Determination of folate transport pathways in cultured rat proximal tubule cells

Chemico-Biological Interactions 114 (1998) 15 – 31

Determination of folate transport pathways in cultured rat proximal tubule cells Pankaj K. Sikka, Kenneth E. McMartin * Department of Pharmacology and Therapeutics, Louisiana State Uni6ersity Medical Center, Shre6eport LA 71130, USA Accepted 20 March 1998

Abstract Deficiency of the vitamin folic acid has recently been linked with increased incidence of neural tube defects and of cardiovascular disease, through elevated plasma homocysteine levels. The kidney has an important role in conserving folate to counteract development of deficiency. Urinary folate excretion is regulated by the degree of reabsorption of folate by the proximal tubule cell. To evaluate an in vitro model for studies of the regulation of urinary folate excretion, the present studies examined the transport of 5-methyltetrahydrofolate (5-CH3-H4PteGlu), the primary form of folate in the glomerular filtrate, by normal rat proximal tubule (RPT) cells in confluent monolayer cultures. Specific binding of 5-CH3H4PteGlu to the apical membrane was saturable (KD = 27 nM), but intracellular transport was not saturated up to 100 nM concentrations. 5-CH3-H4PteGlu transport was decreased 50% by concentrations of folic acid that completely blocked 5-CH3-H4PteGlu binding by the apical folate receptor. Probenecid (10 mM), an anion exchange (reduced folate carrier) inhibitor, reduced 5CH3-H4PteGlu transport by 50% without significantly affecting binding. Aspirin (3 mM) did not alter 5-CH3-H4PteGlu transport, but significantly enhanced the inhibition due to probenecid. Similarly, indomethacin (5 mM) potentiated the inhibition of 5-CH3-H4PteGlu transport by probenecid. These data suggest that RPT cells take up 5-CH3-H4PteGlu by both the folate receptor and the reduced folate carrier, implying a role for both pathways in regulating urinary folate excretion. © 1998 Elsevier Science Ireland Ltd. All rights reserved.

* Corresponding author. Tel.: +1 318 675 7857; fax:+ 1 318 675 7857. 0009-2797/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0009-2797(98)00038-6

16

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

Keywords: Folate receptor; Reduced folate carrier; Urinary folate excretion; 5Methyltetrahydrofolate

1. Introduction Folate coenzymes participate in transfer of one-carbon units in the metabolism of purines, pyrimidines and amino acids, thus playing key roles in intracellular metabolism of nucleic acids and proteins. Folate deficiency in severe cases results in megaloblastic anemia [1], but recent studies have linked marginal deficiency with increased incidence of neural tube defects [2] and elevated plasma homocysteine levels [3], with associated cardiovascular disease [4]. Although folate deficiency can occur as a result of inadequate dietary intake, other processes such as malabsorption, altered hepatic metabolism and increased elimination can also contribute to folate deficiency [5]. The renal regulation of folate is one of the important mechanisms by which the body physiologically maintains folate homeostasis. Increased urinary folate excretion with decreased plasma folate levels is known to occur in pregnancy [6], with oral contraceptive use [7] and with excess consumption of alcohol [8,9]. The mechanisms by which the kidney regulates folate homeostasis are not well defined. 5-Methyltetrahydrofolate (5-CH3-H4PteGlu) is the primary form of folate that is transported in the blood and tissues [10]. The cellular uptake of this vitamin is complex and believed to occur via two distinct cellular transport systems. One system is receptor-mediated [11] occurring via different isoforms of the folate receptor (folate binding protein) (FR). Although these isoforms vary in their affinities for the folate forms, in general they have higher affinity for folic acid (PteGlu) than for reduced folates like 5-CH3-H4PteGlu or 5-formyltetrahydrofolate (5-HCO-H4PteGlu) [12]. The receptor-mediated transport of 5-CH3-H4PteGlu appears to occur via a typical endocytic pathway involving endocytic vacuoles [13] or via a similar pathway involving caveolae [11], where the receptor recycles within the caveolae without dissociating from the plasma membrane. The second distinct system involves a reduced folate carrier (RFC) system which has markedly greater affinities for 5-CH3-H4PteGlu than for PteGlu [14]. This membrane carrier also accumulates methotrexate [15], thus serving as a common transport pathway for folates and many therapeutic anti-folates. The RFC can be inhibited by organic anion inhibitors like probenecid and bromosulfophthalein [16]. To prepare for studies of the renal regulation of 5-CH3-H4PteGlu excretion, we have established in vitro models, including transport by cultured human proximal tubule (HPT) cells [17,18]. In HPT cells, both the FR and the RFC participate significantly in the apically mediated transport of 5-CH3-H4PteGlu [19]. A major drawback of the studies in the human cells is that the origin of these primary cultures cannot be controlled, especially as to their nutritional, medical and pharmacological history. Therefore, the present studies have been conducted to characterize the uptake of 5-CH3-H4PteGlu by the cultured rat proximal tubule

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

17

(RPT) cell. Like HPT cells, RPT cells are normally differentiated, primary cultures which retain characteristics of the proximal tubule in vivo such as apical enzymatic activities and apically mediated sodium-dependent glucose transport [20]. RPT cells are grown in a serum-free hormonally defined medium and can be subcultured [21], both of which allow for transport studies under controlled conditions in homogeneous cultures of renal cells. The RPT cell culture model offers several advantages over the HPT cell system. The physical history of the rats prior to cell harvesting and culturing is well controlled so that inter-isolate variation is minimized. Also, since previous studies of urinary folate excretion have been conducted in rats in vivo [10,13] or in isolated perfused rat kidneys [22,23], the transport of 5-CH3-H4PteGlu needs to be characterized in cultured cells from the same species so direct comparison of results can be made. Finally, the long term effects of pharmacologic and nutritional manipulations on folate transport in renal cells can only be carried out in cells cultured from treated rats, not humans. Hence, the present studies must be conducted first so as to characterize the binding and transport of 5-CH3-H4PteGlu in normal primary cultures of RPT cells and to evaluate the transport roles of the FR and the RFC in these cells.

