In lysinuric protein intolerance (LPI), impaired transport of cationic amino acids in kidney and intestine is due to mutations of theSLC7A7 gene. To assess the functional consequences of the LPI defect in nonepithelial cells, we have characterized cationic amino acid (CAA) transport in human fibroblasts obtained from LPI patients and a normal subject. In both cell types the bidirectional fluxes of arginine are due to the additive contributions of two Na+-independent, transstimulated transport systems. One of these mechanisms, inhibited by N-ethylmaleimide (NEM) and sensitive to the membrane potential, is identifiable with system y+. The NEM- and potential-insensitive component, suppressed by l-leucine only in the presence of Na+, is mostly due to the activity of system y+L. The inward and outward activities of the two systems are comparable in control and LPI fibroblasts. Both cell types expressSLC7A1 (CAT1) and SLC7A2 (CAT2B and CAT2A) as well as SLC7A6 (y+LAT2) and SLC7A7 (y+LAT1). We conclude that LPI fibroblasts exhibit normal CAA transport through system y+L, probably referable to the activity ofSLC7A6/y+LAT2.
- cationic amino acid
- membrane transport
- system y+
- membrane potential
- nitric oxide
- lysinuric protein intolerance
lysinuric protein intolerance (LPI, also known as dibasicoaminoaciduria type II; MIM 222700) is an autosomal recessive disorder in which increased urinary excretion of cationic amino acids (CAA) and protein intolerance are caused by impaired CAA transport across the basolateral membrane of kidney and intestine cells (see Ref. 20 for review). Early works on nonepithelial cells (21, 22) suggested that human fibroblasts (HF) but not erythrocytes from LPI patients express the genetic defect as a reduced transstimulation of CAA efflux.
When those studies were performed, no hints were available about the molecular bases of either LPI defect or, more in general, CAA transport itself; moreover, a single mechanism, named system y+(27, 28), was thought to account for CAA transport in mammalian cells. It is now well known that, in fact, system y+ corresponds to the activity of several carriers of the cationic amino acid transporter (CAT) family (16, 17), none of which, however, has been found mutated in LPI patients (15). Moreover, detailed studies, performed in the last few years (10) indicated that mammalian cells can transport CAA through at least four distinct mechanisms, namely, systems y+, y+L, b0,+, and B0,+ (see Table 1).
Among these, system y+L, first described by Deves et al. (11), is an exchange route that recognizes CAA in the absence of Na+ but requires the cation to interact with neutral amino acids such as leucine. The same authors also found that the activities of systems y+L and y+ can be easily discriminated by taking advantage of the sensitivity of the latter mechanism to N-ethylmaleimide (NEM) inhibition (9). System y+L is a glycoprotein-associated amino acid heterodimeric transporter (18, 24, 26) whose “heavy chain” is the glycoprotein 4F2hc, and the hydrophobic “light chain,” y+LAT1 or y+LAT2, is responsible for the recognition and operational features of the transport process (26) .
Functional and genetic studies have suggested that system y+L is a candidate for the transport mechanism mutated in LPI (24). This hypothesis has been validated independently by two groups (1, 25) that identified SLC7A7, encoding for the light chain y+LAT1, as the gene mutated in LPI patients. This gene is expressed not only in kidney and intestine but also, although to a lesser degree, in nonepithelial tissues (1,25).
In light of these recent developments, we have readdressed here the issue of the expression of LPI defect in nonepithelial tissues. To this aim, we have performed a characterization of bidirectional CAA fluxes in HF and compared the discriminated activities of systems y+ and y+L in control cells and in cells obtained from LPI patients.
MATERIALS AND METHODS
Cell strains, cell culture, and experimental procedures.
