JOURNAL OF CELLULAR PHYSIOLOGY 150:640-646 (1992)
Tunicamycin Reduces Na+-K+-Pump Expression in Cultured Skeletal Muscle SANDRA V. ALBOIM, ASIA BAK, AND SANFORD R. SAMPSON* Health Sciences Research Center and the Otto Meyerhoff Center, Department of l i f e Sciences, Bar-/Ian University, Ramat-Can 52900, lsrael The purpose of this study was to examine effects of tunicamycin (TM), which inhibits core glycosylation of the p-subunit, on functional expression of the Na+-K+ pump in primary cultures of embryonic chick skeletal muscle. Measurements were made of specific-['HI-ouabain binding, ouabain-sensitive 86Rb uptake, resting membrane potential (Em), and electrogenic pump contribution to Em (E,) of single myotubes with intracellular microelectrodes. Growth of 4-6-day-old skeletal myotubes in the presence of TM (1 pg/ml) for 21-24 hr reduced the number of Na+-K+ pumps to 60-90% of control. Na+-K+ pump activity, the level of resting Em and E,, were also reduced significantly by TM. In addition, TM completely blocked the hyperpolarization of Em induced in single myotubes by cooling to 10°C and then re-warming to 37°C. Effects of tunicamycin were compared with those of tetrodotoxin (TTX; 2 x l o p 7 M for 24 hr), which blocks voltage-dependent Na+ channels. TM produced significantly greater decreases in ouabain-binding and Em than did TTX, findings that indicate that reduced Na'-K+ pump expression was not exclusively secondary to decreased intracellular Na+, the primary regulator of pump synthesis in cultured muscle. Similarly, effects of TM were significantly greater than those of cycloheximide, which inhibits protein synthesis by 95%. These findings demonstrated that effects were not due to inhibition of protein synthesis. We conclude that glycosylation of the Na+-K+ pump p-subunit is required for full physiological expression of pump activity in skeletal muscle.
A number of membrane proteins require N-glycosylation for adequate cellular expression. Thus, it has been shown that tunicamycin (TM), which inhibits this process, decreases the density of voltage-dependent Nachannels in a number of cell types including cultured neuroblastoma (Waechter et al., 19831, cultured chick skeletal muscle (Bar-Sagi and Prives, 1983), neurons (Zona et al., 1990) and Xenopus oocytes injected with brain mRNA (Sumikawa et al., 1988). Similarly, carbohydrate moieties are required for expression of acetylcholine receptor (AChR) in cultured chick myotubes (Prives and Olden, 1980) and Xenopus oocytes (Sumikawa and Miledi, 1989), glucose transport in chick embryo fibroblasts (Olden et al., 1979), high affinity carriers for amino acid transport (Deas and Erecinska, 1989), insulin receptors (Ronnet and Lane, 1981), and for maintenance of normal high affinity of growth hormone receptors in rat adipocytes (Szecowa et al., 1990). The importance of N-glycosylation in expression of Na+-K+pump activity is not entirely clear. On the one hand, several studies indicate that @-subunitN-glycosylation may not be essential for expression of Na+-K+pump activity. Thus, Olden et al. (1979) reported that TM treatment did not affect activity of this enzyme in chick fibroblasts, and Tamkun and Fambrough (1986), in a n extensive study of this enzyme in chick sensory neurons, concluded that n-glycosylation of the (3-subunit is not important for either intracellular transport, assembly or membrane insertion of Na+-K+-purnpsub0 1992 WILEY-LlSS, INC
units. They reported similar findings for chick skeletal muscle. The role of N-gycosylation in expression of Na+-K+-pumpactivity, however, was not investigated. On the other hand, Zamofing et al. (1988) found that treatment with TM results in a decrease in amount of newly synthesized @- and a-subunit. They suggested further that efficient cellular accumulation of the a-subunit may depend on @-subunitaccumulation. The same group (Zamofing et al. 1989) later reported that while TM reduced epithelial Na+ transport in toad bladder cells, Na+-K+ transport (Na+-K+ ATPase activity) was not altered. Because of the importance of N-glycosylation in expression of functional activity of other membrane proteins in excitable membranes, we have examined the role of this process for resting and stimulated activity of the Naf-Kt-pump in cultured skeletal muscle. The Na+-K+-pumpin this preparation has been characterized as having a single binding site for ouabain and, according to Scatchard analyses, there is only one class of Na-K pump units (Brodie et al., 1987). In addition, the activity of the Na+-K+-pumpcontributes a n important and easily measurable electrogenic component (E,) to resting membrane potential (Ern;Brodie et al.,
Received June 17,1991; accepted October 21,1991. * To whom reprint requestsicorrespondence should be sent.
TUNICAMYCIN REDUCES Na ’ -K ‘ -PUMP EXPRESSION
1987; Brodie and Sampson, 1985, 1989a,b). In this study, we have investigated effects of TM on the amount and activity of Na-K pump sites as assessed by the specific binding of L3H1-ouabainand uptake of “Rb, and by levels of resting Em and E, in primary cultures of skeletal muscle.
