Copper incorporation into superoxide dismutase in Menkes lymphoblasts.

The incorporation of copper into Cu,Zn-superoxide dismutase (SOD) was examined in Menkes lymphoblasts that express a genetic defect of copper metabolism. SOD activity was ≈40% higher in Menkes than normal lymphoblasts. Since Menkes lymphoblasts contain elevated copper levels, the higher SOD activity is most likely due to near copper saturation of an apoSOD pool that is in normal lymphoblasts. Cycloheximide markedly inhibited 64Cu(II) incorporation into SOD in Menkes lymphoblasts under conditions in which no significant, de novo synthesis of SOD protein was detected with normal lymphoblasts. The maximal amount of 64Cu incorporation into newly synthesized SOD in Menkes lymphoblasts was approximately equal to the maximal amount of 64Cu that could be incorporated into the apoSOD pool in normal lymphoblasts. The increased synthesis of SOD in Menkes lymphoblasts may play a protective role against copper toxicity in Menkes lymphoblasts. The protonophore, CCCP markedly inhibited 64Cu incorporation into SOD in both normal and Menkes lymphoblasts, which is consistent with 64Cu incorporation into SOD within a membrane-bounded compartment in both cell types. When 64Cu-incorporation into SOD was blocked with CCCP, copper accumulated in a Superose column fraction that contains S-adenosylhomocysteine hydrolase (SAHH), which has a high affinity for copper. SAHH may play a role in delivering copper to SOD.

Menkes disease is a fatal, X-linked disease of copper metabolism which affects copper delivery to several copper enzymes, the synthesis of collagen and elastin, and brain development (1)(2)(3). A candidate gene (cDNA) for the Menkes defect was recently cloned (4 -6), and the deduced amino acid sequence is homologous to metal-transporting, P-type, membrane ATPases from procaryotes (7,8). Vulpe et al. (4)postulated that the protein encoded by the Menkes gene was within a subcellular membrane. Menkes lymphoblasts are derived by Epstein-Barr virus transformation of lymphocytes from Menkes patients (9). These cells maintain their characteristic phenotype of excess net copper accumulation when incubated with copper (9) and are also sensitive to copper toxicity.
Normal lymphoblasts were recently found to contain a pool of apoSOD 1 that was activated when these cells were incubated with copper (10). The results reported here suggest that the apoSOD pool is nearly copper-saturated in Menkes lymphoblasts when grown in standard culture medium without added copper. Moreover, new SOD synthesis was detected in Menkes lymphoblasts under conditions in which no de novo synthesis of SOD was detected with normal lymphoblasts. Also, CCCP markedly inhibited 64 Cu incorporation into SOD in both normal and Menkes lymphoblasts. The results suggest that SOD synthesis is induced in Menkes lymphoblasts to replenish the apoSOD pool in a membrane-bounded compartment as a protective mechanism against potential copper toxicity.
Cell Cultures-Human lymphoblastoid cell lines developed by transformation of peripheral B lymphocytes with Epstein-Barr virus were obtained from NIGMS, Human Genetic Mutant Cell Repository (Coriell Institute for Medical Research, Camden, NJ). Normal human lymphoblasts (repository number GM03798) were from a normal 10-year-old Caucasian male. Menkes syndrome (kinky hair disease) human lymphoblasts (repository number GM01982) were from a 2-year-old Caucasian male.
Human lymphoblasts were grown as suspension cultures in RPMI 1640 medium (Sigma) supplemented with 5% fetal calf serum and 5% newborn calf serum (Intergen, Purchase, NY). Cell cultures were maintained in logarithmic growth phase (Ϸ0.7 ϫ 10 6 cells/ml) by replacing cell suspensions with fresh medium as described previously (10). Cells were incubated with varying concentrations of 64 Cu(NO 3 ) 2 , specific activity, Ϸ14 mCi/mg of copper (Buffalo Materials Research Center of the State University of New York at Buffalo) for the indicated durations in the normal growth medium. Cells from three 100-ml culture bottles of Ϸ0.7 ϫ 10 6 cells/ml were washed three times with cold phosphate-buffered saline, pH 7.4, before combining the samples for homogenization.
