Characteristics of a lysosomal membrane transport system for tyrosine and other neutral amino acids in rat thyroid cells.

Tyrosine countertransport was used to demonstrate the existence of a carrier system for neutral amino acids in the lysosomal membrane of FRTL-5 thyroid cells. In addition to tyrosine, the carrier system recognized the neutral amino acids leucine, histidine, phenylalanine, and tryptophan. Cystine and lysine, amino acids for which a lysosomal carrier system has been demonstrated, showed no competition with tyrosine for countertransport. The tyrosine system showed stereospecificity and cation independence. It did not require an acidic lysosome or the availability of free thiols. The apparent Km for tyrosine was approximately 100 microM; the energy of activation of the system was approximately 9.7 kcal/mol. This new lysosomal membrane carrier system for neutral amino acids resembles the plasma membrane L system in 3T3 Chinese hamster ovary cells and melanoma B-16 cells.

mately 100 WM; the energy of activation of the system was approximately 9.7 kcal/mol. This new lysosomal membrane carrier system for neutral amino acids resembles the plasma membrane L system in 3T3 Chinese hamster ovary cells and melanoma B-16 cells.
Several systems for the transport of small molecules across the lysosomal membrane have been described. Carrier-mediated transport of the amino acid cystine was demonstrated in 1982; a deficiency of this transport system was shown to be responsible for the autosomal recessive disease nephropathic cystinosis (1-4). Later, a lysosomal transport system which recognized lysine, other cationic amino acids, and the mixed disulfide cysteine-cysteamine was reported (5). Recently, lysosomal egress of the monosaccharide, sialic acid, was shown to proceed in a concentration-dependent manner, and deficiency of this egress characterized cells of the lysosomal storage disorder, Salla disease (6). In addition, lysosomal accumulation of free vitamin B12 in fibroblasts of a single patient has been attributed to defective transport of cobalamin across the lysosomal membrane (7).
We investigated whether tyrosine and other neutral amino acids also cross the lysosomal membrane in a facilitated fashion. Tyrosine was chosen as a prototype for the neutral amino acids because lysosomes could be readily loaded to high concentrations of tyrosine by exposure to tyrosine methyl ester. The amino acid methyl ester is rapidly cleaved by a lysosomal hydrolase to yield the free amino acid which accumulates within the lysosomes (8,9). The very high levels of intralysosomal tyrosine achieved by this technique have permitted the demonstration of tyrosine countertransport. This phenomenon, in which tracer amounts of a radiolabeled sub-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. stance will cross a membrane at an increased rate if there is a substantial concentration of the nonradioactive substance on the opposite side of the membrane, rejects the concept of a simple pore and constitutes strong evidence for a carriermediated process (10).
Lysosomal tyrosine transport was studied in human polymorphonuclear leucocytes and in a rat thyroid-derived cell line, FRTL-5. A carrier-mediated system for lysosomal transport of tyrosine and other neutral amino acids was demonstrated in the rat cells.
Preparation of Tyrosine-loaded Granular Fractions-Intact human polymorphonuclear leukocytes were exposed to 0.08-0.6 mM [3H] tyrosine methyl ester (final specific radioactivity 10 pCi/pmol) in Hanks' balanced salt solution containing 10 mM sodium phosphate, p H 7, to achieve various tyrosine loadings. Cells were ruptured by limited sonication using a model W 140 cell disrupter with a microtip (Head Systems, Ultrasonic, Inc., Plainview, NY), and a lysosomerich granular fraction, loaded with tyrosine, was prepared by differential centrifugation as previously described (2,4).