EXPERIMENTAL PROCEDURES Preparation of cultures Skeletal muscle cell cultures were prepared by mechanical dissociation of the thigh muscles of 11-12-day chick embryos, a s already described (Brodie and Sampson, 1985). Most fibroblasts were eliminated by preplating, and primary cultures were established by plating a t a density of 3 x lo5 cellsiml (1.5 ml/dish) in 35 mm plastic tissue culture dishes (Nunc) coated with collagen-gelatin. Cultures were grown in a water-saturated atmosphere of 95% air-5% CO, at a temperature of 37°C. The composition of the growth medium was a s follows: Dulbecco’s Minimal Essential Medium (Gibco), 83%; horse serum (Biolab) 15%; chick embryo extract (from 11-12-day old embryos), 2%. Fresh growth medium was replaced every 3 days. Unless otherwise indicated, experiments were done on cultures of age 6-8 days in vitro (DIV). Measurement of r3H]ouabainbinding The binding of [3H]ouabain was determined by a modification (Brodie et al., 1987) of the method of Vandenberg and Kaufman (1981). Cultures were washed three times with 2 ml K-free phosphate buffer (pH 7.4) and then incubated in 1.0 ml ofbuffer containing 75 nM [‘Hlouabain a t 37°C. In preliminary studies, we determined the concentration-binding characteristics of [’Hlouabain by chick myotubes. We found that maximal binding was reached a t a ouabain concentrations of 60-75 nM. Non-specific binding was determined in the presence of 1mM unlabelled ouabain. After incubation for 30 min, cells were washed five times with cold buffer. They were solubilized in 1% Triton x 100, and radioactivity was assayed by standard procedures. Control studies established that ouabain binding by nonmuscle elements was less that 5-10% of the total. Measurement of Na+-K+-pumpactivity 86Rb uptake was measured by a modification of the method of Vandenburgh and Kaufman (1981) as described (Brodie et al., 1987; Brodie and Sampson, 1989a). Rate of u take was measured in 1ml PBS containing 2.5 pCi PRb. Uptake was stopped after 15 min by aspiration of the medium immediately followed by 4 rapid rinses in cold (4°C)PBS. One ml of 1%Triton x 100 was added to each culture dish, and the cells were scraped from the dish and added to 1.0 ml H 0 in a scintillation vial. Samples were counted in a 3H-window of a Tricarb liquid scintillation counter. Uptake specifically related to Na+-K+-pumpactivity was determined by subtraction of the amount taken up in the presence of ouabain (10-3M) from that in its absence. Measurement of Em Intracellular recordings of membrane potentials were obtained by techniques as described in other reports (Brodie et al., 1987, 1989; Brodie and Sampson,
641
1989a,b). For recording, culture dishes were removed from the incubator, and the medium was changed to a phosphate-buffered saline (PBS) solution (pH 7.32) of the following composition (mM): NaCl, 135; KCI, 2.7; Na,HPO,, 1.4; KH,PO,, 1.4; MgCl,, 0.5; CaCl,, 0.7; glucose, 22. The temperature of the culture was controlled at predetermined levels by placing the dish containing the culture in a copper sleeve heated by circulating water from a thermostatically controlled heater, and this assembly was placed on the stage of a n inverted phase-contrast microscope (Nikon). Em was recorded from single myotubes with glass micropipettes filled with 2.8 M KC1 and having internal resistances of 20-50 M a . The microelectrodes were connected to the input of a microelectrode preamplifier (Winston Electronics, San Francisco), and recordings were displayed on one beam of a storage oscilloscope (Tektronix 5113). The verticle signal output from this beam was fed to a pulse height selector-rate meter for measurements of frequency of action potential. Em was measured a s the highest stable (minimum 30 sec) potential recorded during a n impalement. Tip potentials were generally between 1and 3 mV. Earlier studies in our laboratory have shown that ouabain (1 mM) added to chick myotubes in cultures causes a n almost immediate (within 5-10 sec) depolarization of 15-25 mV (Brodie and Sampson, 1989b). Accordingly, the contribution of the Na+-K+ pump to Em was determined a s the difference between the mean in Em recorded within 5-10 min after addition of ouabain (1mM) from that recorded prior to ouabain administration.
Analysis of data For ouabain binding and “Rb uptake studies, measurements were made on triplicate samples of control and treated groups in each individual experiment. For determination of Em, measurements were made from a minimum of 9 cells in each of 2 dishes of control and treated groups in each experiment. The mean values and standard errors were calculated, and data were analyzed statistically by Student’s t-test for unpaired samples or by one-way analysis of variance (ANOVA; Instat, GraphPad, San Diego, CA). Chemicals and drugs Tetrodotoxin (TTX), tunicamycin (TM), and cycloheximide (CH) were obtained from Sigma Chemical Co. (St. Louis, MO). [’Hlouabain (specific activity 16-18 Ci/mM) and “Rb were obtained from Amersham Ltd. (UK). TM was dissolved in absolute ethanol a t a final concentration of 1mg/ml. Treatment of cells On the day of treatment, TM was diluted in DMEM to a concentration of 10 pgiml; 150 pl of this solution was added to the culture dishes which contained 1.5 ml growth medium to give a final TM concentration of 1 pgtml. Myotubes of cultures age 4-6 days were treated with TM for 24 hr. Measurement of total protein content Total protein content of the culture dishes were determined by the method of Lowry et al. (1951).
ALBOIM ET AL.
642 120
1
loo
I 60 -. rcc
0
2R
cultures, effects in 3 experiments ranging from no change to a decrease of 16%. Second, treatment of myotubes with the protein synthesis inhibitor (5 pm) cycloheximide for 24 hr caused less of a n effect on ouabain binding than did TM. Thus, in each of 2 experiments TM decreased the number of ouabain binding sites by 37 and 60%, respectively, while CH in these same experiments decreased the number of binding sites by 23 and 29%, respectively (not illustrated), and the total protein by 27%. This concentration of CH has been shown to decrease protein synthesis by 95% in muscle cell cultures (Shainberg et al., 1971). One way to determine activity of the Na+-K+-pump is by measurement of ouabain-sensitive uptake of 86Rb, which is recognized by the external K+-binding site of the pump. We examined effects of TM on Na+-K+-pump activity in 4 experiments. In each experiment, Rb uptake in TM-treated myotubes was lower than that in control, untreated cells. The TM-induced decrease in ouabain-sensitive "Rb uptake ranged from 41-92% of control in the individual experiments (mean 66.8%; P < 0.02). The contribution of the Na+-K+-pump to resting membrane potential of cultured chick myotubes has been estimated to be of the order of 15-18 mV (Brodie and Sampson, 1989b). In this study, we sought to determine if the decrease in number and activity of the Na'K+-pump obtained by treatment with TM is also expressed by change in the level of Em.This was done by comparing the mean resting Em of cells treated with TM for 24 h r before and after addition of ouabain (1 mM) with that of control, untreated myobutes of the same culture. The results of 3 experiments of this type are summarized in Figure 1. The mean (? S.E.) Em of control myotubes was -69 (? 1.2) mV, while that of TM-treated cells was -61.5 ( 2 0.8) mV (P < 0.001), and the E, was reduced to 31% of that in control myotubes. We considered the possibility that the ability of TM to decrease ouabain binding might be related to a nonselective, general interference with the li and. To examine this, we incubated myotubes in [ HI-ouabain plus various concentrations of TM, from 0.1 to 10 pg/ml, for 30 min. The effects of increasing concentrations of TM are shown in Figure 2, which shows that TM does not have a nonselective or competitive effect on ouabain binding.