Preparation of Cytosols-Cells were homogenized with 150 strokes of a motor-driven (1000 rpm) homogenizer (Thomas Teflon pestle) in 0.4 ml of isotonic HEPES sucrose buffer (0.25 M sucrose, 5 mM HEPES pH 7.4). Phenylmethylsulfonyl fluoride (40 g/ml) and leupeptin (0.5 g/ ml) were added to the homogenization buffer to inhibit proteolysis. Homogenates were centrifuged for 2 min at 1930 ϫ g at 4°C, and the supernatants were centrifuged for 60 min at 100,000 ϫ g. The supernatants were filtered through a 0.22-m Milex GV syringe filter (Millipore, Bedford, MA) before applying to the Superose column. Typical protein concentrations of lymphoblast cytosols were 10 -15 mg/ml as determined by the bicinchoninic acid assay method using bovine serum albumin as a standard (11).
Superose 12 HPLC-The Superose 12 column was equilibrated with 0.05 M HEPES, 0.1 M NaCl, pH 7.4. Samples (200 l) were injected, and the columns were eluted at a flow rate of 0.4 ml/min, as described previously (10). The amount of 64 Cu in each tube (220 l) was determined with a LKB gamma counter (model 1282), correcting for decay by a program within the counter. The radioactivity in each tube was expressed as picograms of 64 Cu/mg of total cytosolic protein, and the total amount of 64 Cu in a column fraction was determined by adding the based spectrophotometric method (12). One unit of SOD activity was the amount of cytosolic protein required to quench the absorbance due to superoxide formation in controls by 50%. Assays were done in quadruplicate, and the statistical significance of differences was determined by a two-tailed, Student's t test. The SOD activity obtained in presence of 1 mM KCN, representing Mn-SOD, was substracted from the total SOD activity to determine the Cu,Zn-SOD activity.
MT-1 Expression Vector-The vector pSG-HEBO was obtained by combining a pSG5 vector (Stratagene, La Jolla, CA), and an episomal pHEBO vector for Epstein-Barr-transfected lymphoblasts (13) (kindly provided by Dr. John Yates, Roswell Park Cancer Institute, Buffalo, NY). The SfiI-SalI fragment from pSG5 carries part of an SV40 early promoter, the rabbit ␤ globin intron 2, a BamHI site that was used for subcloning the MT-cDNA and an SV40 early gene poly(A) recognition sequence. A SfiI-SalI fragment from the pHEBO plasmid carries the remaining part of an SV40 early promoter, Epstein-Barr virus oriP (origin of replication for stable replication in lymphoblasts), a hygromycin resistance gene for selection of transfected lymphoblasts, an Escherichia coli compatible origin of replication (Col E1), and an ampicillin resistance gene for propagation of the vector in E. coli. Stable propagation in Epstein-Barr virus transformed lymphoblasts was accomplished by expression of the EBNA-1 gene provided by Epstein-Barr virus in lymphoblasts. The cDNA for monkey metallothionein 1 (MT-1) was obtained from Dr. Dean Hamer (NIH, Bethesda, MD) (14). The MT-1 cDNA was subcloned into pSG-HEBO through a unique BamHI site to give the pSG-HEBO(MT-1) plasmid.
Transfection of Human Lymphoblasts for Stable Expression of MT-1-Transfection was performed by electroporation by a modification of the procedure described by Anderson et al. (15). Lymphoblasts were pelleted and resuspended in standard growth medium at a concentration of Ϸ1 ϫ 10 7 cells/ml, and 160 l of the cell suspension were transferred to a 0.4-cm electroporation cuvette (Bio-Rad). Plasmid DNA (200 -250 g/ml) in 40 l of RPMI 1640 medium was added, and the cuvette was exposed to a 200-V pulse for 60 -80 ms at an internal capacitance of 960 F (Gene Pulser Electroporation apparatus, Bio-Rad). The cell suspension was transferred to conditioned medium obtained from 48-h lymphoblast cultures, and 48 h after transfection, hygromycin (Sigma) was added to the medium (150 g/ml). Cells were grown in the antibiotic-containing medium until the culture contained only hygromycin-resistant transfected cells (about 20 days).