Normal HF were obtained from a 15-yr-old healthy donor. LPI cells were obtained from skin biopsies of two patients affected by LPI.SLC7A7 cDNA of LPI1 cells presents a deletion of 543 bp at position 197, while LPI2 cells present an insertion of 4 bp (ATCA) at position 1,625 of the SLC7A7cDNA (1, 23). Cells were routinely grown in 10-cm-diameter dishes in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 4 mM l-glutamine, and 5.5 mM glucose. The conditions of culture were as follows: pH 7.4, atmosphere 5% CO2 in air, and temperature 37°C. The experiments were made on fibroblast subcultures resulting from 3 × 104 cells seeded onto 2-cm2 wells of disposable 24-well trays (Nunc) and were incubated for 5–7 days in 1 ml of growth medium. The culture medium was always renewed 24 h before the experiment. Cells were used at a density of 25 ± 10 μg protein/cm2.
For amino acid-free incubation, cell monolayers were washed twice in Earle's balanced salt solution (EBSS) and incubated in the same solution supplemented with 10% dialyzed FBS.
Amino acid influx.
All the experiments were performed by using the cluster tray method for the measurement of solute fluxes in adherent cells (13) with appropriate modifications. Cell monolayers, washed twice with modified bicarbonate-free EBSS buffered at pH 7.4 with 20 mM Tris · HCl, were incubated for 30 s in the same solution containing [14C]arginine. In this interval of time, arginine uptake approached linearity (2). The experiment was terminated by three rapid washes (<10 s) in 0.1 M MgCl2, and cell monolayers were extracted in 0.2 ml ethanol. The radioactivity of cell extracts was determined with Microbeta Trilux (Wallac). Extracted cell monolayers were then dissolved with 0.5% sodium deoxycholate in 1 M NaOH, and protein content was determined directly in the well by using a modified Lowry procedure as previously described (13) .
NEM inhibits selectively system y+ and, thus, can be employed to discriminate the components of CAA uptake (9). In cultured HF, NEM treatment incompletely suppresses arginine influx (Fig. 1). To eliminate the contribution of system y+, NEM was added to the incubation medium 5 min before the uptake assay at a final concentration of 0.4 mM from a 100× stock solution in water. Preliminary experiments had shown that, under the conditions adopted, arginine transport was not further inhibited by the prolongation of NEM treatment.
In the experiments in which Na+-independent transport was to be measured, NaCl and NaH2PO4were replaced, respectively, byN-methyl-d-glucamine (NMG) and choline salts to employ a modified Na+-free EBSS (NMG-EBSS). When the extracellular K+ concentration ([K+]o) was to be changed, KCl replaced NaCl in EBSS and 25 mM HEPES-NaOH or -KOH was employed as a buffer.
The osmolality of the transport solutions was routinely checked with a vapor pressure osmometer (Wescor 5500) and was found to be 280 ± 10 mosmol/kg.
Expression of influx data.
The intracellular fluid volume was estimated by urea distribution space, as described previously (6, 7). Amino acid influx is expressed as micromoles or nanomoles per milliliter of intracellular water per minute.
Kinetic parameters of arginine influx were determined by nonlinear regression analysis by using Eq. 1 for a saturable system plus diffusion Equation 1where V is the initial velocity of amino acid influx,V max is the maximal velocity of transport, [S] is the amino acid substrate concentration, K m is the Michaelis-Menten constant, and K d is the diffusion constant, and by using Eq. 2 for a competitively inhibited system Equation 2where V 0 is the influx velocity in the absence of the inhibitor, [I] is the inhibitor concentration, Imax is the maximal inhibition, and [I]0.5 is the inhibitor concentration required for half-maximal inhibition.
Amino acid efflux.
Determination of arginine efflux was performed by sequentially removing aliquots of the extracellular medium and replacing them with fresh solution as described by Rotoli et al. (19) for the efflux of neutral amino acids. Briefly, after a 6-h depletion in EBSS supplemented with 10% dialyzed FBS, cells were loaded for 15 min with 0.5 μCi/ml of carrier-free [14C]arginine in EBSS (arginine concentration [Arg] = 1.5 μM). After this period, cells were washed twice with fresh EBSS in the presence or in the absence of Na+ and incubated in 0.4 ml of the same solution. [14C]arginine that had escaped from cells was measured after 30 s and 1 min, whereas the amino acid remaining in the cells was determined after extraction of cell monolayers in 0.2 ml of ethanol. Efflux data are expressed as percentages of the total arginine accumulated before the efflux.