1
40 -.
20 -.
0
i
OUAB
Fib
EP
Fig. 1. Effects of TM on expression of Na'-K' pump activity in cultures skeletal muscle as determined by measurements of L3H1-ouabain binding (OUAB), "Rb-uptake (Rb),and electrogenic pump component of resting membrane potential (E ), All values expressed as % of those in control, untreated sister cells. "he data for ouabain binding are the mean (* S.E.) of triplicate values obtained in 7 separate experiments (n = 21; P < 0.001). The data for Rb-uptake were obtained on triplicate samples in each of 4 experiments (n = 12; P < 0.01). The data for Emwere obtained from measurements on 9 cells in 2 dishes in each of 3 experiments (n = 54; P < 0.001).
RESULTS Na+-K+-pumpexpression in cultured skeletal muscle is readily assessed by the measurements of specific [3H]-ouabain binding, ouabain-sensitive "Rb uptake, and the level of transmembrane resting Em recorded with intracellular microelectrodes. The results of studies on the effects of TM on these 3 parameters are summarized in Figure 1. Ouabain is a specific antagonist of the Na+-K+-pump and binds to the external K+-transport site on the enzyme. This ligand has been used to quantitate the number of Na+-K+-pumpsite in a variety of different cells under different conditions. In this study, we compared the number of [3H]-ouabain binding site in TM-treated cells with that in control myotubes from the same culture. The number of ouabain binding sites in control cells ranged from 166 to 243 fmoledmg protein. These values are in the range reported by others for skeletal muscle in culture (Vandenburgh and Kaufman, 1981; Brodie and Sampson, 1989a). In each of 5 experiments, 24-hr incubation of myotubes in M TM resulted in significant decrease (P < 0.01) in the number of ouabain binding sites. In addition to effects on glycosylation, TM in the concentration used in this study has been reported to inhibit protein synthesis. That the ability of TM to impair expression of the Na+-K+-pumpis not likely to be due entirely to a general effect on protein synthesis is indicated by the following. First, TM treatment for 24 h r had only a slight effect on total protein content of the
8
Comparison between effects of TM and TTX on Na+-K+-pumpexpression The results obtained so far demonstrate that treatment of cultured myobutes with TM, which inhibits N-terminal glycosylation, causes a decrease in number of ouabain-binding sites, a decrease in Na-K pump activity, and a decrease in resting Em. These findings suggest that p-subunit glycosylation is, therefore, important for expression of Na+-K+ pump activity. Another explanation, however, may be that TM-treatment causes a decrease in internal Na+ as a result of the decreased expression of voltage-dependent Na-channels, a n effect known to occur in this a s well as other preparations (Bar-Sagi and Prives, 1983). The level of intracellular Na+ has been shown to be a major determinant of Na+-K+ pump density (Wolitzky and Fambrough, 1986; Brodie and Sampson, 1989a). To examine
TUNICAMYCIN REDUCES Na ' -K ' -PUMP EXPRESSION
643
Na-loading cells treated with TTX for 24 h r caused Em to change by the same amount as that obtained in control cells (Fig. 4). This indicates that even though the number of pump sites is reduced slightly by TTX, this fewer number is capable of responding to Na-loading to the same degree as in control.
k
3
E: 1 2 0 1 0
CON
0.1
1
10
TM CONCENTRATION (yg/ml) Fig. 2. Effects of different concentration of TM on ['HI-ouabain binding by cultured myotubes. Values expressed as in Figure 1. Each bar represents the mean f S.E. of triplicate measurements in each of 2 experiments (n = 6).
