Metallothionein, glutathione, and cystine transport in pulmonary artery endothelial cells and NIH/3T3 cells

Irawan Susanto, Shawn E. Wright, Richard S. Lawson, Charnae E. Williams, Susan M. Deneke


Both glutathione (γ-glutamylcysteinylglycine; GSH) and the metalloprotein metallothionein (MT) are composed of approximately one-third cysteine. Both have antioxidant activity and are induced by oxidant stresses and heavy metals. Intracellular cysteine levels may depend on uptake and reduction of extracellular cystine. GSH synthesis can be limited by the activity of the xc cystine transport system, which is induced by oxidants and other stresses. MT is induced by treatments that also increase GSH levels and may compete with GSH for intracellular cysteine. We investigated the induction of MT and GSH and cystine transport in NIH/3T3 cells and bovine pulmonary artery endothelial cells exposed to cadmium (Cd) or arsenite. Cd and arsenite increased MT and GSH in both cells. Increases in MT and GSH were accompanied by increases in cystine uptake. Inhibition of cystine transport by glutamate decreased GSH levels and blocked Cd-induced GSH increases in both cell types. MT levels were not significantly affected, suggesting that MT synthesis is less sensitive to intracellular cysteine levels than GSH synthesis.

  • oxidant stress
  • amino acid transport
  • antioxidants

oxidant stress causes cellular injury that may be decreased if cellular antioxidant defenses are induced. Glutathione (GSH) is an important component of antioxidant defenses in the lung and other tissues. One of the factors regulating cellular GSH is the availability of the cysteine substrate needed for synthesis (6, 12). In the extracellular milieu, cystine (oxidized cysteine) is more abundant and, therefore, is more readily available to the cells than cysteine (3). In many cells, intracellular cysteine levels are regulated by the rate of cystine uptake into the cells (3, 7). After transport into the cell, cystine is reduced to cysteine (3).

Metallothionein (MT) is a metalloprotein that contains approximately one-third cysteine residues and has been demonstrated in various systems to have antioxidant activity (23, 24, 28). It has been postulated that a primary role of MT is to limit the intracellular concentration of heavy metals and thereby protect the cells from the toxic effects of cadmium (Cd), copper, and mercury (28). This metal-scavenging role of MT has been postulated to mediate the induction of tumor resistance to anticancer drugs (17). MT has also been found to be induced in a number of cell types and tissues in response to many of the treatments that also increase cellular GSH levels (23, 28). MT may thus potentially compete with GSH for cysteine under conditions in which the availability of this amino acid is limiting for GSH synthesis.

Cellular GSH levels have been correlated with changes in cystine transport in isolated endothelial cells, smooth muscle cells, macrophages, fibroblasts, and brain cells (3, 4, 6, 7, 10-15,19-22). Cystine transport in these cells occurs primarily through an inducible transport system designated as xc by Bannai (3). Intracellular glutamate (Glu) acts as a countertransport agent, facilitating uptake of extracellular cystine (5). This system, which is sodium independent and inhibited by extracellular Glu or homocysteate, is inducible by a variety of stresses including exposure to metals, sulfhydryl reagents, and oxidants. Deneke and co-workers (10-14) have previously reported that cystine transport in bovine pulmonary artery endothelial cells (BPAEC) is primarily through the xc system (15, 26).

In this study, we investigated the induction of MT as it relates to GSH levels and cystine transport activities after exposure to either Cd or the sulfhydryl reagent sodium arsenite in NIH/3T3 cells and BPAEC. We hypothesized that induction of MT and GSH synthesis requires increased cellular uptake of cystine and that blocking of cystine uptake could interfere with the ability of the cell to respond to various stresses by increasing either MT or GSH levels.


Chemical reagents.

Dulbecco’s phosphate-buffered saline (PBS), RPMI-1640, Dulbecco’s modified Eagle’s medium (DMEM), Fungizone, penicillin-streptomycin, and trypsin-EDTA were purchased from Life Technologies (Gibco BRL, Grand Island, NY). Diethyl maleate (DEM), disulfiram (DSF), sodium meta-arsenite (Ars),l-cysteine,l-cystine,l-glutamic acid, fetal calf serum (FCS), and substrates for the GSH assay were purchased from Sigma (St. Louis, MO). Radiolabeled [14C]cystine was obtained from Amersham (Arlington Heights, IL). Lithium chloride was obtained from Mallinckrodt (Paris, KY).

Cell cultures.