Total RNA from subconfluent cultures of control and LPI fibroblasts, seeded onto 10-cm2 wells, was isolated with Trizol (Life Technologies Italia). RNA (5 μg) that was pretreated with RNase-free DNase, heated at 70°C for 10 min, and placed on ice for 1 min was then incubated with a mixture containing 0.5 mM dNTP mix, 25 ng/μl oligo(dT)15–18 (Life Technologies), 10 mM dithiothreitol, 1× first strand buffer, 10 units of RNase inhibitor (Pharmacia), 200 units of SuperScript RT (Life Technologies), and water to a final volume of 20 μl for 1 h at 42°C. The reaction was stopped by heating at 70°C for 15 min. The duplex of RNA-DNA was treated with 5 units of RNase H (USB) at 37°C for 20 min, and the amount of single-strand cDNA was evaluated by fluorometry (Victor, Wallac) with the fluorescent probe Oligreen (Molecular Probes) by using phage M13+ as single-strand DNA standard.
Single-strand cDNA (200 ng) from each sample was amplified in a total volume of 50 μl with 1.25 units of Taq DNA polymerase (Qiagen), 1× PCR buffer, 0.2 mM each dNTP, and 1.5 mM MgCl2 along with transporter-specific sense and antisense primers (see Table 2) at a concentration of 0.4 mM. The primers were designed according to the known sequences reported in GenBank with the help of the Primer3 program (19a).
The “hot start” technique was used to increase amplification efficiency. After an initial denaturation at 94°C for 3 min, samples were heated to 80°C to reduce formation of nonspecific amplification products. Taq polymerase (1.25 units) was then added, and the reaction went on with the annealing step at 59°C for 45 s, followed by the extension step at 72°C for 2 min. A further 34 cycles were carried out, with the only difference being the denaturation step of 30 s. A final extension of 5 min at 72°C was performed. Amplified mixtures (10 μl each) were electrophoresed through a 2% agarose gel (NuSieve 3:1-FMC), stained with ethidium bromide (1 μg/ml) in 1× TBE buffer (0.1 M Tris, 90 mM boric acid, and 1 mM EDTA, pH 8.4), and visualized by ultraviolet light. Product size was established by comigration with a 100-bp ladder marker (Life Technologies Italia). Pictures of the electrophoresed cDNAs were recorded with a digital DC 120 Kodak camera.
FBS and culture medium (DMEM) were purchased from BioWhittaker. [14C]arginine (312 mCi/mol) was obtained from du Pont de Nemours (Bad-Homburg, Germany), and ethanol was from Carlo Erba (Milan, Italy). Sigma (St. Louis, MO) was the source of all other chemicals.
Characterization of arginine transport in cultured HF.
Little difference was observed when the kinetic analysis of arginine influx was performed in the absence or in the presence of Na+ (Fig. 2). The parameters obtained (Table 3) indicate that, under both conditions, arginine influx can be described as the additive result of a single saturable system, endowed with a moderately high affinity for the substrate (K m ≈ 60 μM), and a nonsaturable component, formally indistinguishable from diffusion in the concentration range adopted. However, the apparent homogeneity of the transport process, although consistent with the linearity exhibited in the Hofstee plot (Fig. 2 B), contrasts with the incomplete inhibition by NEM, which reducedV max of arginine influx by <70%. NEM-resistant arginine influx is mediated by a single saturable agency whose affinity for arginine (K m = 45 μM) is slightly higher than that exhibited by total influx. These results indicate that two main components, both Na+ independent, account for CAA transport in cultured HF.
Figure 3 shows the effects of leucine on arginine influx in either the presence or absence of Na+. In NEM-treated cells (Fig. 3 A), leucine inhibited arginine influx. However, the maximal inhibition by the neutral amino acid was much larger in the presence of Na+, when >80% of the NEM-resistant arginine influx was leucine inhibitable, than in the absence of the cation (Imax < 20%). Leucine inhibition was restricted to the NEM-insensitive component of arginine influx, because the NEM-sensitive fraction was not significantly affected by l-leucine up to a concentration of 1 mM (Fig.3 B). These results indicate that most of the NEM-insensitive, leucine-inhibitable portion of arginine influx is due to the activity of system y+L, while the NEM-sensitive, leucine-resistant fraction can be identified as system y+(10).