Intact rat thyroid FRTL-5 cells were exposed to [3H]tyrosine methyl ester for egress experiments, or nonradioactive tyrosine methyl ester for countertransport, in Hanks' balanced salt solution containing 25 mM sodium phosphate. Cells were centrifuged, resuspended in 0.25 M sucrose, 10 mM Hepes,' p H 7.0, and subjected to limited sonication. The homogenate was centrifuged a t 2,000 X g for 5 min and the supernatant centrifuged a t 17,000 X g for 10 min to prepare a crude granular fraction pellet. This pellet contained approximately 30% of the total hexosaminidase activity in the homogenate and was 1.5-2.5-fold enriched with respect to this lysosomal enzyme. Electron microscopy of the crude granular fraction obtained after fixing in 2.5% glutaraldehyde in 0.1 M cacodylate buffer (13), followed by post-fixation in 1% osmium tetroxide, staining with 1% ' The abbreviation used is: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. in Rat Thyroid Cells aqueous uranyl acetate, and embedding with propylene oxide/Spurr resin mixture (14), revealed many single membrane vesicular structures characteristic of lysosomes, as well as numerous mitochondria and abundant cell debris ( Fig. 1).
Tyrosine Egress-Egress experiments were performed on human leucocytes as well as rat thyroid cells. Crude granular fractions, loaded with [3H]tyrosine, were suspended in sucrose/Hepes and incubated at 37 "C. Aliquots were removed at various times and centrifuged at 17,000 X g for 10 min, and radioactivity was measured in supernatant and pellet. High-voltage electrophoresis (2) verified that the radioactivity represented only free tyrosine and not tyrosine methyl ester. Tyrosine present in the supernatant due to rupture of lysosomes during incubation was corrected for by measurement of soluble hexosaminidase (1).
Tyrosine Countertransport-Tyrosine-loaded and -unloaded granular fractions from FRTL-5 cells were suspended in sucrose/Hepes containing 8 p~ [3H]tyrosine and incubated at 37 "C. At various times, aliquots were removed and centrifuged at 17,000 X g for 10 min at 4 "C. The supernatant was assayed for soluble hexosaminidase activity, whose mean value was 5.6% of total hexosaminidase at zero time and 6.6% after 3 min at 37 "C. The pellet was washed vigorously with 1 ml of sucrose/Hepes, once by resuspension with a Pasteur pipette, and once by resuspension in a glass pestle, with subsequent centrifugation (4). Final recovery of hexosaminidase in the pellet averaged 74% of initial activity. The final pellet was resuspended in 1 ml of H20 and extensively sonicated. The radioactivity in duplicate 100-pl aliquots was determined using a Beckman scintillation counter. t3H]Tyrosine uptake was expressed as counts/min/unit of hexosaminidase at the designated time minus the zero-time value. The difference between the uptake into the loaded lysosomes and uptake into the unloaded lysosomes constituted countertransport and was expressed as picomoles/unit of hexosaminidase/min. Within a single experiment, duplicate determinations differed by 7-10%. Radioactivity was converted to picomoles of tyrosine by dividing by the specific activity of [3H]tyrosine in the incubation medium.
To differentiate binding from true uptake into lysosomes, tyrosineloaded granular fractions, obtained at the end of a typical countertransport experiment and containing large amounts of [3H]tyrosine, were placed on 2.4-cm GF/A glass fiber filters (Whatman) and washed extensively with sucrose/Hepes. Another aliquot was ruptured by sonication and treated similarly. The tyrosine-loaded granular fractions lost 70% of their radioactivity after sonication. This amounted to 12 times the radioactivity lost from unloaded granular fractions due to sonication, indicating that the bulk of [3H]tyrosine uptake into loaded granular fractions was released by lysosomal rupture and represented true uptake rather than enhanced binding.
To assess lysosomal loading with nonradioactive tyrosine, a 100-pl aliquot of the crude granular fraction was assayed for tyrosine using an LKB 4150 amino acid analyzer with a 5-buffer lithium citrate system.