BPAEC were isolated by collagenase treatment of fresh pulmonary arteries and characterized as previously described (26). Cells were seeded at a density of 4–8 × 104/35-mm2dish and grown to confluence (>1 × 106 cells/dish) in RPMI-1640 supplemented with 10% FCS, penicillin-streptomycin, and Fungizone as previously described (26). Cells were grown in 21% O2-5% CO2-balance N2 at 37°C. Media were changed every 24–48 h before experimental exposures. NIH/3T3 cells (American Type Culture Collection, CRL 1658) were maintained in DMEM supplemented with 10% FCS, penicillin-streptomycin, and Fungizone.


Exposures to Cd, Ars, DSF, or DEM were begun after cells reached confluence. Treatments lasted 16–24 h, at which time the inducing agents were rinsed off and MT, GSH, or cystine uptake experiments were begun. Glu, used to inhibit cystine uptake, was added at the same time as Cd.

Cell counts.

Cell counts were performed on the same plates as those used for GSH measurements or on replicate plates for MT or cystine uptake assays using a calibrated Coulter counter as described previously (11, 14). Cell sizes were determined with a Coulter channelizer.

GSH assay.

For GSH measurements, cells were harvested with trypsin-EDTA. An aliquot of the cells was treated with 10% perchloric acid, sonicated, centrifuged, and immediately frozen for later GSH assay by the method of Tietze (29) as described by Akerboom and Sies (1).

Amino acid uptake.

Amino acid uptake studies were performed as previously described (10). [14C]cystine was used for all uptake experiments. BPAEC or 3T3 cell monolayers were washed four times with warm PBS containing 14 mM glucose and incubated for 60 min in PBS-glucose. After two more rinses, labeledl-cystine was added to each dish at a concentration of 1 mCi/ml. The total cystine concentration was 0.06 mM in PBS-glucose. Cells were incubated for 10 min, followed by four rinses. The supernatant was aspirated, and the cells were dissolved in 1 ml of 1% Triton X-100. A 0.5-ml aliquot was counted in Ecolite (ICN; Costa Mesa, CA), using a β-counter.

Cellular MT determination.

Intracellular MT was measured in cell lysates using a modification of109Cd-labeled hemoglobin binding assay (16). Cells were scraped into 400 ml of CdCl2 in tris(hydroxymethyl)aminomethane (Tris) buffer and then subjected to repeated freeze-thaw cycles and sonicated. After 10 min of centrifugation, 25 ml of the supernatant were diluted to 200 ml using a Tris-Cd solution, and an equal volume of109CdCl2was added. The sample was vortexed briefly and allowed to sit for 10 min before 100 ml of hemoglobin were added. The mixture was heated for 4 min and centrifuged, and the supernatant was removed. Hemoglobin binding was repeated once, and then the supernatant was counted using a γ-counter.

Calculations and statistics.

MT levels, GSH levels, and rates of cystine uptake were expressed per 106 cells for both cell types. Significant differences between the various groups were determined for paired samples by Student’s t-test or for multiple samples by analysis of variance with the post hoc Scheffé’s test for groups with significant differences.


Effects of Cd on cells.

BPAEC and 3T3 cells were exposed to 5 mM CdCl2 for 24 h. MT and GSH levels and rates of cystine uptake were assayed after the exposure and were compared with untreated controls. The results are summarized in Table1. In BPAEC, Cd exposure resulted in MT levels that were 302 ± 60% of control levels and GSH levels that were 239 ± 27% of control levels. Exposure of 3T3 cells to Cd resulted in a similar percent increase in MT (353 ± 13% of control) but lower increases in GSH (137 ± 4.8% of control). Cystine uptake was modestly induced in BPAEC to 145 ± 16% of control. Cystine uptake in 3T3 cells was also slightly induced. Basal cystine transport activity in 3T3 cells was similar to cystine uptake in control BPAEC (Table 2); however, both basal MT and GSH levels were higher in the 3T3 cells. The cystine transport systems in both cells were found to be predominantly sodium independent (Table 2), characteristic of the xc system.

View this table:
Table 1.

Effects of 5 μM Cd on MT, cystine uptake, and GSH

View this table:
Table 2.

Control levels of MT, cystine uptake, and GSH in NIH/3T3 cells and BPAEC

Effects of Ars, DEM, and DSF on cells.