Operational features of systems y+and y+L in cultured HF.
The kinetic analysis was repeated under discriminating conditions in both fed and depleted cells to study the effect of changes in the intracellular amino acid pool on the arginine influx pathways. The results indicate that both systems y+ and y+L are significantly transstimulated (Fig. 4). The effect was much larger for system y+, whose Vmax is more than fourfold enhanced in fibroblasts with an intact amino acid pool compared with that in cells depleted for 6 h in an amino acid-free saline solution. TheV max of system y+L was only doubled in fed cells compared with their depleted counterparts (Table4).
While the dependence on membrane potential has been conclusively described for system y+ (14), no clear-cut data are available about effects of voltage changes on the activity of system y+L in nucleated cells, although in human erythrocytes (8) and in plasma membrane vesicles from placenta (12) this system has been described as poorly dependent on membrane potential. In the experiment shown in Fig.5, membrane potential of cultured human fibroblasts was changed by increasing [K+]ofrom 5 to 150 mM in the presence of the K+ ionophore valinomycin (100 μM). Under these conditions, the membrane potential of HF is dominated by the K+ transmembrane gradient and ranges from −4 to −72 mV (3). To reduce the possible influence of transstimulation by the intracellular amino acid pool on transport activities, we extensively depleted cells with a 6-h incubation in EBSS + 10% FBS. As expected, the activity of system y+ exhibited an evident dependence on [K+]o. The relationship between system y+ activity and log[K+]o is fairly described by a linear regression (slope −11.13 ± 0.94;P < 0.01; R 2 = 0.979). On the contrary, system y+L activity was not significantly affected by changes in [K+]o (slope −1.199 ± 1.5; not significantly different from 0,P = 0.59). This result indicates that, in cultured HF, system y+L activity is substantially voltage insensitive.
Expression of genes related to CAA transport in cultured HF.
Figure 6 shows the results of RT-PCR analysis of CAA transporters expressed in normal HF. System y+activity corresponded to the expression of SLC7A1 (coding for hCAT1 transporter), SLC7A2 (coding for hCAT2B and 2A transporters), and, possibly, SLC7A4 genes (Fig.6 A). The expression of two genes for y+L-related light chains, SLC7A6 (coding for y+LAT2) andSLC7A7 (coding for y+LAT1), was also detected in fibroblasts, along with a clear-cut expression of the geneMDU1 for the corresponding heavy chain 4F2hc/CD98.
Comparison of discriminated systems for arginine influx in control and LPI cells.
Figure 7 shows the results of an experiment in which the discriminated arginine influx was measured in control cells and in fibroblasts obtained from two LPI patients carrying different mutations of SLC7A7 (see Cell strains, cell culture, and experimental procedures). In both the presence and absence of Na+, the values of total and discriminated arginine influx determined in control fibroblasts were intermediate between those obtained in the two LPI strains. No clear-cut quantitative or qualitative alteration was therefore detectable in LPI cells compared with control fibroblasts. These data, obtained in fed cells, were confirmed in amino acid-depleted fibroblasts (data not shown); under the latter condition, total arginine influx, as well as its system-specific components, was lower than corresponding values obtained in fed cells, as expected for transstimulated transport mechanisms.
Comparison of discriminated arginine efflux in control and LPI cells.
The transport defect in LPI is thought to cause an impairment of CAA efflux through the basolateral membrane of kidney and intestine epithelium (26), suggesting that the malfunction of system y+L in LPI cells is restricted to CAA efflux. To assess this hypothesis, the contribution of systems y+ and y+L to arginine efflux was discriminated and compared in control (Fig. 8, A andB) and LPI1 fibroblasts (Fig. 8, Cand D). After an extensive depletion of the intracellular amino acid pool, cells were incubated with tracer amounts of arginine for 15 min. In a subset of cells, NEM was added during the last 2 min of uptake so as to block system y+ operation selectively. NEM-treated cells exhibited a slower efflux with respect to untreated cells in both the presence (Fig. 8, A and C) and absence of Na+ (Fig. 8, B and D), as expected from the inhibition of system y+. However, when arginine efflux was performed in the presence of extracellular leucine (100 μM), the efflux rate was significantly increased. Leucine-induced transstimulation was observed only in the presence of Na+, and, therefore, it is clearly attributable to the operation of system y+L. The results obtained in the other LPI cell strain, LPI2 fibroblasts, were similar (data not shown).