Assay of Hezosarninidose Activity-The extent of lysosomal rupture was assessed by dividing supernatant hexosaminidase activity by total hexosaminidase activity in the granular fraction suspension. Hexosaminidase activity was assayed by 3 min incubation in 13 mM citric acid, 20 mM sodium phosphate, pH 4.4, as previously described (1). One unit of hexosaminidase hydrolyzed 1 nmol of substrate/min at 37 "C.
Protein was measured by the method of Lowry et al. (15).

in Rat Thyroid
Cells 17109

RESULTS
Loading of lysosomes with tyrosine was accomplished by exposure of intact cells to tyrosine methyl ester. For whole FRTL-5 cells, tyrosine loading was maximal a t p H 7. 2 (Fig. 2a) and increased with duration of tyrosine methyl ester exposure through 20 min (Fig. 2b). Therefore, a pH of 7.2 and an exposure time of 20 min were chosen for all subsequent loadings. Under these conditions, granular fraction tyrosine loading increased linearly with tyrosine methyl ester concentrations up to 3.0 mM (data not shown).
For human leucocytes and rat FRTL-5 cells, the crude granular fractions obtained from cells exposed to 0.6-1.0 mM tyrosine methyl ester contained 10-20 times the amount of free tyrosine compared with granular fractions from cells not exposed to the methyl ester. This high level of tyrosine loading permitted determination of a rate of tyrosine egress out of granular fractions for each cell type (Fig. 3). Tyrosine exited leucocyte granular fractions very slowly but left the FRTL-5 cell granular fractions at a high rate. In view of this finding, and because we could not demonstrate tyrosine countertransport in the leucocyte granular fractions (data not shown), subsequent experiments were performed using the FRTL-5 cells.
Tyrosine countertransport was investigated in these cells by measuring granular fraction uptake at 8 FM [3H]tyrosine. L-Tyrosine-loaded granular fractions took up 5 times more ['H]tyrosine than did unloaded granular fractions (Fig. 4); the difference comprised tyrosine countertransport. To eliminate the possibility that the difference in [3H]tyrosine content represented nonspecific uptake, it was demonstrated that cystine-loaded granular fractions took up no more [3H]tyrosine than did unloaded granular fractions (Table I).
Since tyrosine countertransport increased linearly with time through at least 3 min (Fig. 4), an initial velocity of [3H] tyrosine uptake, in picomoles of tyrosine/unit of hexosaminidase/min, could be measured after 3 min of incubation. This initial velocity increased with increased loading of nonradioactive tyrosine up to approximately 1.5 nmol of tyrosine/ unit of hexosaminidase and then leveled off (Fig. 5). This  2. a, [3H]tyrosine loading of FRTL-5 cells after exposure to [3H]tyrosine methyl ester at different pH values. FRTL-5 cells were exposed to 1 mM [3H]tyrosine methyl ester at 37 "C in 4 ml of Hanks' balanced salt solution buffered by 25 mM sodium phosphate at different pH values for 20 min. Cells were washed twice in 3 ml of phosphate-buffered saline; the cell pellet was resuspended in 420 pl of H 2 0 and sonicated, and tyrosine/mg of protein was determined for each loading condition. Results are means of duplicate determinations. The identity of the radioactivity in the tyrosine peak was confirmed by high-voltage electrophoresis. Tyrosine radioactivity was converted to picomoles using the calculated specific radioactivity of tyrosine in the incubation mixture. b, time course of [3H]tyrosine loading into FRTL-5 cells exposed to [3H]methyl ester. FRTL-5 cells were exposed to 1 mM [3H]tyrosine methyl ester at 37 "C and treated as described for a. The pH was 7.2.  to 1 mM tyrosine methyl ester were placed in sucrose/Hepes medium containing 8 PM [3H]tyrosine at 37 "C. Aliquots were removed at different times, centrifuged, washed twice, and assayed for [3H]tyrosine and for total hexosaminidase activity. Soluble hexosaminidase activity ranged from 5-7% at zero time to 11-15% after 12 min at 37 "C. demonstration of saturation kinetics confirmed that tyrosine was being transported by a carrier-mediated system. To ensure maximal [3H]tyrosine uptake in subsequent experiments, the granular fractions were routinely loaded to at least 1.5 nmol of tyrosine/unit of hexosaminidase; this level was generally achieved by exposing the whole cells to 1-2 mM tyrosine methyl ester.