2. Materials and methods

2.1. Cell culture RPT cells were isolated and grown to confluency, as previously described [20]. Briefly, two male Sprague – Dawley rats (Harlan, Houston, TX) weighing between 150 and 250 g were anesthetized and the kidneys were excised, decapsulated and cut into four pieces to permit visual identification of cortical tissue. Cortical tissue was sliced away from the medulla and subjected to deoxyribonuclease –collagenase (Sigma, St. Louis, MO) digestion for three incubations of 20 min each. The solution containing isolated cells was filtered and centrifuged to yield a proximaltubule-rich cell pellet. The isolated proximal tubule cells were suspended in a serum-free growth medium (1:1 mixture of Ham’s F-12 and Dulbecco’s modified Eagle’s medium, GIBCO Laboratories, Grand Island, NY) supplemented with 0.05% bovine serum albumin (BSA, fraction V, Miles, Kankakee, IL), insulin (10 mg/ml), selenium (10 ng/ml), hydrocortisone (36 ng/ml), epidermal growth factor (10 ng/ml) (Collaborative Biomedical Research, Bedford, MA), triiodothyronine (4 pg/ml, Sigma) and glutamine (2 mM, GIBCO). The RPT cells were grown to confluency (5 – 7 days) on rat-tail collagen (Sigma) coated plastic surfaces (75-cm2 flasks, Costar, Cambridge, MA) in a humidified atmosphere containing 5% CO295% air. The RPT cells were then subcultured using trypsin-EDTA (0.05:0.02%, GIBCO) and seeded onto 24-well plates (Costar) for experimental studies. Confluent RPT cells from passage 2 were used for all the studies described, unless otherwise mentioned.

18

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

2.2. PteGlu transport and binding experiments The growth medium was aspirated and the confluent RPT cells were washed twice with 1 ml of pH 7.4 incubation buffer containing (in mM) 107 NaCl, 5.3 KCl,1.9 CaCl2, 1.0 MgCl2, 26.2 NaHCO3, 7.0 D-glucose and 20 N-2-hydroxyethylpiperazineN%2-ethanesulphonic acid (HEPES) (Sigma) [17]. The cells were incubated at 37°C with 0.5 ml of this buffer containing 25 nM [3%,5%,7,9-3H]-6S-5-CH3-H4PteGlu (30 Ci/mmol, Moravek Biochemicals, Brea, CA) for 10–240 min (time course). This 5-CH3-H4PteGlu concentration was initially chosen since the total folate plasma concentration in male Sprague – Dawley rats is about 50 nM [10]. Nonspecific controls contained a 1000-fold excess of unlabeled 5-CH3-H4PteGlu (Sigma). In some studies [14C]inulin (0.15 mCi/ml, 1.7 mCi/g, Dupont, Boston, MA) was added to the incubation buffer as a marker to determine the integrity or leakiness of the cell monolayer [18]. After the appropriate time period, the incubation buffer was aspirated and the cells were washed three times with 1 ml of cold (4°C) incubation buffer. The cells were then treated for 30 s first with 0.5 ml of an acid buffer (acetate, pH 3.0, containing 150 mM NaCl) and then with 0.5 ml of incubation buffer. The two samples were combined and saved for analysis of bound radioactivity [24]. The cells were solubilized in 0.5 ml of 0.1 N NaOH overnight and the samples were analyzed for transported radioactivity. For concentration experiments, the RPT cells were incubated for 120 min at 37°C with 6.25 – 100 nM 5-CH3-H4PteGlu. For inhibitor studies, a stock solution of probenecid (100 mM, Sigma) was prepared by dissolving it in 0.1 N NaOH and adjusting the pH to 7.4. Stock solutions of aspirin (50 mM, Sigma) and indomethacin (10 mM, Sigma) were prepared in absolute ethanol. These solutions were diluted with the incubation buffer to achieve the desired concentrations: probenecid, 0.1–20 mM [25]; aspirin, 0.1 – 6 mM [26] and indomethacin, 1–10 mM [27]. For folate analogue studies, PteGlu or 5-HCO-H4PteGlu were added to the incubation buffer to achieve concentrations of 0 – 500 nM and 0–2000 nM, respectively. For the inhibitor studies, the RPT cells were pre-incubated for 30 min with the various concentrations of these drugs, following which [3H]-5-CH3-H4PteGlu (25 nM) in incubation buffer was added for 120 min incubations as above. Incubations were carried out in duplicate; experiments were conducted on three or four separate cell preparations as noted in the figure legends.

2.3. Transport and binding analysis The amount of 3H-radioactivity was measured in the acid-wash samples (binding to apical cell membrane) and in the solubilized samples (intracellular transport) by liquid scintillation analysis. The amount of total transport and binding was determined in incubations containing only [3H]-5-CH3-H4PteGlu. Nonspecific control incubations contained a 1000-fold excess of unlabeled 5-CH3-H4PteGlu to saturate the specific binding and transport sites [17]. Specific transport or binding values were calculated by subtracting the nonspecific values from the total values. The protein content in the solubilized cells was determined by the method of Bradford [28].