RESULTS
The Distribution of 64 Cu in Menkes Lymphoblast Cytosols after Incubating with 64 Cu(II)-Menkes and normal lymphoblasts were incubated for 2 h with 7 M 64 Cu(II). The cell cytosols were isolated and fractionated by Superose chromatography. No significant differences were detected in the amount of 64 Cu bound to proteins in the void volume fractions from normal and Menkes lymphoblasts (Fig. 1). Three additional 64 Cu-binding fractions were reproducibly detected with both normal and Menkes lymphoblast cytosols. Fraction I ( Fig. 1) contains SAHH which was recently proposed to play a role in copper metabolism (16,17). The 64 Cu-protein fraction peaking at tube 37 is SOD (10), and the fraction labeled MT contains metallothionein(s) as indicated below. Menkes cytosols showed much higher 64 Cu binding in the MT-containing fraction than normal cytosols, which is consistent with previous studies (9). Menkes lymphoblast cytosols also showed less incorporation of 64 Cu into SOD, and significantly more 64 Cu-binding in fraction I than normal lymphoblasts (Fig. 1). However, by 15 h, the amounts of incorporation of copper into SOD and copper binding to a fraction I protein(s) in Menkes lymphoblasts were similar to what was detected with normal lymphoblasts (see below). As with normal lymphoblasts (10), no 64 Cu incorporation or isotopic exchange into SOD was detected when 64 Cu was added directly to cytosols rather than to whole lymphoblasts (data not shown).  2). The estimated maximum amounts of 64 Cu incorporated into SOD were similar in normal and Menkes lymphoblasts (4.6 and 4.9 ng 64 Cu per mg of protein, respectively). However, the total cytosolic 64 Cu concentration at half-maximal levels of copper incorporation into SOD in Menkes lymphoblasts (206.5 ng/mg of protein) was significantly higher than with normal lymphoblasts (6.3 ng/mg of protein) (Fig. 2). Since the higher total cytosolic copper of Menkes cytosols at any extracellular copper concentration is, for the most part, due to higher levels of copper bound to elevated MT (9), these results imply that either MT⅐Cu is unavailable to SOD and/or that higher concentrations of copper are required for copper incorporation into SOD in Menkes lymphoblasts.
The cytosolic copper concentration dependence for binding of 64 Cu to a protein(s) in fraction I, which contains SAHH, was also determined. The results are plotted in Fig. 3 as percent of maximal 64 Cu bound versus cytosolic copper. Interestingly, in both normal and Menkes lymphoblasts, 64 Cu-binding in fraction I showed the same relative concentration dependence (percent of maximal) as 64 Cu incorporation into SOD (Fig. 3).
The Effect of Cycloheximide on 64 Cu Incorporation into SOD in Menkes Lymphoblasts-Menkes lymphoblasts were preincubated for 3 h with 5 g/ml cycloheximide and then incubated for an additional 15 h with 7 M 64 Cu(II) with cycloheximide. Cytosols were isolated and fractionated on Superose. Inhibition of protein synthesis by cycloheximide caused a significant decrease in the amount of 64 Cu binding to MT (Fig. 4) in Menkes lymphoblasts as was also observed with normal lymphoblasts (10). However, in contrast to normal lymphoblasts, cycloheximide markedly inhibited incorporation of 64 Cu into SOD in Menkes lymphoblasts (Fig. 4). The effect of cycloheximide on decreasing cytosolic 64 Cu by Menkes lymphoblasts was similar in magnitude to its effect on normal lymphoblasts (10) and not large enough to account for its large inhibitory effect on 64 Cuincorporation into SOD. Thus, in striking contrast to normal lymphoblasts (10), copper seems to be incorporated mainly into newly synthesized SOD in Menkes lymphoblasts rather than into a preexisting pool of apoSOD.
The  (Table I). Surprisingly, although cycloheximide markedly inhibited 64 Cu incorporation into newly synthesized SOD (Fig. 4), that had no significant effect on the total SOD activity in Menkes lymphoblasts (Table  I). Also, in marked contrast to normal lymphoblasts, no significant increase in SOD activity was detected with Menkes lymphoblasts when incubated with copper (Table I) Although, an approximately 8.4-fold increase in 64 Cu bound within the MT fraction was detected, this led to a small decrease (Ϸ19%) in the amount of copper incorporated into SOD rather than any increase (Fig. 6, Table II).