Exposure of the cells to low levels of thiol reactive agents including Ars, DSF, or DEM for 6–24 h has previously been reported to induce cystine transport activity and increase GSH levels in cells utilizing the xc cystine transport system (3, 6, 7). DEM at higher levels has been used as an effective agent to deplete GSH in these cells, but, as previously shown, induction of xc -mediated cystine transport activity occurs at concentrations low enough that no measurable GSH depletion occurs (3, 11). Both Ars and DEM have been shown to be very effective at inducing MT synthesis in vivo (2, 8). We wanted to determine the relative effects of thiol reactive agents and Cd on MT levels and GSH levels in our cells. In the experiments reported here, exposure to 2.5 μM Ars increased MT levels in both 3T3 cells and BPAEC; however, Ars was not as effective as Cd at inducing MT in either cell type (Table 3). Deneke (10) has previously reported that 24 h of exposure of BPAEC to low levels of Ars resulted in increases in both GSH levels and cystine transport in these cells. The data in Table 3 confirm these previous observations. These increases were also seen in 3T3 cells exposed to Ars (Table 3).

View this table:
Table 3.

Effects of 2.5 μM Ars on MT, cystine uptake, and GSH

In attempting to determine the effect of DEM on MT, GSH, and cystine transport levels in 3T3 cells, we found that 3T3 cells were extremely sensitive to DEM. DEM treatments at levels that induced cystine transport in BPAEC completely killed the 3T3 cells. DSF, another sulfhydryl reactive compound, has been reported to increase cystine transport and GSH in BPAEC (13). DSF (25 μM) also caused significant increases in intracellular MT (229 ± 73% of control) and cystine transport (294 ± 6% of control) in 3T3 cells. Changes in GSH in DSF-treated 3T3 cells were not significant (117 ± 25% of control).

As we have noted, when 3T3 cells were exposed to DEM, the DEM was very toxic at levels well below those that have previously been reported to be nontoxic in BPAEC and type II cells (9, 11, 20). We have determined that 3T3 cells are extremely susceptible to GSH depletion by DEM (Fig.1). Essentially 100% of the total cell GSH was depleted by exposure to 50 μM DEM for 4 h. In contrast, it has previously been reported that treatment of BPAEC or type II lung epithelial cells with 50 μM for 2–4 h depleted >10% of the total cellular GSH (3, 11, 19). The fact that DEM depletes GSH primarily through a reaction catalyzed by glutathione transferase (12) suggests that the 3T3 cells have a significantly higher level of transferase activity than BPAEC or other cells discussed in the literature (2, 9, 15, 19, 22).

Fig. 1.

Diethyl maleate (DEM) was added in increasing concentrations to NIH/3T3 cells, and intracellular levels of glutathione (GSH) were measured after 4 h. Values are means ± SD for 4 replicate plates.

Effects of inhibition of cystine uptake on MT and GSH levels.

Glu has been shown to be an effective competitive inhibitor of cystine transport by the xc system in a number of cell types (7, 12). When 5 mM Glu was added to 3T3 cells, GSH levels were decreased to <25% of control levels by 24 h (Fig.2). Increases in GSH after exposure to Cd were also prevented by the presence of Glu. In contrast, MT levels in the control cells were not significantly blocked by Glu nor was the induction of MT synthesis by Cd prevented. Similar effects were seen in BPAEC (Fig. 3). GSH levels were also significantly depleted by exposure to Glu for 24 h, and Cd-induced increases in GSH were substantially blocked by Glu. MT levels were not significantly affected by Glu in either control or Cd-treated BPAEC.

Fig. 2.

CdCl2 (5 μM) and/or glutamic acid (pH 7.4; 5 mM) was added to 3T3 cells. After 24 h, cells were harvested, and assays for metallothionein (MT) and GSH were performed on each group. Values are means ± SD for 10 individual plates from 2 separate experiments. Cd, cadmium. * Means were less than values for equivalent groups without glutamate.

Fig. 3.

CdCl2 (5 μM) and/or glutamic acid (pH 7.4; 5 mM) was added to bovine pulmonary artery endothelial cells (BPAEC). After 24 h, cells were harvested, and assays for MT and GSH were performed on each group. Values are means ± SE for 10 individual plates from 2 separate experiments. * Means were less than values for equivalent groups without glutamate.


We have shown that both GSH and MT levels can be increased in BPAEC and 3T3 cells after exposure to Cd and Ars. The magnitude of the increases varies between the cell types. GSH level expressed as a percentage of control level is increased significantly more in BPAEC than in 3T3 cells exposed to either Ars or Cd. Cystine uptake was also increased, at least to some extent, in both cell types. Increases in cystine uptake in BPAEC tended to be higher than in 3T3 cells, but the differences were not statistically significant.

Our data (Table 2) show that 3T3 cells also have higher basal levels (2- to 3-fold) of both MT and GSH per cell. We have noted, however, that the 3T3 cells are somewhat larger than the BPAEC. We have determined using cell size measurements obtained by our Coulter counter channelizer that the 3T3 cells had an average volume of 1.95 × 10−3 compared with 1.16 × 10−3 fl/cell for BPAEC. This could account for some, but not all, of the difference in total GSH and MT expressed per cell.