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

19

2.4. Efflux experiments After aspirating the growth medium and washing the RPT cells as described above, the cells were loaded with 0.5 ml of incubation buffer containing 1000 nM CH3-H4PteGlu (labeled:unlabeled =1:9) for 120 min at 37°C. After loading the cells, the incubation medium was aspirated and the cells were rapidly washed three times with 1 ml of the incubation buffer. Fresh incubation buffer (0.5 ml), with or without 10 mM probenecid, was then added to the cells. The cells were further incubated for another 5 – 120 min at 37°C, after which the incubation buffer was collected and analyzed for effluxed radioactivity. The cells were then washed twice with 1 ml of cold (4°C) incubation buffer, incubated with acid buffer to determine apical binding, and solubilized with NaOH to determine retained cellular radioactivity by liquid scintillation analysis, as described above.

2.5. Statistical analysis Statistical analysis was carried out by using Sigma Stat (Jandel Scientific). Group data were analyzed by Student’s t-test or by ANOVA for multiple comparisons using Student – Newman – Keuls multiple comparison test for differences among the groups. The KD and Bmax values were determined by non-linear regression analysis. The level of significance for all studies was PB0.05. The values cited in the text represent the mean9S.E.M.

3. Results

3.1. Kinetics of 5 -CH3 -H4PteGlu transport and binding The time course for the intracellular transport and apical membrane binding of 5CH3-H4PteGlu (25 nM) by RPT cells is shown in Fig. 1A, B. 5-CH3-H4PteGlu transport increased linearly for 60 min (4.09 0.5 fmol/mg protein/min) and reached a near steady state by 240 min. Binding showed a similar slow increase to a steady state, which was reached by 120 min. These data may reflect the need of the 5-CH3H4PteGlu to replace a tightly bound endogenous ligand [29], which was present because the cells were not pretreated with acid buffer (in order to preserve proper tight junction configuration by the cells) [18]. The nonspecific control experiments showed that nonspecific binding and transport were minimal over 240 min, about 4 and 17% of the total values, respectively. Hence, only the specific transport and binding values are shown in subsequent experiments. The uptake of inulin was minimal (B 0.01% of the dose in 240 min), suggesting that the cells formed a confluent, nonleaky monolayer. When the transport and binding of 5-CH3-H4PteGlu was measured in RPT cells maintained for four passages, both uptake values were only about 50% of those at passage 2 (data not shown). A similar decrease in the ability of RPT cells to take up glucose as the passage number increased was noted earlier [20].

20

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

Fig. 1. Time course of intracellular transport and apical binding of 5-CH3-H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were incubated at 37°C in incubation buffer containing 25 nM [3H]-5-CH3-H4PteGlu for the indicated time periods (incubations for total uptake). Nonspecific control incubations contained a 1000-fold excess of unlabelled 5-CH3-H4PteGlu. Transport (A) and binding (B) values were determined after the incubation as described in Section 2. Specific uptake values were calculated by subtracting the nonspecific values from the total values. Points represent the means 9 S.E.M. of three experiments.

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

21

When confluent monolayers of RPT cells were incubated for 120 min with various concentrations of [3H]-5-CH3-H4PteGlu (6–100 nM), the transport was generally linear with no evidence of saturation by the highest concentration (Fig. 2). Apical binding was saturated at the 50 nM concentration. Transport and binding values were similar up to a 25-nM concentration, but then the transport of 5-CH3-H4PteGlu became quantitatively greater than its binding. Analysis of these kinetic data by nonlinear regression showed that the binding constant (KD) was 27 9 14 nM and the maximal binding (Bmax) was 3.29 0.5 fmol/mg protein/min.

3.2. Effect of folate analogues on 5 -CH3 -H4PteGlu uptake These experiments were conducted to determine whether the folate compounds PteGlu and 5-HCO-H4PteGlu, which have widely different affinities for the FR and the RFC, affect the transport or binding of 5-CH3-H4PteGlu by the RPT cell. PteGlu has a greater affinity for the FR than 5-CH3-H4PteGlu [12] with essentially no affinity for the RFC [14] and should specifically inhibit the receptor-mediated uptake of 5-CH3-H4PteGlu [24]. 5-HCO-H4PteGlu has a much lower affinity for the FR than 5-CH3-H4PteGlu, but similar affinity for the RFC [30], so its interaction

Fig. 2. Concentration dependency of specific transport and binding of 5-CH3H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were incubated for 120 min at 37°C in incubation buffer containing increasing concentrations of [3H]-5-CH3H4PteGlu for the indicated time periods (incubations for total uptake). Nonspecific control incubations contained a 1000-fold excess of unlabelled 5-CH3H4PteGlu. Transport and binding values were determined after the incubation as described in Section 2. Specific uptake values were calculated by subtracting the nonspecific values (9.5 and 5.4% of total transport and binding, respectively, data not shown) from the total values. Points represent the means 9 S.E.M. of three experiments.