Moreover, under these conditions, 64 Cu incorporation into SOD was slightly less than predicted from the cytosolic 64 Cu concentration dependence for lymphoblasts containing normal amounts of MT. Thus, 64 Cu incorporation into SOD was clearly not increased by the large increase in 64 Cu bound to MT in lymphoblasts transfected with the MT expression vector.
The Effects of CCCP on 64 Cu Incorporation into SOD, and the Distribution of 64 Cu in Menkes Lymphoblast Cytosols-Increased net copper accumulation due to decreased copper efflux is a phenotypic characteristic of Menkes lymphoblasts. In the presence of CCCP, net copper uptake by normal lymphoblasts is elevated to approximately the same abnormally high level as observed with Menkes lymphoblasts (18). Since dicyclohexylcarbodiimide, which is a specific inhibitor of ATP-synthase proton pumps had no effect on copper accumulation, the effect of CCCP was most likely due to its action as a general protonophore. Although CCCP increased net copper uptake, it markedly inhibited 64 Cu incorporation into SOD in both normal and Menkes lymphoblasts (Fig. 7). Concomitant large increases in 64 Cu in fraction I were detected (Fig. 7). This is an unusual result because impaired copper utilization in cells usually leads to MT induction and increased copper binding to MT rather than increased copper binding to other copper binding proteins. CCCP had only small inhibitory effects on copper binding to MT in normal and Menkes lymphoblasts, which is consistent with minimal effects of CCCP on protein synthesis under the   conditions of these experiments. Thus, CCCP specifically affected copper incorporation into SOD and 64 Cu binding by a fraction I protein(s).

DISCUSSION
Normal lymphoblasts were recently found to contain a significant pool of apoSOD, which was estimated to be Ϸ35% of the total SOD in these cells (10). Copper was incorporated into this apoSOD pool, and SOD activity increased proportionally when copper was added to normal lymphoblast culture media (10). The data reported here indicate that the activity of SOD in Menkes lymphoblasts grown under standard conditions without added copper was about the same (Ϸ40% higher) as the maximal possible SOD activity (Ϸ35% higher) in normal lymphoblasts after copper was added (10). Since Menkes lymphoblasts accumulate and retain higher levels of copper than normal lymphoblasts, the higher SOD activity in Menkes lymphoblasts probably reflects near saturation of the apoSOD pool in Menkes lymphoblasts.
A striking abnormal characteristic of Menkes lymphoblasts is significant induction of new synthesis of SOD protein over a Ϸ15-h incubation period. No significant SOD synthesis was detected under equivalent conditions with normal lymphoblasts (10), and the increased time dependence for copper incorporation into SOD in Menkes lymphoblasts was probably due to the requirement for new SOD synthesis. Curiously, the new holo-SOD synthesis did not appear to significantly contribute to the total SOD activity in Menkes lymphoblasts, because cycloheximide had no significant effect on SOD activity. This is despite the fact that the maximal amount of copper incorporation into newly synthesized SOD in Menkes lymphoblasts was of similar magnitude to the maximal level of copper incorporation into apoSOD in normal lymphoblasts. The lack of a significant effect of cycloheximide on SOD activity in Menkes lymphoblasts suggests that the induction of SOD synthesis and the rate of degradation or removal of SOD may both be regulated to maintain the level of holo-SOD in Menkes lymphoblasts near a maximal possible level. Equal rates of SOD synthesis and removal would also account for the fact that, unlike with normal lymphoblasts (10), 64 Cu(II) incorporation into SOD was not associated with any significant increase in total SOD activity in Menkes lymphoblasts.
Although the function of the apoSOD pool is unknown, the similar possible maximal levels of copper incorporation into apoSOD in normal lymphoblasts and newly synthesized SOD in Menkes lymphoblasts suggests a compensatory response to depletion of the apoSOD pool in Menkes lymphoblasts. Induction of SOD synthesis and removal of SOD to maintain a maximal possible steady-state level of holo-SOD may function as a protective response against potential copper toxicity. Menkes fibroblasts are sensitive to copper toxicity (19), and Menkes lymphoblasts also show increased sensitivity to copper toxicity (unpublished results). Thus, increased SOD synthesis in Menkes lymphoblasts may provide a pool of SOD which can sequester excess copper, which is then rapidly removed to minimize copper toxicity.