this possibility, we treated myotubes with tetrodotoxin (TTX), which blocks voltage-dependent Na-channels, thereby decreasing internal Na concentration and down-regulating Na+-K+ pumps (Wolitzky and Fambrough, 1986; Brodie and Sampson, 1989a1, and compared the effects of TTX with those of TM on specific [3H]-ouabain binding and on Em. The comparison between TM and TTX on ouabain binding and Em is shown in Figure 3. Treatment with TTX for 24 h r caused a 24% decrease in ouabain binding compared with that in control, sister cells (Fig. 3A). This effect was not statistically significant (P > 0.05) (see, however, Brodie and Sampson, 1989a). In contrast, TM caused a decrease in specific ouabain binding of 48% (P < 0.001) compared with control myotubes. The relatively large standard error can be attributed to the fact that in one of the 4 experiments of this type, TM reduced the number of ouabain binding sites by 74%, to 26% of control. Similarly, TM caused a greater decrease in Em than did TTX following treatment for 24 h r (Fig. 3B). Thus, Emwas reduced from a mean ( * S.E.) control value of -77.6 mV (+ 1.2) to -72.4 mV (+ 2.3) by treatment with TTX (P > 0.051, but to -61.3 mV (k 0.9) by treatment with TM (P < 0.001). In another series of experiments, we examined effects of TM and TTX on Na+-K+ pump activation by Naloading. This was accomplished by first cooling the cells to 10°C for 30 min and then rewarming them to 37°C. Em was recorded 10-20 min after the cells were rewarmed. In control cells, Na-loading caused Em to increase significantly (P < 0.0001) from -67.7 ( + 4.5) to -74.5 (+ 4.7), whereas in TM-treated cells, this procedure had no effect on Em (before cooling, -57.9 + 3.9 mV; rewarmed, -58.2 ? 4.5 mV; P > 0.4). In contrast,
Effects of TM on [3Hl-ouabainbinding in fibroblasts Other studies regarding inhibition of n-glycosylation by TM on Na+-K' pump expression were done on fibroblasts derived from various sources (Olden et al., 1979; Zamofing et al., 1988,1989).These studies have yielded conflicting results. On the possibility that fibroblasts and skeletal muscle cells may differ with regard to the importance of glycosylation in activity of Na+-K+ pumps, we examined effects of TM on ouabain-binding by preparations of fibroblasts prepared from chick limb muscle. As shown in Figure 5, which summarizes the results obtained in 3 separate experiments, TM-treatment caused ouabain-bindingby fibroblasts to decrease by over 70% compared with control. DISCUSSION TM has been used in many studies to assess the role of core-glycosylation of protein subunits in a number of membrane receptor and transport systems. It has been reported that the concentration of TM used in this study will block glycosylation by 85-90%, with less than a 20% decrease in protein synthesis (Prives and Olden, 1980; Merlie et al., 1982; Waechter et al., 1983; Deas and Erecinska, 1989). This results in impaired expression of ion channels (Waechter et al., 1983; BarSagi and Prives, 1983; Sumikawa et al., 1988; Zona et al., 1990), ACh-receptors (Prives and Olden, 1980; Merlie et al., 1982; Sumikawa and Miledi, 1989), high-affinity amino acid carriers in C6 glioma cells (Deas and Erecinska, 1989), insulin receptors (Ronnet and Lane, 19811, the hexose carrier in mouse 3T3 cells (Kitagawa et al., 1985), and in maintenance of normal high affinity of growth hormone receptors in r a t adipocytes (Szecowa et al., 1990). In all these cases, it has been concluded that N-glycosylation of the appropriate subunit is required for expression of the functional protein. In this study, we found that TM markedly reduced functional expression of the Naf-K+ pump in cultured skeletal muscle as determined by measurements of the s ecific binding of L3H]-ouabain,the specific uptake of "Rb, and by the level of resting Em and its electrogenic component measured with intracellular microelectrodes. Thus, after 24 h r of incubation in TM a t a concentration of 1 p.g/ml, ouabain binding was reduced to 60% of control, Rb uptake decreased by a mean of nearly 35%, resting Em fell by 10-15%, and E, was reduced by almost 70%. In addition to reducing expression of resting pump activity, TM completely eliminated the increase in Em as a result of Na+-K' pump activation by Na-loading. All these findings indicate that the Na+-K+pump does not function normally after treatment with TM. It is possible that the reduced functional activity of the Na+-K+pump after TM treatment is due to reduced pump expression secondary to decreased intracellular Na+ following effects of this inhibitor of glycosylation
644
ALBOIM ET AL.
B
s E
c,
c
0 0
100 -
-80
v
u-
0 CTRL
0
(3 60
-40
z a I
a
0
TREATMENT
TREATMENT
Fig. 3. Comparison between effects of W X with those of TM on Na'-K +-pumpexpression in cultured skeletal muscle. A: Ouabain binding. Values, representing the mean ? S.E. of measurements on triplicate samples in each of 3 experiments (n = 91, are expressed as % of control values determined from untreated, sister cultures. B: Mem-
brane Potential (mV). The data were obtained from measurements on 9 cells in 2 dishes in each of 3 experiments (n = 54; P < 0.001). Data for both ouabain binding and En,were obtained from cells from the same cultures. Cells were treated with either 2 x 10-7M TTX (back crosshatch) or 1 pgiml TM (solid bars) for 24 hr.
to reduce Na-channel expression (Waechter e t al., 1983; Bar-Sagi and Prives, 1983; Sumikawa et al., 1988;Zona et al., 1990). Studies from this and other laboratories (Brodie and Sampson, 198913; Wolitzky and Fambrough, 1986) have shown that the level of intracellular Na+ plays a major role in regulation of Na+-K+ pump synthesis in skeletal muscle. Thus, we earlier reported in a study of Na+-K+ pump regulation in rat skeletal muscle that reduction of intracellular Naf by treatment with TTX decreased the number of L3H]-ouabain binding sites, ouabain-sensitive "Rb uptake, and the size of the electrogenic pump component of Em (Brodie and Sampson, 1989a) in a manner similar to that reported here with TM. For this reason, we compared effects of TM and TTX on these functional properties. In every case, effects of TM treatment were significantly greater than those of TTX treatment. It should be pointed out that effects of TTX on Em of cultured rat muscle are somewhat different from those of cultured chick myotubes. Thus, in the former Em is reduced significantly by TTX, a n effect entirely attributable to the decrease in E (Brodie and Sampson, 1989a). In contrast neither nor E, was changed significantly by TTX treatment in this study, even though the number of ouabain-binding sites was decreased. This indicates that the reduced number of pump sites in chick muscle was capable of maintaining Em a t control levels. Moreover, these fewer pump sites were capable of responding sufficiently to the challange of Na-loading and cause hyperpolarization of Em to the same level a s control myotubes. This is in contrast to effects of TM which reduced both the number of pump sites and the ability of existing pumps to maintain Em and respond to Na-
loading. Hence, effects of TM on functional Na+-K+ pump activity do not appear to be exclusively secondary to reduced Na-channel expression. TM, in the concentration used in this study, does have a slight inhibitory effect on protein synthesis (Prives and Olden, 1980; Merlie et al., 1982; Waechter et al., 1983; Deas and Erecinska, 1989). To examine the possibility that TM effects on Na'-Kc pump expression might be due to inhibition of protein synthesis, we compared TM with cycloheximide, a well-known inhibitor of protein synthesis, on ouabain binding. TM effects were consistently greater than those of cycloheximide. In addition, TM did not consistently affect the total protein content of the cultures over 24 h r of incubation. Thus, inhibition of protein synthesis can not entirely account for the effects of TM on the various properties measured in this study. Instead, the data indicate that N-glycosylation of the 6-subunit is required for functional expression of the Na+-Kf pump in cultured skeletal muscle. Our findings thus appear to be in marked contrast to some of those reported for the Na+-K+ pump in other preparations. Olden et al. (1979) reported that TM treatment of chick embryo fibroblasts affected neither Na+-K+pump activity nor other membrane-associated processes, such as adenylate cyclase activity and passive glucose uptake. Such treatment did, however, cause defective transport of glucose and amino acids. Similarly, Tamkun and Fambrough (1986), in a study of chick sensory neurons, reported that core glycosylation of the (3-subunit is not important for either intracellular accumulation of the Na+-K+pump or its transport to and assembly in the plasma membrane, or its
k,
TUNICAMYCIN REDUCES Na+-K ' -PUMP EXPRESSION
645
-I
CON CONTROL
TM
m
TREATMENT Fig. 4. Effects of TM and TTX on increase in Emas a result of activation of the Na'-K' pump by Na-loading. Values obtained and expressed as in Figure 3B. On the day of recording (24 h r after treatment), control mean Em t- S.E. was first determined (clear bars). Then, the cells were cooled to 4 4 ° C for 30 min following which they were re-warmed to 37°C. Ten minutes after re-warming, mean Em was again recorded (cross-hatched bars). The bars represent values obtained on 9 cells in 2 dishes in each of 3 experiments (n = 54).