22

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

Fig. 3. Effect of PteGlu on transport and binding of 5-CH3-H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were preincubated for 30 min at 4°C in incubation buffer containing the indicated concentrations of PteGlu. Then, the RPT cells were further incubated with 25 nM [3H]-5-CH3H4PteGlu for 120 min at 37°C. Transport and binding values of 5-CH3-H4PteGlu were determined after the incubation as described in Section 2. Points represent the means 9 S.E.M. of three experiments. Asterisks (*) indicate significant difference from the control value (0 nM PteGlu) (P B0.05).

with 5-CH3-H4PteGlu uptake should be independent of the receptor. After preincubation with the unlabeled folate analogues to saturate the transport systems, RPT cells were incubated with 25 nM [3H]-5-CH3H4PteGlu in the presence of the other folate. Results in Fig. 3 show that folic acid from 5 to 500 nM significantly inhibited both the binding and transport of 5-CH3-H4PteGlu. At equimolar concentrations of PteGlu and 5-CH3-H4PteGlu, 5-CH3-H4PteGlu binding and transport values were decreased by 87 and 38%, respectively. At the highest concentration, PteGlu completely blocked 5-CH3-H4PteGlu binding, but could only inhibit 5-CH3H4PteGlu transport by 50%. The data in Fig. 4 show that the binding, but not the transport, of 5-CH3-H4PteGlu was inhibited by lower concentrations of 5-HCOH4PteGlu (25 and 100 nM). At 500 nM and above, 5-HCO-H4PteGlu significantly decreased both the binding and transport of 5-CH3-H4PteGlu.

3.3. Effect of probenecid and organic anions on 5 -CH3 -H4PteGlu uptake Probenecid is a inhibitor of organic anion transport in the kidney [31] and has been shown to inhibit the transport of folates by the RFC in various cell systems [32,33]. To study the role of the RFC in 5-CH3-H4PteGlu uptake by the RPT cells, confluent monolayers of cells were pre-incubated with various concentrations of

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

23

probenecid (0 – 20 mM) for 30 min, then for 120 min with 25 nM [3H]-5-CH3H4PteGlu along with probenecid. At high concentrations, probenecid markedly inhibited the intracellular transport of 5-CH3-H4PteGlu, with decreases of 52 and 77% at 10 and 20 mM, respectively (Fig. 5). Probenecid did not significantly affect the apical membrane binding of 5-CH3-H4PteGlu at any concentration, except the highest (20 mM), where binding was decreased by 43%. Low concentrations of probenecid had little effect on either parameter, except for a slight stimulation of transport at 1 mM. Aspirin is a substrate for the organic anion secretory carrier in the renal proximal tubule that is commonly inhibited by probenecid [34]. Whether this secretory carrier is related to the RFC in the kidney has not been determined. The effects of aspirin and probenecid were studied alone or in combination in order to elicit the role of the renal anion transport system in the transport of 5-CH3-H4PteGlu by RPT cells. Confluent monolayers were preincubated for 30 min with aspirin (0.1–6 mM), probenecid (10 mM) or aspirin (3 mM) plus probenecid (10 mM), followed by a 120 min incubation with 25 nM [3H]-5-CH3-H4PteGlu along with these agents. Concentrations of aspirin alone up to 3 mM did not affect the binding or transport of 5-CH3-H4PteGlu, although higher concentrations affected the viability of RPT cells

Fig. 4. Effect of 5-HCO-H4PteGlu on transport and binding of 5-CH3-H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were preincubated for 30 min at 4°C in incubation buffer containing the indicated concentrations of 5-HCO-H4PteGlu. Then, the RPT cells were further incubated with 25 nM [3H]-5-CH3-H4PteGlu for 120 min at 37°C. Transport and binding values of 5-CH3-H4PteGlu were determined after the incubation as described in Section 2. Points represent the means 9S.E.M. of three experiments. Asterisks (*) indicate significant difference from the control value (0 nM 5-HCO-H4PteGlu) (P B0.05).

24

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

Fig. 5. Effect of probenecid on transport and binding of 5-CH3-H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were preincubated for 30 min at 37°C in incubation buffer containing the indicated concentrations of probenecid. Then, the RPT cells were further incubated with 25 nM [3H]-5-CH3-H4PteGlu for 120 min at 37°C. Transport and binding values of 5-CH3-H4PteGlu were determined after the incubation as described in Section 2. Points represent the means 9S.E.M. of three experiments. Asterisks (*) indicate significant difference from the control value (0 mM probenecid) (P B0.05).

(data not shown). The data in Fig. 6 show that probenecid alone (10 mM) inhibited 5-CH3-H4PteGlu transport by 40%, while aspirin alone (3 mM) did not affect transport. However, when aspirin was combined with probenecid, 5-CH3-H4PteGlu transport was inhibited by 61%, which was significantly greater than in cells treated with probenecid alone. Neither aspirin nor probenecid, alone or in combination, affected the apical binding of 5-CH3-H4PteGlu. Indomethacin, like aspirin, is transported by the organic anion system and is a potent inhibitor of prostaglandin synthesis [35]. To test whether the potentiation of probenecid’s inhibition of 5-CH3-H4PteGlu transport is a specific effect of aspirin, studies were conducted to determine the combined effects of probenecid and indomethacin on 5-CH3-H4PteGlu uptake by RPT cells. Confluent monolayers were preincubated for 30 min with indomethacin (5 mM), probenecid (10 mM) or indomethacin (5 mM) plus probenecid (10 mM), followed by a 120-min incubation with 25 nM [3H]-5-CH3-H4PteGlu along with these agents. As above, probenecid inhibited the transport by 43%, but not the binding of 5-CH3-H4PteGlu (Fig. 7). However, when indomethacin was combined with probenecid, the transport of 5-CH3-H4PteGlu was inhibited by 72%, which was significantly greater than that with probenecid alone (Fig. 7). The binding of 5-CH3H4PteGlu was not affected by the combined use of indomethacin and probenecid. Unlike the situation with

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

25

aspirin, indomethacin alone significantly increased the transport of 5-CH3H4PteGlu, by about 60% (Fig. 7).

