Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

By selective reabsorption and secretion, transporters in the kidney play a role in maintaining total body and tissue-specific homeostasis of naturally occurring nucleosides (Griffiths et al., 1991; Gutierrez et al., 1992; Gutierrez and Giacomini, 1993; Ritzel et al., 1997; Wang et al., 1997). These transporters may be particularly important in regulating the local concentrations of nucleosides such as adenosine, which have pronounced effects on renal function (Kuttesch and Nelson, 1982; Thorn and Jarvis, 1996). There have been limited studies investigating the disposition of endogenous nucleosides in the human kidney. Kuttesch and coworkers (Kuttesch and Nelson, 1982; Nelson et al., 1983) observed that adenosine was reabsorbed, whereas the 2'-deoxynucleosides, 2-chloro-2'-deoxyadenosine (2CdA or cladribine) and 2'-deoxyadenosine, were actively secreted in patients lacking adenosine deaminase or treated with the deaminase inhibitor, deoxycoformycin.

With the cloning of several Na⁺-dependent nucleoside transporters (Huang et al., 1994; Che et al., 1995; Ritzel et al., 1997; Wang et al., 1997) it is now possible to understand the molecular function of these carrier proteins. Recently, the cDNA encoding a purine-selective, Na⁺-dependent nucleoside transporter was cloned from human kidney in this laboratory (Wang et al., 1997). Northern blot analysis revealed a broad distribution of multiple hSPNT1 transcripts, with expression in liver, intestine, pancreas, heart, skeletal muscle, and in kidney (Wang et al., 1997). In contrast, mRNA transcripts of the rat homolog, rSPNT, were not detected in rat kidney by Northern blotting methods (Che et al., 1995). These data suggest that there may be species differences in the renal handling of purine nucleosides. Furthermore, it is not known whether there are intrinsic differences in the functional characteristics of the cloned human and rat purine-selective transporters. This information is particularly important for understanding the mechanisms involved in the renal transport of nucleoside analogs, including the antiviral agent 2',3'-dideoxyinosine (ddI) and the antineoplastic agent, 2CdA, with hSPNT1.

Mammalian expression systems have been used previously in this and other laboratories to characterize the function of the cloned purine- and pyrimidine-selective nucleoside transporters from rat (Fang et al., 1996; Schaner et al., 1997). The goal of this study was to develop a transiently transfected mammalian expression system for hSPNT1 and to elucidate its functional characteristics and interactions with nucleosides and nucleoside analogs. In particular, the role of hSPNT1 in the disposition of the endogenous nucleosides, inosine, uridine, and adenosine, and various nucleoside analogs (e.g., 2'- 2CdA, 2'-deoxyadenosine, and ddI) was examined.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

HeLa cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA). Minimum essential media, fetal bovine serum, penicillin-streptomycin, and fungizone were obtained from the University of California, San Francisco Cell Culture Facility. 2', 3'-Dideoxyadenosine (ddA); 2', 3'-dideoxycytidine (ddC); 2',3'-dideoxyinosine (ddI); uridine; thymidine; inosine; adenosine; guanosine; nitrobenzylthioinosine; 2CdA; 2'-deoxyadenosine; and acyclovir were purchased from Sigma Chemical Co. (St. Louis, MO). The DNA isolation kit was obtained from Qiagen (Santa Clarita, CA). The radioisotopes [³H]inosine [specific activity (S.A.), 27.8 Ci/mmol], [³H]uridine (S.A., 43.8 Ci/mmol), [³H]adenosine (S.A., 45.8 Ci/mmol), [³H]2'-deoxyadenosine (S.A., 27.8 Ci/mmol), [³H]2CdA (S.A., 3.8 Ci/mmol), and [³H]ddI (S.A., 38 Ci/mmol) were obtained from Moravek (Brea, CA). Lipofectamine and Opti-minimum essential media were supplied by Gibco/BRL (Gaithersburg, MD). Cell culture plates purchased from Nunc (Cambridge, MA) and 12-well plates purchased from Costar (Milpitas, CA) were used to maintain the cells and for uptake studies, respectively. The Bradford reagent for protein studies was purchased from Bio-Rad Labs. (Hercules, CA). The pcDNA3 vector was purchased from Invitrogen (Carlsbad, CA).