TM
TREATMENT Fig. 5. Effects of TM on specific ["HI-ouabain binding by fibroblasts. Values obtained and expressed as in Figure 1. Cultured fibroblasts were treated for 24 hr with TM (1 Fgiml).
suggested that the difference between their findings (and possibly ours) and those of Tamkun and Fambrough (1986) might be explained on the basis of the relatively high degree of inhibition (60%) of protein synthesis in the latter's study. Zamofing et al. sugested transport function. And, Zamofing et al. (1989) re- that this situation may be ''unfavorable'' to assess efported that p-subunit glycosylation is not important for fects on expression of specific proteins. The precise role of the P-subunit, itself, in functional the hydrolytic or transport properties of the pump in toad urinary bladder cells. These studies indicate non- activity of the Na+-Kf pump is also not clear, nor are glycosylated P-subunits are readily assembled to a-sub- the mechanisms underlying defective expression of the units, and the assembled a@ complex is inserted into Na+-K+ pump in TM-treated cells. Nonetheless, sevthe membrane, where its catalytic functions presum- eral possibilities have been raised by Zamofing et al. ably continue unperturbed. Our findings indicate oth- (1988). They found that nonglycosylated @subunits erwise (i.e., that non-glycosylated $3 complexes do not synthesized in the presence of TM were much more function normally). We have no clear explanation for sensitive to trypsin than core-glycosylated or fully glythe differences between our results on cultured skeletal cosylated forms. Thus, as they stated, acquisition of muscle and those reported for the other systems. One core sugars may be important for correct folding of the possibility may be related to a higher level of pump p-chain o r may protect the protein from cellular proactivity in cultured muscle than in preparations of fi- teases. A number of studies have addressed the quesbroblasts because of the spontaneous electrical and me- tion of P-subunit involvement in Na'-K+ pump regulachanical activity (Brodie and Sampson, 1989a). This tion and have presented evidence that the P-subunit Na-leak may alter pump affinity which may be affected may have several important roles. Sabatini et al. (1981) by TM, as mentioned by Zamofing et al. (1989). This is and Hiatt et al. (1984) suggested that the p-subunit is not likely to be the sole or even main factor, however, involved in stabilization and orientation of the a-subbecause we also found that [3H]-ouabain binding by unit during biogenesis of the Na+-K+pump, a concept fibroblasts was also markedly reduced by TM treat- supported by several recent studies (Zamofing et al., 1988; Ackermann and Geering, 1990; Noguchi et al., ment. Our findings are in good agreement with a n earlier 1990). In recent years evidence has accumulated that study of Zamofing et al. (1988))who reported that TM the P-subunit may actually regulate the number of treatment of toad bladder cells (TBM) caused a decrease Naf-K+ pumps transported to the plasma membrane of newly synthesized a- and p-subunits. Treatment of (i.e., that up-regulation of p-subunits is the key mechaTBM with TM for 21 h r caused a decrease of about 35% nism for regulation of Na+-K+pump abundance) (Mcin total cellular content of a-subunit, and this was asso- Donough et al., 19901.Zamofing e t al. (1988) also found ciated with a corresponding reduction Na+-K+ATPase that the amount of a-subunit, which is not glycosylated, activity. These changes are of the same order of magni- was decreased by TM treatment to a similar extent a s tude we report in this study. Zamofing et al. (1988) the P-subunit. They suggested that efficient cellular
646
ALBOIM ET AL.