3.4. Effect of probenecid on 5 -CH3 -H4PteGlu efflux The efflux of the folate analogue methotrexate has been shown to occur by anion carrier-like mechanisms in L1210 cells [25]. To test whether probenecid affected the transport of 5-CH3-H4PteGlu from inside the cell to the extracellular medium, a concentration gradient was first established from the outside to the inside of the cells by incubating confluent monolayers of RPT cells with 1000 nM [3H]-5-CH3H4PteGlu for 120 min. To measure the efflux of 5-CH3-H4PteGlu from these preloaded cells, the original incubation buffer was removed and replaced with fresh buffer containing probenecid (0 or 10 mM) for a further 120 min incubation. Probenecid inhibited the efflux of 5-CH3-H4PteGlu by about 30% over the 120 min period (Fig. 8). The effect of probenecid was significant at time periods over 15 min. This inhibition of efflux was reinforced by the fact that probenecid significantly increased the amount of [3H]-5-CH3H4PteGlu remaining in the RPT cell after 120 min (from 333 9 30 fmol/mg protein in controls to 482944 fmol/mg, PB 0.05).

Fig. 6. Effect of aspirin and probenecid on transport and binding of 5-CH3-H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were preincubated for 30 min at 37°C in incubation buffer in the absence (control) or presence of aspirin (3 mM), probenecid (10 mM), or aspirin (3 mM) plus probenecid (10 mM). Then, the RPT cells were further incubated with 25 nM [3H]-5-CH3-H4PteGlu for 120 min at 37°C. Transport and binding values of 5-CH3-H4PteGlu were determined after the incubation as described in Section 2. C, control; A, aspirin; P, probenecid; A + P, aspirin plus probenecid. Bars represent the means 9 S.E.M. of four experiments. Bars with different letters (a – d for transport and x – y for binding) were significantly different from each other (ANOVA, Student – Newman – Keuls comparison test, PB 0.05).

26

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

Fig. 7. Effect of indomethacin and probenecid on transport and binding of 5-CH3 H4PteGlu by cultured RPT cells. Confluent monolayers of RPT cells were preincubated for 30 min at 37°C in incubation buffer in the absence (control) or presence of indomethacin (5 mM), probenecid (10 mM), or indomethacin (5 mM) plus probenecid (10 mM). Then, the RPT cells were further incubated with 25 nM [3H]-5-CH3H4PteGlu for 120 min at 37°C. Transport and binding values of 5-CH3-H4PteGlu were determined after the incubation as described in Section 2. C, control; I, indomethacin; P, probenecid; I +P, indomethacin plus probenecid. Bars represent the means 9S.E.M. of three experiments. Bars with different letters (a-d for transport and x–y for binding) were significantly different from each other (ANOVA, Student – Newman–Keuls comparison test, PB 0.05).

4. Discussion 5-CH3-H4PteGlu, after filtration by the renal glomerulus, is presented to the proximal tubule cell, which normally reabsorbs 90–95% of the lumenal folate [22]. Decreased reabsorption of 5-CH3-H4PteGlu will result in an increase in urinary folate excretion [23], which can contribute to the development of folate deficiency. Reabsorption of 5-CH3-H4PteGlu occurs via apically mediated transport of folate into the proximal tubule cell [36], apparently by two protein-based transport systems, the FR and the RFC [19]. Hence, changes in the transport by either of these pathways could result in decreased reabsorption and increased excretion. The present studies have shown that nutritional and pharmacologic studies of the regulation of urinary folate excretion can be conducted in the RPT cell culture model, since the folate transport pathways in the rat kidney are similar to those in the human kidney. Several lines of evidence in the present studies have indicated that both the FR and the RFC participate materially in the uptake of 5-CH3-H4PteGlu by the RPT cell. First, kinetic experiments showed that transport continued to occur at concentrations above those that saturated the FR, suggesting the role of an additional

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

27

pathway. Apical binding by RPT cells was saturated at concentrations consistent with KD values reported for the FR in other tissues (1–50 nM) [12], while transport could not be saturated at concentrations up to 100 nM. Hence, transport of 5-CH3-H4PteGlu by RPT cells could occur at low concentrations (B 25 nM) by the FR, while another pathway would mediate transport at high concentrations. 5-CH3-H4PteGlu transport was suppressed by a 1000-fold excess of unlabelled 5-CH3-H4PteGlu, indicating that the second pathway was a saturable, low affinity pathway like the RFC. The Km for 5-CH3-H4PteGlu transport by the RFC in other cells is generally 0.1 – 1 mM [37], which is consistent with these data. Second, inhibitor experiments showed that PteGlu completely blocked the apical binding of 5-CH3-H4PteGlu, while decreasing its transport into RPT cells by only 50%. PteGlu, which has a much greater affinity for the FR than does 5-CH3-H4PteGlu, but essentially no affinity for the RFC, blocks FR-mediated transport in various cells [24,38]. Hence in the RPT cell, the FR appeared to mediate about 50% of the transport of 5-CH3-H4PteGlu, while another pathway (not blocked by PteGlu) must be responsible for the remaining 50%. Third, probenecid, which inhibits the RFC-mediated transport of 5-CH3-H4PteGlu in human lymphocytes [33], decreased 5-CH3-H4PteGlu transport in the RPT cell by 40–50% at a concentration (10 mM) that did not inhibit apical binding. These data suggest that probenecid did not

Fig. 8. Effect of probenecid on efflux of 5-CH3-H4PteGlu from cultured RPT cells. Confluent monolayers of RPT cells were incubated for 120 min at 37°C in incubation buffer containing 1000 nM of 5-CH3-H4PteGlu (labeled:unlabeled = 1:9). Then, the RPT cells were rapidly washed and further incubated at 37°C for the indicated time periods in the presence or absence of 10 mM probenecid. After the incubations, the incubation buffer was collected and analyzed for effluxed radioactivity as described in Section 2. Points represent the means 9 S.E.M. of three experiments. Asterisks (*) indicate significant difference from the control value (0 mM probenecid) (P B0.05).