Methods

Transfection. The cDNA encoding hSPNT1 was subcloned into the pcDNA3 vector. Plasmid DNA was prepared using the Qiagen Maxi-prep kit as previously described (Schaner et al., 1997). An initial titering of the lipid demonstrated that significant expression of the transporter was observed with DNA to lipid ratios between 1:4 and 1:6. Maximal expression was observed at a ratio of 1:6 for the cDNA encoding hSPNT1; therefore this ratio was used for the study. HeLa cells were seeded at a density of 2 × 10⁵ cells/well 24 h before transfection. Transfection was carried out as previously described (Schaner et al., 1997). The efficiency of transfection was not measured, but studies under identical transfection conditions with a green fluorescent protein chimera of hSPNT1 suggest that about 30 to 40% of the cells express the chimera hSPNT1 (from confocal microscopy studies, data not shown).

Permeant Studies. Uptake studies were carried out 36 to 72 h after transfection. Uptake was measured in the presence of Na⁺ (128 mM NaCl, 4.73 mM KCl, 1.25 mM CaCl₂, 1.25 mM MgSO₄, and 5 mM HEPES, pH 7.4) or absence of Na⁺ (Na⁺ was replaced by 128 mM choline). The uptake was stopped by aspirating the reaction mixture and washing three times with ice-cold Na⁺-free buffer. Cells were then solubilized with 1 ml 0.5% Triton X-100 and 0.5 ml was sampled for liquid scintillation counting (Beckman Instruments, Palo Alto, CA).

Inhibition Studies. For inhibition studies, (a tracer concentration of labeled [³H]inosine (~70 nM plus 1 µM unlabeled inosine) was used as the permeant, and the amount of unlabeled compound used is indicated in the table or figure legends. In general, this concentration was 0.5 mM but varied in the IC₅₀ studies. These experiments were carried out in duplicate or triplicate for 5 min. Data are presented as the mean ± S.D. Uptake assays were stopped as stated above. Inhibition studies were carried out in the presence and absence of Na⁺. As a control, Na⁺-dependent uptake was analyzed in cells transfected with empty vector.

Data Analysis. In general, data are expressed as mean ± S.D. of uptake values obtained in at least two to three wells. Data are representative of a minimum of two experiments carried out on different days. For Michaelis-Menten studies, rate of uptake was expressed as pmol/mg protein/5 min for inosine and pmol/mg protein/2 min for uridine. Data were fit to the equation V = V_max[S]/(K_m+[S]) where V is the rate of inosine or uridine uptake and [S] represents the concentration of inosine or uridine. The Kaleidagraph fitting program was used to fit the data. Parameter estimates are expressed as mean ± S.E. For IC₅₀ studies, data were fit to the equation V = V_o/[1+(I/IC₅₀)ⁿ] where V is the uptake of inosine in the presence of inhibitor, V_o is inosine uptake in the absence of inhibitor, I is the inhibitor concentration, and n is the Hill coefficient. An unpaired Student's t test from Primer of Biostatistics software (Version 3, by Stanton A. Glantz, McGraw-Hill, 1991) was used for determination of statistical significance, and P < .05 was considered statistically significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

In this study we demonstrated that hSPNT1 can be transiently expressed in HeLa cells using the technique of lipofection. Similar techniques have been used previously in this laboratory to transiently transfect HeLa with the cDNA of the rat homolog, rSPNT (Schaner et al., 1997). The experimental conditions required for optimum expression of hSPNT1 were essentially identical to those described for rSPNT except that higher lipid to DNA ratios were used in this study (6:1 in comparison with 3:1 used previously).

Uptake and Inhibition Studies. The uptake of [³H]inosine (~70 nM plus 1 µM unlabeled inosine) in the presence of Na⁺ was linear up to 30 min (data not shown) in HeLa cells transiently transfected with hSPNT1 cDNA. An initial time point of 5 min was used for further kinetic studies. At 0.5 mM, the purine nucleosides, adenosine, guanosine, and inosine, significantly inhibited Na⁺-dependent [³H]inosine uptake, whereas the pyrimidine nucleosides, thymidine and cytidine, did not (Fig. 1). The pyrimidine nucleoside, uridine, which is a substrate of all known mammalian nucleoside transporters (Crawford et al., 1990; Crawford and Belt, 1991; Plagemann, 1991; Ritzel et al., 1997), also significantly inhibited [³H]inosine uptake (P < .05).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Effect of nucleosides on Na+-dependent uptake of [3H]inosine in HeLa cells transiently transfected with cDNA encoding hSPNT1. Solid columns indicate data obtained in cells transfected with hSPNT1 cDNA and gray columns indicate data obtained in cells transfected with empty vector. A, adenosine; I, inosine; T, thymidine; U, uridine; G, guanosine; C, cytidine. Bars represent mean (± S.D.) of data obtained from two to three wells in a representative of at least two experiments.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   A, effect of ddI on Na+-dependent uptake of [3H]inosine was determined in HeLa cells transfected with cDNA encoding hSPNT1. [3H]Inosine uptake was measured in the presence of Na+ and of increasing concentrations of ddI. Data were fit to equation: V = Vo/[1+(I/IC50)n +3.69] using Na+-dependent uptake data and adjusted for nonspecific uptake with Na+-independent uptake data. Points represent mean (± S.D.) of data obtained from two to three wells in a representative of two experiments. IC50 value for this experiment was 19 ± 8 µM. B, effect of 2CdA on inosine uptake in transfected cells. Uptake of [3H]inosine was measured at 5 min in presence of various concentrations of 2CdA in cells transfected with cDNA encoding hSPNT1. Points represent mean (± S.D.) of data obtained from two to three wells in a representative of two experiments. Data were fit to the equation: V = Vo/[1+(I/IC50)n +3.69] as described in A. IC50 value for hSPNT1 was 371 ± 209 µM.