accumulation of the a-subunit, the site of ion transport and hydrolytic function, depends on accumulation of the P-subunit. And, Horowitz et al. (1990) reported that P-subunit mRNA decreases and the P-subunit protein virtually not expressed in hypothyroid skeletal muscle. Taormino and Fambrough (1990) found that veratridine up-regulation of Na+-K+pumps in chick skeletal muscle is correlated with up-regulation of P-subunit mRNA. Finally, Lescale-Matys et al. (1990) reported that low K+-induced increases in Naf-Kf pumps in LLC-PKlIC41 cells occurs as a result of a differential increase in p- hut not in a-subunit mRNA. It has, thus, become obvious that the P-subunit of the Na'-K' pump is a primary determinant of its normal and regulated expression. Our data on ouabain binding are consistent with this concept. In addition, the differences in effects of TM and TTX, both of which decrease pump number, may indicate further that post-translational glycosylation of the @-subunit must occur for the ciP subunit complexes to function normally after insertion into the membrane. ACKNOWLEDGMENTS This work was supported by the Ben and Effie Raber Neuroscience Research Fund, the Charles Krown Health Sciences Research Fund, and the Otto Meyerhoff Center. S.R.S. in the incumbent of the Louis Fisher Chair in Cellular Pathology. LITERATURE CITED Ackerman, U., and Geering, K. (1990) Mutual dependence of Na-KATPase 01- and 6-subunits for correct posttranslational processing and intracellular transport. FEBS Lett., 269t105-108. Bar-Sagi, D., and Prives, J . (1983) Tunicamycin inhibits the expression of surface Na ' channels in cultured muscle cells. J. Cell. Physiol., 114.77-81. Brodie, C., and Sampson, S.R. (1985) Contribution of electrogenic sodium-potassium ATPase to resting membrane potential of cultured rat skeletal myotubes. Brain Res., 347t28-35. Brodie, C., and Sampson, S.R. (1987) Nerve growth factor supports growth of rat skeletal myotubes in culture. Brain Res., 435t393397. Brodie, C., and Sampson, S.R. (1989a)Regulation of the sodium-potassium pump in cultured rat skeletal myotubes by intracellular sodium ions. J. Cell. Physiol., 140t131-137. Brodie, C., and Sampson, S.R. (1989b) Characterization of resting membrane potential and its electrogenic pump component in cultured chick myotubes. Int. J Dev. Neurosci. 7t165-172. Brodie, C., Bak, A., Shainberg, A,, and Sampson, S.R. (1987) Role of Na-K ATPase in regulation of resting membrane potential of cultured rat skeletal myotubes. J . Cell. Physiol., 130t191-198. Deas, J., and Erecinska, M. (1989) Effect of tunicamycin, a n inhibitor of protein glycosylation, on the high-affinity transport of acidic amino acid neurotransmitters in C6 gliona cells. Brain Res., 483t84-90. Hiatt, A., McDonough A.A., and Edelman, I.S. (1984) Assembly of the (Na +-K')-adenosine triphosphatase. Post-translational membrane integration of the a subunit. J . Biol. Chem., 259t2629-2635. Horowitz, B., Hensley, C.B., Quintero, M., Azuma, K.K., Putnam, D., and McDonough, A.A. (1990) Differential regulation of Na,K-ATPase a l , a 2 and 6 subunit mRNA and protein levels by thyroid hormone. J. Biol. Chem., 265t14308-14314. Kitagawa, K., Nishima, H., and Iwashima, A. (1985) Effect of tunicamycin on hexose transport in mouse embryo fibroblast Swiss 3T3 cells. Biochim. Biophys. Acta, 821t67-71.
Lescale-Matys,L., Hensley, C.B., Crnkovic-Markovic, R., Putnam, D., and McDonough, A.A. (1990)Low K+ increases Na-K-ATPase abundance in LLC-PK,/CI, cells by differentially increasing p, and not 01, subunit mRNA. J. Biol. Chem., 265t17935-17940. Lowry, O.H., Rosenbrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurement with the folin phenol reagent. J . Biol. Chem., 193t265-2 75. McDonough, A.A., Geering, K., and Farley, R.A. (1990) The sodium pump needs its p subunit. FASEB J., 4:159€L1605. Merlie, J.P., Sebbane, R., Tzartos, S.,and Lindstrom, J. (1982) Inhibition of glycosylation with tunicamycin blocks assembly of newly synthesized acetylcholine receptor subunits in muscle cells. J . Biol. Chem., 257t2694-2701. Noguchi, S., Higashi, K., and Kamamura, M. (1990) A possible role of the p subunit of (Na-K)-ATPasein facilitating correct assembly of the cu-subunitinto the membrane. J . Biol. Chem., 265t15991-15995. Olden, K., Pratt, R.M., Jaworski, C., and Yamada, K.M. (1979) Evidence for role of glycoprotein carbohydrates in membrane transport: Specific inhibition by tunicamycin. Proc Natl. Acad. Sci. U.S.A. 76t791-795. Prives, Y., and Olden, K. (1980) Carbohydrate requirement for expression and stability of acetylcholine receptor on the surface of embryonic muscle cells in culture. Proc. Natl. Acad. Sci. U.S.A. 77,52635267. Ronnett, G.V., and Lane, M.D. (1981) Post-translational glycosylation-induced activation of aglycoinsulin receptor accumulated during tunicamycin treatment. J. Biol. Chem., 256t47044707. Sabatini, D., Colman, D., Sabban, E., Sherman, J., Morimoto, T., Kreibich, G., and Adesnik, M. (1981)Mechanisms for the incorporation of proteins into the plasma membrane. Cold Springs Harb. Symp. Quant. Biol., 46t807-818. Shainberg, A., Yagil, G., and Yaffe, D. (1971)Alterations ofenzymatic activities during muscle differentiation in vitro. Dev. Biol., 25:l-29. Sumikawa, K., and Miledi, R. (1989) Assembly and N-glycosylation of all ACh receptor subunits are required for their efficient insertion into plasma membranes. Mol. Brain Res., 5t183-192. Sumikawa, K., Parker, I., and Miledi, R. (1988) Effect of tunicamycin on the expression of functional brain neurotransmitter receptors and voltage-operated channels in Xenopus oocytes. Mol. Brain Res., 4t191-199. Szecowka, J., Tai, L.-H., and Goodman, H.M. (1990) Effects of tunicamycin on growth hormone binding in rat adipocytes. Endocrinology, 126t1834-1841. Tamkun, M.M., and Fambrough, D.M. (1986) The (Na'-K+)-ATPase of chick sensory neurons. Studies on biosynthesis and intracellular transport. J . Biol. Chem., 261t1009-1019. Taormino, J.P., and Fambrough, D.M. (1990) Pre-translational regulation of the (Na' -K+J-ATPasein response to demand for ion transport in cultured chicken skeletal muscle. J. Biol. Chem., 265t41164123. Vandenburgh, H.H., and Kaufman, S. (1981) Stretch-induced growth of skeletal myotubes correlated with activation of the sodium pump. J . Cell. Physiol., 109t205-214. Waechter, C.J., Schmidt, J.W., and Catterall, W.A. (1983) Glycosylation is required for maintenance of functional sodium channels in neuroblastoma cells. J . Biol. Chem., 2585117-5123. Wolitzky, B.A., and Fambrough, D.M. (1986) Regulation of the (Na+K+)-ATPasein cultured chick skeletal muscle. Regulation of expression by the demand for ion transport. J. Biol. Chem., 261t99909999. Zamofing, D., Rossier, B.C., and Geering, K. (1988)Role of the Na-KATPase P-subunits in the cellular accumulation and maturation of the enzyme as assessed by glycosylation inhibitors. J . Membr. Biol., 104t69-79. Zamofing, D., Rossier, B.C., and Geering, K. (1989) Inhibition of N-glycosylation affects transepithelial Na' but not Na+-K+-ATPase transport. Am. J. Physiol., 256tC95€LC966. Zona, C., Eusebi, F., and Miledi, R. (1990)Glycosylation is required for maintenance of functional voltage-activated channels in growing neocortical neurons of the rat. Proc. R. SOC.Lond., B239t119-127.