28

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

affect the FR-mediated pathway in the RPT cell, rather that it decreased the RFC-mediated transport of 5-CH3-H4PteGlu. We have obtained similar results when examining the transport of 5-CH3H4PteGlu by primary cultures of HPT cells [19]. In those studies, PteGlu decreased apical binding by 95%, while decreasing transport by about 50%. Probenecid (10 mM) decreased apical transport by 50%, without an effect on apical binding. Colchicine, which disrupts vesicular trafficking such as would occur in an endocytic process like that mediated by the FR, decreased apical transport by 50–70%, without affecting the basolateral uptake of 5-CH3-H4PteGlu. The addition of probenecid to incubations containing either PteGlu or colchicine produced a nearly complete block in apical transport. These studies showed that probenecid affected the RFC, not the FR, since the latter activity was blocked already by PteGlu or colchicine. Hence in both RPT and HPT cells, the apically-directed transport of 5-CH3-H4PteGlu was mediated by both the FR and the RFC, with roughly equal proportions at 5-CH3-H4PteGlu concentrations in the physiologic range (10–25 nM). HPT and RPT cells represent primary cultures of normal tissues in which the cells retain normal differentiation and functional and structural characteristics of the proximal tubule in vivo [20,39]. These primary cultures may utilize both the FR and the RFC for 5-CH3-H4PteGlu transport because the FR is not expressed at high levels in normal cells [40]. Indeed the studies by Fort et al. [33] in normal human lymphocytes have demonstrated transport of folates by the RFC, independent of any FR activity. The studies with aspirin and indomethacin produced interesting results, with one interpretation being that prostaglandins may be involved in folate transport. Aspirin and indomethacin are weak organic anions that inhibit the renal organic anion transporter by acting as competitive substrates [41]. In combination with probenecid, both aspirin and indomethacin significantly exacerbated the inhibitory effect of probenecid on 5-CH3-H4PteGlu transport. Since probenecid was given in high concentrations that should have blocked both the RFC and the organic anion carrier, the potentiation by these anti-inflammatory drugs would seem to occur by a mechanism independent of their action on the anion carrier. Aspirin and indomethacin are also known to be potent inhibitors of prostaglandin synthesis [42]. The potentiation of the probenecid-induced decrease in 5-CH3-H4PteGlu transport by these drugs may result from their inhibition of prostaglandin synthesis. Studies by Henderson and Tsuji [43] have also suggested a role for prostaglandins in folate transport, since the efflux of the folate analogue methotrexate by the L1210 cell is potently inhibited by prostaglandin A, [43] and since PteGlu is also a substrate for the methotrexate efflux pathway. If this pathway is likewise affected in RPT cells, a decreased synthesis of prostaglandins might lead to increased folate efflux and, in combination with probenecid to block the RFC-mediated influx, decreased intracellular folate accumulation. Alternatively, the effects of aspirin and indomethacin in combination with probenecid may indirectly result from an effect of the latter on aspirin and indomethacin transport. As noted above, aspirin and indomethacin are weak organic anions that act as competitive substrates for the organic anion carrier in the kidney. Since this transporter is distinct from the RFC,

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

29

aspirin and indomethacin by themselves would not be able to inhibit 5-CH3H4PteGlu transport by the RFC. In our experiments, the high concentrations of probenecid would have blocked several renal transport systems, including the transport of aspirin and indomethacin by the organic anion transporter. Then sufficient aspirin or indomethacin might be present extracellularly to affect 5-CH3H4PteGlu transport by the RFC. Further studies will be needed to distinguish whether the effect of these drugs on folate transport is related to their structure as anions or their ability to inhibit prostaglandin synthesis. In the RPT cells, probenecid decreased the efflux of preloaded [3H]-5-CH3H4PteGlu, suggesting that an anion carrier transported 5-CH3-H4PteGlu across the apical membrane into the lumen. However, previous studies have shown that 5-CH3-H4PteGlu can be metabolized by renal cells to other folate and nonfolate derivatives [22,44]. In the present studies, the chemical form of the effluxed radioactivity was not characterized, so the anion carrier may have been transporting a 5-CH3-H4PteGlu metabolite. In the perfused kidney, secretion of other folates occurs via a probenecid-sensitive transport pathway [22]. In the HPT cell, preloaded 5-CH3-H4PteGlu is transported out of the cell, with about 60% as unchanged 5-CH3-H4PteGlu, the rest as metabolites [44]. These results suggest that the renal proximal tubule cell may secrete 5-CH3-H4PteGlu or its metabolites across the apical membrane by an anion carrier, possibly the RFC. One drawback of the current studies is that the high concentrations of probenecid ( \ 1 mM) would be nonselective such that many renal anion transport systems would be affected, including the one that transports 5-CH3-H4PteGlu (the RFC) and those that transport other organic anions. Hence, conclusions as to the nature of the anion transporter that carries 5-CH3-H4PteGlu cannot be made. However, by considering the results of all of the studies, we can conclude that 5-CH3-H4PteGlu is transported by the RPT cell by both the FR and by an organic anion transporter, most likely the RFC. Because these results resemble those in human cells, studies that determine how nutritional and pharmacologic manipulations alter folate transport in RPT cells can now be used to assess the mechanisms by which these manipulations alter urinary folate excretion.