The velocity of tyrosine countertransport increased as temperature increased (Fig. 6), with a Qlo of approximately 1.9. The energy of activation was 9.7 kcal/mol. The technique of tyrosine countertransport was also used to determine which compounds compete with [3H]tyrosine for uptake into tyrosine-loaded granular fractions; such com-

Effect of different loading conditions on pH]tyrosine uptake
In different experiments FRTL-5 cells were loaded to 1.2-4.1 nmol of half-cystine/unit of hexosaminidase or 0.21-0.73 nmol of tyrosine/ unit of hexosaminidase and exposed to 8 PM [3H]cystine or [3H] tyrosine for 20 min at 37 "C. The radioactivity remaining in the washed pellet was converted to picomoles of each compound by using the calculated specific radioactivity in the incubation mixture. Uptake of [3H]tyrosine into cystine-loaded lysosomes was no different from uptake into unloaded lysosomes; the relatively low uptake into tyrosine-loaded lysosomes was due to suboptimal tyrosine loading of the granular fractions.

FIG. 5. ['HITyrosine uptake into FRTL-5 granular fractions loaded to different levels with nonradioactive tyrosine.
FRTL-5 granular fractions, loaded to different levels of tyrosine by exposure of the whole cells to 0.1-3 mM tyrosine methyl ester, were assayed for their ability to take up 13H]tyrosine, as described in the legend to Fig. 4. [3H]Tyrosine uptake was determined after 3 min. Results represent differences between loaded and unloaded values; different symbols represent different experiments. Intralysosomal concentrations of tyrosine were measured using an amino acid analyzer. Hex, hexosaminidase. pounds would be expected to share the transport system with tyrosine. Nonradioactive L-tyrosine itself competed, apparently saturating the tyrosine uptake system a t approximately 100 PM (Fig. 7). When expressed as the percentage of the control amount of [3H]tyrosine taken up by tyrosine-loaded granular fractions, 0.5 mM nonradioactive L-tyrosine routinely competed 88-94% (Table 11). In contrast, D-tyrosine at 0.5 mM competed only negligibly. Cystine, lysine, glycine, glutamate, proline, and alanine, which did not compete a t all with [3H]tyrosine for uptake, were apparently not recognized by the carrier.
Other amino acids could be divided into two groups. Group 11, including valine and methionine, showed moderate competition (Table 11). Amino acids in group I11 were apparently well recognized by the tyrosine carrier and competed for [3H] tyrosine uptake to the same extent as tyrosine itself. This group of neutral amino acids included leucine, histidine, tryptophan, phenylalanine, and isoleucine (Table 11).
The alkalinizing agent ammonium chloride was used to determine whether lysosomal tyrosine transport required an  7. ['HITyrosine countertransport in the presence of different concentrations of tyrosine. FRTL-5 granular fractions, loaded to 1.7 nmol of tyrosine/unit of hexosaminidase, were added to incubation medium containing trace amounts of [3H]tyrosine plus different amounts of nonradioactive tyrosine. The initial velocity of [3H]tyrosine uptake was determined at 3 min; radioactivity inside the lysosomes was converted into picomoles of tyrosine using the specific activity of tyrosine in the incubation medium. acidic lysosome. Ammonium chloride (20 mM) had no effect on tyrosine countertransport, nor did the free-thiol trapping agent, N-ethylmaleimide, at 5 mM (Table 111). High concentrations (100 mM) of sodium or potassium chloride also did not alter the initial velocity of [3H]tyrosine uptake into tyrosine-loaded granular fractions ( Table 111). The presence of 2 mM magnesium chloride, magnesium-ATP, or sodium-ATP neither enhanced nor inhibited tyrosine countertransport (Table 111).