Tunicamycin Reduces Na+-K+-Pump Expression in Cultured Skeletal Muscle SANDRA V. ALBOIM, ASIA BAK, AND SANFORD R. SAMPSON* Health Sciences Research Center and the Otto Meyerhoff Center, Department of l i f e Sciences, Bar-/Ian University, Ramat-Can 52900, lsrael The purpose of this study was to examine effects of tunicamycin (TM), which inhibits core glycosylation of the p-subunit, on functional expression of the Na+-K+ pump in primary cultures of embryonic chick skeletal muscle. Measurements were made of specific-['HI-ouabain binding, ouabain-sensitive 86Rb uptake, resting membrane potential (Em), and electrogenic pump contribution to Em (E,) of single myotubes with intracellular microelectrodes. Growth of 4-6-day-old skeletal myotubes in the presence of TM (1 pg/ml) for 21-24 hr reduced the number of Na+-K+ pumps to 60-90% of control. Na+-K+ pump activity, the level of resting Em and E,, were also reduced significantly by TM. In addition, TM completely blocked the hyperpolarization of Em induced in single myotubes by cooling to 10°C and then re-warming to 37°C. Effects of tunicamycin were compared with those of tetrodotoxin (TTX; 2 x l o p 7 M for 24 hr), which blocks voltage-dependent Na+ channels. TM produced significantly greater decreases in ouabain-binding and Em than did TTX, findings that indicate that reduced Na'-K+ pump expression was not exclusively secondary to decreased intracellular Na+, the primary regulator of pump synthesis in cultured muscle. Similarly, effects of TM were significantly greater than those of cycloheximide, which inhibits protein synthesis by 95%. These findings demonstrated that effects were not due to inhibition of protein synthesis. We conclude that glycosylation of the Na+-K+ pump p-subunit is required for full physiological expression of pump activity in skeletal muscle.
A number of membrane proteins require N-glycosylation for adequate cellular expression. Thus, it has been shown that tunicamycin (TM), which inhibits this process, decreases the density of voltage-dependent Nachannels in a number of cell types including cultured neuroblastoma (Waechter et al., 19831, cultured chick skeletal muscle (Bar-Sagi and Prives, 1983), neurons (Zona et al., 1990) and Xenopus oocytes injected with brain mRNA (Sumikawa et al., 1988). Similarly, carbohydrate moieties are required for expression of acetylcholine receptor (AChR) in cultured chick myotubes (Prives and Olden, 1980) and Xenopus oocytes (Sumikawa and Miledi, 1989), glucose transport in chick embryo fibroblasts (Olden et al., 1979), high affinity carriers for amino acid transport (Deas and Erecinska, 1989), insulin receptors (Ronnet and Lane, 1981), and for maintenance of normal high affinity of growth hormone receptors in rat adipocytes (Szecowa et al., 1990). The importance of N-glycosylation in expression of Na+-K+pump activity is not entirely clear. On the one hand, several studies indicate that @-subunitN-glycosylation may not be essential for expression of Na+-K+pump activity. Thus, Olden et al. (1979) reported that TM treatment did not affect activity of this enzyme in chick fibroblasts, and Tamkun and Fambrough (1986), in a n extensive study of this enzyme in chick sensory neurons, concluded that n-glycosylation of the (3-subunit is not important for either intracellular transport, assembly or membrane insertion of Na+-K+-purnpsub0 1992 WILEY-LlSS, INC
units. They reported similar findings for chick skeletal muscle. The role of N-gycosylation in expression of Na+-K+-pumpactivity, however, was not investigated. On the other hand, Zamofing et al. (1988) found that treatment with TM results in a decrease in amount of newly synthesized @- and a-subunit. They suggested further that efficient cellular accumulation of the a-subunit may depend on @-subunitaccumulation. The same group (Zamofing et al. 1989) later reported that while TM reduced epithelial Na+ transport in toad bladder cells, Na+-K+ transport (Na+-K+ ATPase activity) was not altered. Because of the importance of N-glycosylation in expression of functional activity of other membrane proteins in excitable membranes, we have examined the role of this process for resting and stimulated activity of the Naf-Kt-pump in cultured skeletal muscle. The Na+-K+-pumpin this preparation has been characterized as having a single binding site for ouabain and, according to Scatchard analyses, there is only one class of Na-K pump units (Brodie et al., 1987). In addition, the activity of the Na+-K+-pumpcontributes a n important and easily measurable electrogenic component (E,) to resting membrane potential (Ern;Brodie et al.,
Received June 17,1991; accepted October 21,1991. * To whom reprint requestsicorrespondence should be sent.
TUNICAMYCIN REDUCES Na ’ -K ‘ -PUMP EXPRESSION
1987; Brodie and Sampson, 1985, 1989a,b). In this study, we have investigated effects of TM on the amount and activity of Na-K pump sites as assessed by the specific binding of L3H1-ouabainand uptake of “Rb, and by levels of resting Em and E, in primary cultures of skeletal muscle.