Permeant Studies. Although inhibition studies do show interaction with the carrier, such studies do not necessarily identify substrates (or permeants) of the transporter. To identify other substrates of hSPNT1, the uptake of several ³H-labeled nucleosides and nucleoside analogs was determined in HeLa cells transiently transfected with the cDNA of hSPNT1. Uptake of [³H]inosine, [³H]uridine, [³H]thymidine, [³H]adenosine, [³H]2'-deoxyadenosine, [³H]2CdA, and [³H]ddI was measured for 5 min in the presence and absence of Na⁺ in HeLa cells transiently transfected with hSPNT1 cDNA. [³H]Inosine and [³H]uridine demonstrated significant (P < .05) uptake by this carrier, whereas [³H]ddI and [³H]thymidine did not (Fig. 3). [³H]Adenosine showed significant Na⁺-dependent uptake in the cells expressing hSPNT1, whereas [³H]2'-deoxyadenosine and [³H]2CdA did not (Fig. 4). Similar results were also obtained when uptake was carried out at 1 min (data not shown). The kinetics of uridine and inosine uptake were further characterized. In this study a time point of 2 min was used for uridine, because uridine uptake was linear up to this time point but was not linear at 5 min. Both compounds demonstrated saturable uptake in the transfected cells (Table 2); however, in comparison to the interaction of inosine (K_m = 13.7 ± 8.09 µM; V_max = 182 ± 40 pmol/mg protein/5 min), the interaction of uridine was characterized by a lower affinity and higher capacity (K_m = 116 ± 26 µM; V_max = 728 ± 98 pmol/mg protein/2 min). The Na⁺-dependent uptake of [³H]adenosine (1 µM) was significantly higher in HeLa cells transfected with the cDNA encoding rSPNT versus cells transfected with the cDNA encoding hSPNT1 (Fig. 5).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Uptake of 3H nucleosides and 3H nucleoside analogs in HeLa cells transiently transfected with cDNA encoding hSPNT1. Uptake was carried out for 5 min in the presence (black columns) and absence (gray columns) of Na+. Compounds that demonstrated significant Na+-dependent uptake (P < .05) are indicated by an asterisk. Each bar represents data obtained from two to three wells from a representative of two or more experiments.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Uptake of [3H]adenosine, [3H]2'-deoxyadenosine (dA), and [3H]2CdA (1 µM) in HeLa cells transiently transfected with cDNA encoding hSPNT1. Uptake was carried out for 5 min in presence (black columns) and absence (gray columns) Na+; *indicates substrate with significant Na+-dependent uptake (P < .05). Each bar represents mean ± S.D. of data determined from two to three wells. Data are representative of two or more experiments.