DISCUSSION
Carrier-mediated transport of tyrosine and other neutral amino acids across the plasma membrane has been described for 3T3 Chinese hamster ovary cells (16) and melanoma B-16 cells (17). The system has an apparent K,,, for tyrosine of 75 PM and is sodium-independent (16, 17). It was characterized by measuring uptake of radiolabeled amino acids into whole cells.
in Rat Thyroid Cells FRTL-5 granular fractions, loaded to tyrosine levels above 1.5 nmol/unit of hexosaminidase, were placed in sucrose/Hepes containing 8 PM [3H]tyrosine and the added compound. [3H]Tyrosine countertransuort was measured and expressed as a percentage of the control (i.e. with no added compound).

Compound
Concentration Transport  Effects of ammonium chloride, N-ethylmuleimide, various cations, and A T P o n tyrosine countertransport in FRTL-5 granular fractions FRTL-5 granular fractions were loaded to over 2.1 nmol of tyrosine/unit of hexosaminidase. [3H]Tyrosine uptake was measured in the presence of the added compounds and expressed as a percentage of control uptake, which averaged 1.85 f 0.7 pmol of [3H]tyrosine uptakelunit of hexosaminidaselmin.

92
Such uptake experiments are more difficult with subcellular vesicles such as lysosomes, although this technique has recently been used by Pisoni et al. (18) to elucidate a transport system for proline and other neutral amino acids in Percollpurified fibroblast lysosomes. An alternative method for studying lysosomal membrane transport is to load the lysosomes with a substance and measure its egress from the vesicles (1,2,5,6,9) or its stimulation of uptake into the vesicles (4). Exposure of cells to an amino acid methyl ester has been shown specifically to load lysosomes with the free amino acid (8); the amino acid esterase activity co-purifies with lysosomes, and lysosomal accumulation of the free amino acid is inhibited by the lysosomotropic alkalinizing agent, chloroquine (9). The crude granular fractions which we studied after whole cell exposure to tyrosine methyl ester contained electron-dense bodies typical of lysosomes and no other organelles known to have hydrolytic activity against amino acid methyl esters (Fig. 1). Furthermore, high-voltage electrophoretic analysis verified that the tyrosine content of granular fractions from methyl ester-treated cells was increased 10-20-fold compared with granular fractions from untreated cells.
It has been shown that lysosomes could be loaded with each of the neutral amino acids leucine (9), phenylalanine (19), and tyrosine by exposure to the respective amino acid methyl ester. However, Reeves (9) was unable to demonstrate a trans effect of external leucine on leucine exodus in rat liver lysosomes, Pisoni et al. (5) could not detect a trans effect by neutral amino acids on phenylalanine exodus from human fibroblast lysosomes, and we were unable to show a trans effect on egress of tyrosine using human leucocyte lysosomes; furthermore, we could not demonstrate countertransport of tyrosine in the same system (data not shown). These findings suggest that, for reasons perhaps related to the saturability of the lysosomal carrier or to species variability, the rat liver and human cells studied were not appropriate for the demonstration of a neutral amino acid carrier in the lysosomal membrane.