Acknowledgements This work was supported by NIH grant R01-AA05308 and by a grant from Louisiana State University Medical Center Biomedical Research Foundation of Northwest Louisiana. The authors wish to thank Geneva Evans for helping with the preparation of the manuscript.

References [1] N. Colman, Folate deficiency in humans. In: H.H. Draper (Ed.), Advances in Nutritional Research., Vol 1. Plenum Press, New York, 1977, pp. 77 – 124.

30

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

[2] MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet 338 (1991) 131 – 137. [3] L.E. Brattstrom, B. Israelsson, J.O. Jeppsson, B.L. Hultberg, Folic acid — an innocuous means to reduce plasma homocysteine, Scand. J. Clin. Lab. Invest. 48 (1988) 215 – 221. [4] R. Clarke, L. Daly, K. Robinson, E. Naughten, S. Cahalane, B. Fowler, I. Graham, Hyperhomocysteinemia an independent risk factor for vascular disease, New. Engl. J. Med. 324 (1991) 1149–1155. [5] D.G. Weir, P.G. McGing, J.M. Scott, Folate metabolism, the enterohepatic circulation and alcohol, Biochem. Pharmacol. 34 (1985) 1–7. [6] M.J. Landon, F.E. Hytten, The excretion of folate in pregnancy, J. Ob. Gyn. Brit. Commonwealth. 78 (1971) 769–776. [7] A.M. Shojania, The effect of oral contraceptives on folate metabolism: III: Plasma clearance and urinary folate excretion, J. Lab. Clin. Med. 85 (1975) 185 – 190. [8] K.E. McMartin, Increased urinary folate excretion and decreased plasma folate levels in the rat atter acute ethanol treatment, Alcoholism Clin. Exp. Res. 8 (1984) 172 – 178. [9] K.E. McMartin, T.D. Collins, B.H. Eisenga, et al., Effects of chronic ethanol and diet treatment on urinary folate excretion and development of folate deficiency in the rat, J. Nutr. 119 (1989) 1490–1497. [10] B.H. Eisenga, T.D. Collins, K.E. McMartin, Effects of acute ethanol on urinary excretion of 5-methyltetrahydrofolic acid and folate derivatives in the rat, J. Nutr. 119 (1989) 1498 – 1505. [11] K.G. Rothberg, Y. Ying, J.F. Kolhouse, et al., The glycophospholipid-linked folate receptor internalizes folate without entering the clathrin-coated pit endocytic pathway, J. Cell. Biol. 110 (1990) 637–649. [12] X. Wang, F. Shen, J.H. Freisheim, L.E. Gentry, M. Ratnam. Differential stereospecificities and affinities of folate receptor isoforms for folate compounds and antifolates., Biochem. Pharm. 44 (1992) 1898–1901. [13] H. Birn, J. Selhub, E.I. Christensen, Internalization and intracellular transport of folate-binding protein in rat kidney proximal tubule, Am. J. Physiol. 264 (1993) C302 – C310. [14] G.B. Henderson, M.R. Suresh, K.S. Vitols, et al., Transport of folate compounds in L1210 cells: kinetic evidence that folate influx proceeds via the high-afffinity transport system for 5-methyltetrahydrofolate and methotrexate, Canc. Res. 46 (1986) 1639 – 1643. [15] G. Jansen, G.R. Westerhof, G. Rijksen, et al., Interaction of N-hydroxy(sulfo)succinimide active esters with the reduced folate/methotrexate transport system from human leukemic CCRF – CEM cells, Biochim Biophys Acta 985 (1989) 266 – 270. [16] E.E. Sierra, K.E. Brigle, M.J. Spinella, I.D. Goldman, Comparison of transport properties of the reduced folate carrier and folate receptor in murine L1210 leukemia cells, Biochem. Pharmacol. 50 (1995) 1287–1294. [17] K.E. McMartin, K.M. Morshed, D.J. Hazen-Martin, D.A. Sens, Folate transport and binding by cultured human proximal tubule cells, Am. J. Physiol. 263 (1992) F841 – 848. [18] K.M. Morshed, K.E. McMartin, Transient alterations in cellular permeability in cultured human proximal tubule cells: implications for transport studies, In Vitro Cell. Dev. Biol. 31 (1995) 107–114. [19] K.M. Morshed, D.M. Ross, K.M. McMartin, Folate transport proteins mediate the bidirectional transport of 5-methyltetrahydrofolate in cultured human proximal tubule cells, J. Nutr. 127 (1997) 1137–1147. [20] P.K. Sikka, K.E. McMartin, Normal rat kidney proximal tubule cells in primary and multiple subcultures, In Vitro Cell Dev. Biol-Animal 32 (1996) 285 – 291. [21] P.B. Hatzinger, J.L. Stevens, Rat kidney proximal tubule cells in defined medium: The roles of cholera toxin, extracellular calcium and serum in cell growth and expression of g-glutamyltransferase, In Vitro Cell Dev. Biol. 25 (1989) 205 – 212. [22] R.T. Muldoon, B.H. Eisenga, K.M. Morshed, K.E. McMartin, 5-Methyltetrahydrofolate is reabsorbed and metabolized in isolated perfused rat kidney, J. Nutr. 122 (1992) 2415 – 2423. [23] R.T. Muldoon, K.M. McMartin, Ethanol acutely impairs the renal conservation of 5-methyltetrahydrofolate in the isolated perfused rat kidney, Alcoholism Clin. Exp. Res. 18 (1994) 333 – 339.