The fraction of total cellular copper that is in the cytosolic fraction is much higher in Menkes than normal lymphoblasts, 2 and this copper is mostly bound to MT, which is elevated in Menkes lymphoblasts (9). Elevated copper in the cytosolic fraction was also reported for animal models of Menkes disease (20,21). This may reflect impairment of copper delivery to, and entry into, various organelles due to impaired function of the Menkes protein and induction of MT in response to the accumulation of nonutilizable copper. The results reported here with normal cells containing elevated levels of MT are consistent with elevated MT being a response to elevated copper rather than elevated MT being a significant contributing factor to impaired utilization of cellular copper in Menkes cells, as elevated MT levels had little or no effect on copper delivery to apoSOD in normal lymphoblasts.
The large inhibitory effect of the protonophore, CCCP on copper incorporation into SOD in normal lymphoblasts suggests that the apoSOD pool is within an intracellular, mem-2 N. Petrovic, A. Comi, and M. J. Ettinger, unpublished data. brane-bounded compartment rather than in the cytoplasm. The fact that copper does not activate the apoSOD when added directly to lymphoblast cytosols rather than to whole cells is also consistent with compartmentalization of the apoSOD pool. Since CCCP had similar effects on copper incorporation into SOD in normal and Menkes lymphoblasts, copper incorporation into SOD in Menkes lymphoblasts may occur in the same cell compartment as in normal lymphoblasts. That is consistent with the hypothesis that new SOD synthesis in Menkes lymphoblasts may represent a cellular response to replenish the apoSOD pool which is apparently depleted in Menkes lymphoblasts, and is also consistent with the similar maximal 64 Cu incorporation into apoSOD in normal lymphoblasts and newly synthesized SOD in Menkes lymphoblasts. While SOD has been detected in nuclei and perioxosomes (22)(23)(24)(25), the compartment hypothesized for the apoSOD pool in lymphoblasts remains to be determined.
As a general protonophore, CCCP may inhibit copper delivery to SOD by dissipating a proton gradient that is required for co-transport of copper into the cellular compartment where copper incorporation into SOD is suggested to occur. Interestingly, the procaryote homologues of the putative Menkes ATPase, copper transporter may be H ϩ -antiport systems (26). The defect in the Menkes gene in the patient whose lymphoblasts were used for our studies was recently reported. A fivebase deletion resulting in a frameshift and presumably inactive protein was detected by reverse transcription-polymerase chain reaction (27). Thus, the similar results with CCCP with normal and Menkes lymphoblasts and the high activity of SOD in Menkes lymphoblasts suggests that a subcellular membrane copper transporter other than the Menkes protein is involved in delivery of copper to this SOD pool in lymphoblasts. This putative copper transport protein may be the homologue of the Menkes protein in humans that was recently identified as a candidate for the defect in Wilson's disease (28 -31), or alternatively, another copper transporter that requires proton cotransport. Alternatively, the effect of CCCP may be due to sulfhydryl modification of a specific protein, as previously suggested (18).
The molecular species, pool, and mechanisms involved in copper delivery to SOD remain unknown. The results reported here suggest that MT⅐Cu is not a major source of SOD copper, and that MT⅐Cu apparently is not in rapid equilibrium with the cellular copper pool(s) that is a source(s) for SOD-Cu because increased copper on MT had no significant effect on the delivery of copper to apoSOD in normal lymphoblasts. However, increased copper in the Superose copper-binding fraction I was detected in each case in which incorporation of copper into SOD was decreased (10), and the apparent concentration dependence, normalized to maximal possible incorporation, was the same for copper incorporation into SOD and a protein(s) in fraction I in both normal and Menkes lymphoblasts. That suggests that the level of 64 Cu binding to a fraction I copperbinding protein is determined, in part, by the degree of saturation of the apoSOD pool. The results with CCCP were particularly striking in that a marked accumulation of copper in fraction I occurred when copper incorporation into SOD was blocked by CCCP. All of these results are consistent with copper bound to a fraction I protein(s), either residing within a copper pool that supplies copper to SOD or being in equilibrium with a pool or species that delivers copper to SOD. A protein that has been identified in the Superose fraction I that has a high affinity for copper is SAHH, and this protein has been proposed to have a role in intracellular copper trafficking (16,17).