In contrast, lysosome-rich granular fractions from rat FRTL-5 cells provided a suitable system for the study of lysosomal tyrosine transport (Fig. 3). They were appropriately loaded by exposure to tyrosine methyl ester (Fig. 2) and displayed several characteristics of carrier-mediated transport. The first and the most important criterion of facilitated transport was the demonstration of countertransport itself; [3H]tyrosine uptake was greater in tyrosine-loaded compared with -unloaded granular fractions (Fig. 4). This was true uptake, rather than some nonspecific binding enhanced by prior exposure to tyrosine methyl ester, because bound [3H] tyrosine (determined by radioactivity retained on a glass fiber filter after extensive sonication) was the same for loaded and unloaded fractions exposed to [3H]tyrosine. The countertransport was also specific for tyrosine, compared with another amino acid, cystine, transported by a lysosomal membrane system (1)(2)(3)(4). That is, cystine loading did not enhance [3H]tyrosine uptake into granular fractions, and tyrosine loading did not enhance [3H]cystine uptake (Table I). Second, the tyrosine transport system was stereoselective for the L-isomer. D-Tyrosine at 0.5 mM competed with [3H]tyrosine uptake only negligibly compared with L-tyrosine (Table 11). Third, tyrosine countertransport appeared a saturable process with respect to intralysosomal tyrosine loading. Finally, the Qlo for tyrosine transport (1.9) was in a range consistent with carriermediated transport rather than simple diffusion; the Qlo closely resembled that for lysosomal cystine transport (2.0) in human leucocytes (4).
In the rat FRTL-5 cells, the lysosomal carrier for tyrosine also recognized other neutral amino acids, including leucine, phenylalanine, tryptophan, histidine, and isoleucine ( Table  11). The relatively broad specificity for ligands resembled that reported for the lysine (5) and proline (18) carriers in fibroblast lysosomes, but contrasted with the strict ligand requirements of the human leucocyte lysosomal cystine transport system (4). The ligand affinity for the tyrosine and cystine systems also differed. Although the cystine carrier had an estimated K, for cystine of 0.5 mM (4), approximately 0.1 mM tyrosine competed 50% with [3H]tyrosine for uptake (Table  11). However, as for the cystine system, lysosomal tyrosine transport was not influenced by alkalinization with ammo-nium chloride (ZO), by trapping of free thiols with N-ethylmaleimide (1, 4), or by high concentrations of sodium or potassium (4) ( Table 111). Therefore, there was no evidence that the carrier operated through a structural change involving a protonated intermediate or that it required free thiols for binding, translocation, or release of tyrosine; furthermore, the system did not appear cation-dependent. The lysosomal tyrosine carrier also resembled the cystine carrier in being relatively unaffected by magnesium chloride or ATP at pH 7.0 (4) ( Table 111).
The rat cell system for lysosomal tyrosine transport resembles the plasma membrane L system in ligand specificity (Table 11), approximate K , for tyrosine binding (17) (Table  11), and sodium independence (Table 111). These similarities, along with the fact that the lysosomal system does not depend upon intravesicular acidification (Table HI), prompt speculation that the same gene might code for this particular neutral amino acid carrier in the plasma membrane and in the lysosomal membrane, with conservation among different species. The implication that an amino acid transport protein can be targeted to either the plasma membrane or the lysosomal membrane has been suggested by Pisoni et al. ( 5 ) , based upon the resemblance of the lysosomal lysine carrier system and the plasma membrane system y+ (21-23). This targeting hypothesis may be tested using the tyrosine-neutral amino acid systems by employing a photoaffinity probe, such as pazidophenylalanine as suggested by Tabb et al. (24), to purify both the L system carrier from plasma membranes (16, 17) and the tyrosine carrier from rat lysosomal membranes. These investigations would comprise the first steps in comparing the plasma and lysosomal membrane carriers and in elucidating the normal pathway for incorporation of carrier proteins into their respective membranes.
The transport system for tyrosine reported here is the most recent lysosomal membrane carrier system to be described, following those for cystine (1-4), lysine (5), sialic acid (6), and proline (18). However, it is the only lysosomal system in which countertransport is demonstrable in growing cells.
Countertransport has been employed to show the gene-dosage effect of a deficiency of the lysosomal cystine carrier (25), implying that the technique may permit a determination of the abundance of a given carrier within the lysosomal membrane. If so, then manipulating the intracellular metabolism of growing rat thyroid cells, with subsequent assay of the number of lysosomal tyrosine carriers, could reveal important characteristics about the biochemistry of lysosomal membrane carrier proteins.