P.K. Sikka, K.E. McMartin / Chemico-Biological Interactions 114 (1998) 15–31

31

[24] B.A. Kamen, A. Capdevila, Receptor-mediated folate accumulation is regulated by the cellular folate content, Proc. Natl. Acad. Sci. USA. 83 (1986) 5983 – 5987. [25] G.B. Henderson, E.M. Zevely, Inhibitory effects of probenecid on the individual transport routes which mediate the influx and effflux of methotrexate in L1210 cells, Biochem. Pharmacol. 34 (1985) 1725–1729. [26] M. Kirkova, E. Ivancheva, E. Russanov, In vitro effects of aspirin on malondialdehyde formation and on activity of antioxidant and some metal-containing enzymes, Comp. Biochem. Physiol. Pharmacol. Toxicol. Endocrinol. 108 (1994) 145 – 152. [27] B. Cantabrana, J.R. Vallina, L. Menendez, A. Hidalgo, Spasmolytic and calmodulin inhibitory effect of non-steroidal anti-inflammatory drugs in vitro, Life Sci. 57 (1995) 1333 – 1341. [28] M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248 – 254. [29] J. Selhub, A.C. Gay, I.H. Rosenberg, Effect of anions on folate binding by isolated brush border membranes from rat kidney, Biochim. Biophys. Acta 557 (1979) 372 – 384. [30] G.R. Westerhof, S. Rijnboutt, J.H. Schornagel, H.M. Pinedo, G.J. Peters, G. Jansen, Functional activity of the reduced folate carrier in KB, MA104 and IGROV-1 cells expressing folate-binding protein, Cancer Res. 55 (1995) 3795 – 3802. [31] J.B. Pritchard, D.S. Miller, Mechanisms mediating renal secretion of organic anions and cations, Physiol Rev 73 (1993) 765–796. [32] B.A. Kamen, A.K. Smith, R.G.W. Anderson, The folate receptor works in tandem with a probenecid-sensitive carrier in MA104 cells in vitro, J. Clin. Invest. 87 (1991) 1442 – 1449. [33] D.W. Fort, R.H. Lark, A.K. Smith, M. Marling-Cason, S.D. Weitman, B. Shane, B.A. Kamen, Accumulation of 5-methyltetrahydrofolic acid and folylpolyglutamate synthetase expression by mitogen stimulated human lymphocytes, Br. J. Haematol. 84 (1993) 595 – 601. [34] T.F. Yu, A.B. Gutman, Study of the paradoximal effects of salicylate in low, intermediate and high dosage on the renal mechanisms for excretion of urate in man, J. Clin. Invest. 38 (1959) 1298 – 1315. [35] M.D. Skeith, P.A. Simkin, L.A. Healey. The renal excretion of indomethacin and its inhibition by probenecid., Clin. Pharmacol. Ther. 9 (1968) 89 – 93. [36] J. Selhub, S. Nakamura, F.A. Carone, Renal folate absorption and the kidney folate binding protein. II. Microinfusion studies, Am. J. Physiol. 252 (1987) F757 – 760. [37] G.B. Henderson, B. Grzelakowska-Sztabert, E.M. Zevely, F.M. Huennekens, Binding properties of the 5-methyltetrahydrofolate/methotrexate transport system in L1210 cells, Arch. Biochem. Biophys. 202 (1980) 144–149. [38] P.D. Prasad, V.B. Mahesh, F.H. Leibach, V. Ganapathy, Functional coupling between a bafilomycin A1-sensitive proton pump and a probenecid- sensitive folate transporter in human placental choriocarcinoma cells, Biochem. Biophys. Acta. 1222 (1994) 309 – 314. [39] J.G. Blackburn, D.J. Hazen-Martin, C.J. Detrisac, et al., Electrophysiology and ultrastructure of cultured human proximal tubule cells, Kidney Int. 33 (1988) 508 – 516. [40] S.D. Weitman, R.H. Lark, L.R. Coney, D.W. Fort, V. Frasca, V.R. Zurawski, B.A. Kamen, Distribution of folate receptor GP38 in normal and malignant cell lines and tissues, Cancer Res. 52 (1992) 3396–3401. [41] R.K. Verbeeck, Pharmacokinetic drug interactions with nonsteroidal anti inflammatory drugs, Clin. Pharmacokin. 19 (1990) 44–66. [42] J.C. Frolich, A classification of NSAIDs according to the relative inhibition of cyclooxygenase isoenzymes, Trends. Pharmacol. Sci. 18 (1997) 30 – 34. [43] G.B. Henderson, J.M. Tsuji, Identification of cholate as a shared substrate for the unidirectional efflux systems for methotrexate in L1210 mouse cells, Biochim. Biophys. Acta 1051 (1990) 60 – 70. [44] K.M. Morshed, K.E. McMartin, Reabsorptive and secretory 5-methyltetrahydrofolate transport pathways in cultured human proximal tubule cells, Am. J. Physiol. (Renal) 272 (1997) F380 – 388.

.