View this table: [in this window] [in a new window]   TABLE 2 Michaelis-Menten parameters for nucleoside uptake in HeLa cells transfected with either cDNA of rSPNT (Schaner et al., 1997) or hSPNT1 Range of concentrations used in establishing kinetic parameters varied among experiments (range, 0 to 50 µM for inosine to 0 to 250 µM for uridine in interacting with hSPNT1).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Adenosine uptake in HeLa cells transfected with cDNA of rSPNT or hSPNT1. Uptake of [3H]adenosine (1 µM) was measured at 5 min in HeLa cells expressing either rSPNT or hSPNT1 in presence (black columns) or absence (gray columns) of Na+. Data from two to three wells of a representative of two experiments are presented (mean ± S.D.).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The kidney plays a vital role in maintaining total body and local purine homeostasis. Limited data suggest that 2'-deoxyribonucleosides are handled differently from ribonucleosides by the human kidney. Namely, 2'-deoxyadenosine and 2CdA are both actively secreted, whereas adenosine is actively reabsorbed. Previously, we demonstrated that both adenosine and 2'-deoxyadenosine inhibit [³H]inosine uptake in Xenopus laevis oocytes expressing hSPNT1; however, the IC₅₀ of adenosine was lower than that of 2'-deoxyadenosine (23 µM versus 110 µM). In this study, we addressed the question of whether adenosine and 2'-deoxyadenosine are permeants of hSPNT1. Our data demonstrate that adenosine is a permeant of hSPNT1, whereas neither 2'-deoxyadenosine nor 2CdA showed significant uptake by the transporter (Fig. 4). However, both compounds showed a slight, albeit not significant, Na⁺-dependent uptake, suggesting that these compounds may be poor permeants of hSPNT1. Collectively, the data suggest that hSPNT1 may play a role in the reabsorption of adenosine in the human kidney but does not appear to play a role in the secretion of either 2'-deoxyadenosine or 2CdA. Understanding the sorting of hSPNT1 to the apical or basolateral membrane of the proximal tubule will provide further information on the physiological role of the transporter.

In this study, the K_m of uridine for hSPNT1 (116 µM) was similar to the value obtained previously in this laboratory in the X. laevis oocyte expression system (80 µM, Table 3). These values are somewhat higher than the K_m of uridine (45 µM) in interacting with the human pyrimidine-selective transporter, hCNT1(Ritzel et al., 1997), suggesting that although both hSPNT and hCNT1 transport uridine, the pyrimidine-selective transporter has a higher affinity for uridine. Both transporters may play a role in the transport of uridine (as well as adenosine) in the human kidney.

Similar to 2'-deoxyadenosine and 2CdA, the antiviral nucleoside analog, ddI, is actively secreted in the human kidney. Our data demonstrate that ddI is a potent inhibitor (IC₅₀ of 19 µM) of [³H]inosine uptake in HeLa cells expressing hSPNT1; however, ddI is not a permeant. These data suggest that hSPNT1 does not contribute to the secretory clearance of either dideoxy or 2'-deoxynucleosides.

Our results demonstrate possible species differences between the human transporter and that cloned from rat (Table 3). The IC₅₀ of 2CdA (371 µM) in inhibiting [³H]inosine uptake in cells expressing hSPNT1 was markedly different from the IC₅₀ of 2CdA (13.7 µM) in cells expressing rSPNT. Furthermore, consistent with previous studies in X. laevis oocytes, the uptake of [³H]2CdA in hSPNT1 cDNA transfected cells was only slightly enhanced. In contrast, in HeLa cells transfected with the cDNA of rSPNT Na⁺-dependent uptake of [³H]2CdA was enhanced more than 2-fold over that in the absence of Na⁺ (Schaner, 1997). Because 2CdA is an antineoplastic agent with numerous clinical applications (Saven et al., 1995, 1996; Pott and Hiddemann, 1997; Stine et al.,1997; Tobinai et al., 1997), a knowledge of species differences in the interaction of this compound and other 2'-deoxynucleoside analogs with the purine-selective nucleoside transporters may be significant for future evaluation of these compounds in animal models.

Adenosine (1 µM) exhibits a higher rate of uptake in cells expressing rSPNT in comparison with those expressing hSPNT1. This higher uptake rate may be due to differences in the K_m of adenosine in interacting with the rat and human transporters; alternatively, the expression level of rSPNT may be greater than that of hSPNT1. The species differences observed for both adenosine and 2CdA in interacting with hSPNT1 and rSPNT may be due to differences in the molecular structures of the two clones. A recent study from this laboratory (Wang and Giacomini, 1997) examined the molecular basis for substrate selectivity using the purine- and pyrimidine-selective rat clones. Similar studies between hSPNT1 and rSPNT may elucidate important species differences. Interestingly, rSPNT is not expressed (to a significant extent) in rat kidney (Che et al., 1995; our unpublished data). This may suggest an important difference in the renal handling of purine nucleosides between species, which must be considered when evaluating such compounds in animal models.

In summary, the human purine-selective, Na⁺-dependent transporter (hSPNT1) was expressed in HeLa cells. The characteristics of hSPNT1 in interacting with a number of nucleosides and nucleoside analogs were determined. Notable species differences in the function of hSPNT1 when compared with the rat homolog, rSPNT, were observed. These results suggest that caution should be taken when evaluating purine nucleosides and nucleoside analogs in the rat. The underlying molecular mechanisms that contribute to such species differences remain to be elucidated. Transient transfection of the cDNA encoding hSPNT1 in HeLa cells represents a useful model to further characterize the interactions of nucleosides and nucleoside analogs with the transporter.