The behavior of the three cell strains was thus comparable not only with respect to the overall arginine efflux (NEM absent, leucine absent) but also to the basal (NEM treated, leucine absent) or even the transstimulated (NEM treated, leucine present) outward activity of system y+L.
Expression of y+L-related genes in control and LPI fibroblasts.
The comparable phenotype of system y+L activity exhibited by control and LPI fibroblasts prompted us to study the expression of genes coding for y+L-related light chains in the three cell strains employed in this study (Fig. 9). The expression of SLC7A6, coding for y+LAT2 (24), was clearly detectable and similar in the three cell strains. On the contrary, the expression of SLC7A7, which codes for y+LAT1 light chain (1, 18, 24, 25), was evident only in control and LPI2 fibroblasts, although it was also detectable in LPI1 fibroblasts under more sensitive amplification conditions (1).
Arginine transport was previously characterized in cultured HF and was found to be dependent on the activity of a single transport system, identified as system y+ (28). At variance with those data, the present contribution demonstrates that HF transport arginine through two main Na+-independent routes. One of these routes exhibits the typical features of system y+ and is referable to the expression of several members of CAT family (cf. Fig. 6). The other mechanism involved in arginine influx can be identified with system y+L because of its inhibition by leucine in the presence, but not in the absence, of Na+ and its insensitivity to NEM treatment. The results of the kinetic analysis presented in Table 4 indicate that the affinity of system y+L for arginine is of the same order of magnitude as that exhibited by system y+ and, therefore, is markedly lower in cultured HF than in red blood cells (10). The roughly comparable affinities exhibited by systems y+ and y+L explain why the heterogeneity of CAA transport had been overlooked in previous works with HF.
The small inhibition of arginine influx by leucine, detectable in the absence of Na+ (cf. Fig. 3), could point to the presence of the so-called system b0,+ (10). However, because this fraction accounts for at best 7–8% of the total arginine influx at either 20 μM (cf. Fig. 3) or physiological plasma concentrations (not shown), the contribution of transport systems other than y+ or y+L to arginine influx appears marginal in HF.
Previous studies on CAA transport in HF indicated that system y+ operates as an exchange strictly dependent on the membrane potential. Because of this characteristic, the activity of the system, measured as either the initial influx of l-arginine or the steady-state accumulation of the amino acid, has been employed as a voltage indicator (2, 3). Conversely, the evidence presented here demonstrates that, in the same cells, as in other models (8, 12), system y+L exhibits poor, if any, sensitivity to voltage changes. However, the discriminated contribution of system y+ to the total influx of arginine is substantial and can well explain why the intracellular levels of the CAA exhibit a clear-cut voltage dependence.
The absence of voltage dependence of system y+L activity indicates the substantial electroneutrality of the rate-limiting step of the transport process mediated by the system and points to an as yet uncharacterized coupling of CAA influx with cation efflux or anion influx. Because chloride substitution with impermeant organic anions does not affect the activity of the system (8), the electroneutrality of the transport process mediated by the system should be referred to the efflux of a cation. On the other hand, K+ countertransport seems unlikely because system y+L activity is not affected by the increase of the extracellular concentration of this cation (8, 12; present study). Therefore, we favor the hypothesis that the system works as a homoexchange or, alternatively, a heteroexchange, coupling CAA influx with a combined efflux of intracellular Na+ and neutral amino acids. In fact, Chillaron and coworkers (4) failed to detect any efflux of neutral amino acids promoted by arginine influx in y+L-expressing oocytes. These authors, however, demonstrated in the same model that the expression of system y+L causes an increased transmembrane ratio of its substrates and, as a consequence, defined the system as a tertiary active amino acid transport system (4), although it is not clear what kind of energy source the system employs to build up the transmembrane substrate gradient. The definition of the bioenergetic features of system y+L deserves, therefore, further investigations.
