.. Admon. A. & Shane. B. (1992) Expression cloning of a human cDNA encoding folylpoly(y-glutamate) synthetase and determination of its primary structure. Proceedings of the Natioricil AcaderriyofSriericesofthe UriitedStatesofAnicrica. 89, 91 51-91 55.

Giannini. E.H., Brewer, E.J.. Kuzmina, N.. Shaikov, A., Maximov. A., Vorontsov, I.. Fink, C.W., Newnian, A.J., Cassidy, J.T.&Zemel, L.S. ( 1 992) Methotrexate in resistant juvenile rheumatoid arthritis. New England Journal of Medicine. 326, 1043-1049.

Hayman. R.J. & van der Weyden, M.B. (1980) Phytohemagglutinin- stimulated normal human peripheral blood lymphocytes in folate- depleted medium: a n in vitro model for megaloblastic hemopoiesis.

Kamen, B.A. ( 1 987) Folic acid antagonists. Mctabofisnr and Artiori of

51, 5329-5338.

Blood. 55, 863-865.


Anti-cancer Drugs (ed. by G. Powis and R. A. Prough). pp. 141- 162. Taylor and Francis, London.

Kamen. B.A.. Smith, A.K. & Anderson, R.G.W. (1991) The folate receptor works in tandem with a probenecid-sensitive carrier in MA104 cells in vitro. Journal of Clinical Investigation, 87, 1442- 1449.

Kamen, B.A.. Wang. M-T., Streckfuss. A.]., Peryea. X. & Anderson. R.G.W. ( 1 988) Delivery of folates to the cytoplasm of MA104 cells is mediated by a surface membrane receptor that recycles. lournill of Biological Clieniistrg. 263, 13602-1 3609.

Lacey, S.W.. Sanders, J.M.. Rothberg, K.G., Anderson, R.G.W. 81 Kamen. B.A. (1 989) Complementary DNA for the folate binding protein correctly predicts anchoring to the membrane by glycosyl- phosphatidylinositol. Journal ofClinical Investigation, 84, 71 5-720.

Matthews. J.H. & Wickramasinghe, S.N. (1986) Acquired folate deficiency in phytohaemagglutinin-stimulated human lymphocytes. British Journal of Haematology, 63. 281-291.

Sambrook, 1.. Fritsch. E.F. & Maniatis. T. (1989) Molecular Clotling: a laboratory manual. 2nd edn. pp. 7.10-7.11. Cold Spring Harbor Laboratory Press.

Sirotnak. F.M. (1985) Obligate genetic expression in tumor cells of a fetal membrane property mediating 'folate' transport: biological

significance and implications for improved therapy of human cancer. Cafirer Research. 45, 3992-4000.

van der Weyden. M.B.. Hayman. R.J., Rose, I S . & Brumley. 1. (1 99 l a ) Folate-deficient human lymphoblasts: changes in deoxy- nucleotide metabolism and thymidylate cycle activities. European Journal of Haematologg, 47, 109-1 14.

van der Weyden. M.B., Rose, I S . & Newitt. P. (1991b) Folate- deficient human lymphoblasts: changes in de novo purine and pyrimidine synthesis and phosphoribosylpyrophosphate. European Journal of Haematologg. 47, 2 13-2 18.

Weinstein, G.D.. Jeffes. E. & McCullough. J.L. (1990) Cytotoxic and immunologic effects of methotrexate in psoriasis. Journal oj Investigative Dermatology, 95, 49s-52s.

Weitman. S.D.. Lark, R.H.. Coney, L.R., Fort, D.W., Frasca, V.. Zurawski. V.R. & Kamen, B.A. (1992a) Distribution of the folate receptor GP38 in normal and malignant cell lines and tissues. Curver Researrh. 52, 3396-3401.

Weitman, S.D.. Weinberg. A.G., Coney, L.R., Zurawski, V.R., Jennings. D.S. & Kamen, B.K. (1992b) Cellular localization of the folate receptor: potential role in drug toxicity and folate homeostasis. Cancer Research. 52, 6708-671 1 .

Documents

Accumulation of 5-methyltetrahydrofolic acid and folylpolyglutamate synthetase expression by mitogen stimulated human lymphocytes