It has previously been reported that xc , the oxidant-inducible and sodium-independent transport system for cystine, is common to a variety of cells from various species (3-7, 10-15). In addition to BPAEC, this system is found in murine peritoneal macrophages, human diploid fibroblasts, bovine smooth muscle cells, and fibroblasts isolated from rat lungs. Cystine uptake increases in these cells occurred in parallel with GSH increases, indicating that increased cystine uptake might be used to provide the necessary intracellular cysteine for GSH synthesis when the cells were subjected to various stresses.

The NIH/3T3 cells appear to have a cystine transport system that is similar to the xc system. The 3T3 cells responded qualitatively similarly to sulfhydryl active agents, as did other cell types. One exception is the response of the cells to DEM. The 3T3 cells were very sensitive to DEM. It was toxic to these cells at levels well below those commonly used to deplete GSH in other cell types. Depletion of GSH in 3T3 cells occurred at levels well below those previously reported for fibroblasts and endothelial cells (see Fig. 1; Refs. 3, 4, 7, 11, 20).

Both BPAEC and 3T3 cells appear to be dependent on cystine transport for the synthesis of GSH in control cells as well as in cells exposed to Cd (Figs. 2 and 3). Glu is unlikely to inhibit GSH synthesis directly. Glu itself is a substrate for GSH synthesis and actually has been reported to increase GSH synthesis in vitro by blocking feedback inhibition of γ-glutamylcysteine synthetase by GSH (18). Treatment of 3T3 cells with Glu depleted GSH to a greater extent than similar treatment of BPAEC. This suggests various possibilities. For example, GSH turnover in 3T3 cells may be more rapid than that in BPAEC; the cystine transport system in 3T3 cells may be more sensitive to competitive inhibition by Glu; or basal levels of cysteine may be lower in 3T3 cells than in BPAEC, and thus GSH synthesis might be more dependent on continuing cystine uptake.

In contrast, the synthesis of MT in either cell type was not significantly inhibited by Glu in either Cd-treated or untreated cells. The fact that MT and GSH presumably compete for the same cysteine pool suggests that available intracellular cysteine is preferentially used for formation of the cysteinyl-tRNA complex required for protein synthesis and is only secondarily available for the non-tRNA-mediated two-step enzymatic synthesis of GSH from Glu, cysteine, and glycine. From the data in Tables 1 and 2, one can compare the total amounts of cysteine in MT and GSH in control and Cd-treated cells in the two cell types. (Calculations are based on the fact that cysteine is approximately one-third of the total weight for both GSH and MT.) Total cysteine in intracellular MT is ∼5.2% of total cysteine in GSH in control BPAEC and 6.5% of the total GSH cysteine in cells treated with CdCl2. In 3T3 cells, however, MT contained ∼11.7% of the cysteine in GSH for control cells and up to 32% of the total amount of GSH cysteine after treatment with CdCl2. Thus cysteine utilization for MT synthesis is unlikely to interfere significantly with GSH requirements for BPAEC but may significantly compete with GSH synthesis for available cysteine in the 3T3 cells, particularly in cells in which MT synthesis is induced by Cd or other agents.

The competition for available cysteine is demonstrated even more strongly in Fig 2. When Cd stress is combined with inhibition of cystine uptake in the 3T3 cells, the GSH levels are actually depleted, whereas MT levels are increased. This results in the total cell cysteine in MT being approximately twice the total cell cysteine in GSH. This type of situation may have in vitro analogs. Generally, in liver and other tissues, a much larger fraction of the cysteine pool is incorporated into GSH than into MT. In livers of rats at birth, however, the total amount of cysteine in MT is nearly equal to that in GSH (27). Whether the low GSH levels in newborn animals reflect an immaturity of the mechanisms by which cells utilize extracellular cysteine sources or whether they reflect deficiencies in extracellular cysteine and its precursors was not determined.

Reports from in vivo studies confirm that decreasing available cysteine in adult animals can result in lower GSH levels but not in lower levels of MT. Fasting has been reported to deplete liver GSH in rats up to 50% (25). In these studies, MT levels were not correspondingly reduced but actually elevated, perhaps because of oxidant stress resulting from inadequate GSH levels. Similar results have been reported for rats fed a sulfur-amino acid-deficient soya-based diet (27). Our data confirm at the cellular level that MT synthesis is less affected than GSH synthesis by treatments that reduce intracellular cysteine availability.


  • Address for reprint requests: I. Susanto, Div. of Pulmonary Diseases/Critical Care Medicine, Dept. of Medicine, The Univ. of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr., San Antonio, TX 78284-7885.


View Abstract