Arginine efflux has been characterized with an approach similar to that employed for influx discrimination. The results presented in Fig. 8indicate that CAA efflux is very rapid with almost 80% of the intracellular tracer exiting cells in the first minute of the amino acid-free incubation. NEM treatment markedly lowers arginine efflux, while the NEM-resistant component is significantly transstimulated by leucine in the presence, but not in the absence, of Na+. These data indicate that CAA efflux is shared by systems y+L and y+ in HF and point to the bidirectional activity of both systems in mesenchymal cells.
This conclusion is of great potential interest for the elucidation of the pathogenesis of extrarenal alterations present in LPI. Indeed, while the main defect in this rare autosomal recessive disease is a decreased absorption of CAA in kidney, the phenotype of LPI patients also involves alterations in different organs, such as hemophagocytic lymphohistiocytosis or alveolar proteinosis (20), that are hardly accounted for by the kidney defect. In 1987 Smith et al. (22) actually found that transstimulation of homoarginine efflux was deficient in LPI fibroblasts and concluded that LPI defect is expressed in mesenchymal tissues.
The recent identification of SLC7A7 as the gene mutated in LPI (1, 25) now allows a direct reassessment of the conclusions reached by Smith et al. (22).SLC7A7 codes for y+LAT1, one of the proteins that associates with 4F2hc/CD98 to yield system y+L activity (18,24). However, at least one other protein, y+LAT2, can bind to 4F2hc to express y+L transport activity (18, 24,26). Moreover, the different affinities exhibited by system y+L in oocytes and erythrocytes suggest that different members of the SLC7A family account for system y+L in the various models (18). In this report we have demonstrated that HF are endowed with system y+L activity and express at least two members of the SLC7Afamily related to this transport system. However, the activity of the system appears comparable in cells obtained from LPI patients or from a normal control subject. It should be stressed that LPI cells employed here were derived from two patients carrying different mutations. Under the same conditions, the expression of SLC7A7 is evident in LPI2 and control cells but not in LPI1fibroblasts, although these cells do indeed express the gene (1). This result could point to a lower content ofSLC7A7 mRNA in LPI1 cells, which carry a heavily deleted gene. On the other hand, in LPI2 cells the gene has a missense mutation, and its expression is comparable to that detected in control cells. However, qualitative and quantitative analysis of bidirectional arginine transport yields comparable results in the two LPI strains. Because all the cell strains clearly expressSLC7A6, encoding for y+LAT2, it is likely that the activity of system y+L in HF is accounted for by this member ofSLC7A family. Alternatively, it is possible that both y+LAT1 and y+LAT2 are functional in normal HF. In this case y+LAT2 activity would somehow compensate for the absence of functional y+LAT1 in LPI cells. The elucidation of possible differential expression of y+LAT2 protein in LPI and control fibroblasts requires the availability of specific antibodies for the transporter.
In conclusion, LPI-associated mutations of SLC7A7/y+LAT1 do not cause any significant functional alteration in CAA transport in HF. Consequently, other cell models should be employed to define the metabolic role of y+LAT1 in nonepithelial cells.
This study was partially supported by Ministero dell' Università e della Ricerca Scientifica e Tecnologica Integrated Project “Metabolic and Molecular Bases of Inherited Diseases” (V. Dall'Asta and G. Sebastio), Consiglio Nazionale delle Ricerche Target Project “Biotechnology” (Rome, Italy) (V. Dall'Asta), and Telethon-Italy Grants E.652 (to G. Sebastio) and 29cp (to M. P. Sperandeo). M. P. Sperandeo is an Assistant Telethon Scientist.
Address for reprint requests and other correspondence: V. Dall'Asta, Dipartimento di Medicina Sperimentale, Sezione di Patologia Generale e Clinica, Plesso Biotecnologico Integrato, Universitàdegli Studi di Parma, Via Volturno 39, 43100 Parma, Italy (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society