EXPERIMENTAL PROCEDURES Preparation of cultures Skeletal muscle cell cultures were prepared by mechanical dissociation of the thigh muscles of 11-12-day chick embryos, a s already described (Brodie and Sampson, 1985). Most fibroblasts were eliminated by preplating, and primary cultures were established by plating a t a density of 3 x lo5 cellsiml (1.5 ml/dish) in 35 mm plastic tissue culture dishes (Nunc) coated with collagen-gelatin. Cultures were grown in a water-saturated atmosphere of 95% air-5% CO, at a temperature of 37°C. The composition of the growth medium was a s follows: Dulbecco’s Minimal Essential Medium (Gibco), 83%; horse serum (Biolab) 15%; chick embryo extract (from 11-12-day old embryos), 2%. Fresh growth medium was replaced every 3 days. Unless otherwise indicated, experiments were done on cultures of age 6-8 days in vitro (DIV). Measurement of r3H]ouabainbinding The binding of [3H]ouabain was determined by a modification (Brodie et al., 1987) of the method of Vandenberg and Kaufman (1981). Cultures were washed three times with 2 ml K-free phosphate buffer (pH 7.4) and then incubated in 1.0 ml ofbuffer containing 75 nM [‘Hlouabain a t 37°C. In preliminary studies, we determined the concentration-binding characteristics of [’Hlouabain by chick myotubes. We found that maximal binding was reached a t a ouabain concentrations of 60-75 nM. Non-specific binding was determined in the presence of 1mM unlabelled ouabain. After incubation for 30 min, cells were washed five times with cold buffer. They were solubilized in 1% Triton x 100, and radioactivity was assayed by standard procedures. Control studies established that ouabain binding by nonmuscle elements was less that 5-10% of the total. Measurement of Na+-K+-pumpactivity 86Rb uptake was measured by a modification of the method of Vandenburgh and Kaufman (1981) as described (Brodie et al., 1987; Brodie and Sampson, 1989a). Rate of u take was measured in 1ml PBS containing 2.5 pCi PRb. Uptake was stopped after 15 min by aspiration of the medium immediately followed by 4 rapid rinses in cold (4°C)PBS. One ml of 1%Triton x 100 was added to each culture dish, and the cells were scraped from the dish and added to 1.0 ml H 0 in a scintillation vial. Samples were counted in a 3H-window of a Tricarb liquid scintillation counter. Uptake specifically related to Na+-K+-pumpactivity was determined by subtraction of the amount taken up in the presence of ouabain (10-3M) from that in its absence. Measurement of Em Intracellular recordings of membrane potentials were obtained by techniques as described in other reports (Brodie et al., 1987, 1989; Brodie and Sampson,
641
1989a,b). For recording, culture dishes were removed from the incubator, and the medium was changed to a phosphate-buffered saline (PBS) solution (pH 7.32) of the following composition (mM): NaCl, 135; KCI, 2.7; Na,HPO,, 1.4; KH,PO,, 1.4; MgCl,, 0.5; CaCl,, 0.7; glucose, 22. The temperature of the culture was controlled at predetermined levels by placing the dish containing the culture in a copper sleeve heated by circulating water from a thermostatically controlled heater, and this assembly was placed on the stage of a n inverted phase-contrast microscope (Nikon). Em was recorded from single myotubes with glass micropipettes filled with 2.8 M KC1 and having internal resistances of 20-50 M a . The microelectrodes were connected to the input of a microelectrode preamplifier (Winston Electronics, San Francisco), and recordings were displayed on one beam of a storage oscilloscope (Tektronix 5113). The verticle signal output from this beam was fed to a pulse height selector-rate meter for measurements of frequency of action potential. Em was measured a s the highest stable (minimum 30 sec) potential recorded during a n impalement. Tip potentials were generally between 1and 3 mV. Earlier studies in our laboratory have shown that ouabain (1 mM) added to chick myotubes in cultures causes a n almost immediate (within 5-10 sec) depolarization of 15-25 mV (Brodie and Sampson, 1989b). Accordingly, the contribution of the Na+-K+ pump to Em was determined a s the difference between the mean in Em recorded within 5-10 min after addition of ouabain (1mM) from that recorded prior to ouabain administration.
Analysis of data For ouabain binding and “Rb uptake studies, measurements were made on triplicate samples of control and treated groups in each individual experiment. For determination of Em, measurements were made from a minimum of 9 cells in each of 2 dishes of control and treated groups in each experiment. The mean values and standard errors were calculated, and data were analyzed statistically by Student’s t-test for unpaired samples or by one-way analysis of variance (ANOVA; Instat, GraphPad, San Diego, CA). Chemicals and drugs Tetrodotoxin (TTX), tunicamycin (TM), and cycloheximide (CH) were obtained from Sigma Chemical Co. (St. Louis, MO). [’Hlouabain (specific activity 16-18 Ci/mM) and “Rb were obtained from Amersham Ltd. (UK). TM was dissolved in absolute ethanol a t a final concentration of 1mg/ml. Treatment of cells On the day of treatment, TM was diluted in DMEM to a concentration of 10 pgiml; 150 pl of this solution was added to the culture dishes which contained 1.5 ml growth medium to give a final TM concentration of 1 pgtml. Myotubes of cultures age 4-6 days were treated with TM for 24 hr. Measurement of total protein content Total protein content of the culture dishes were determined by the method of Lowry et al. (1951).
ALBOIM ET AL.
642 120
1
loo
I 60 -. rcc
0
2R
cultures, effects in 3 experiments ranging from no change to a decrease of 16%. Second, treatment of myotubes with the protein synthesis inhibitor (5 pm) cycloheximide for 24 hr caused less of a n effect on ouabain binding than did TM. Thus, in each of 2 experiments TM decreased the number of ouabain binding sites by 37 and 60%, respectively, while CH in these same experiments decreased the number of binding sites by 23 and 29%, respectively (not illustrated), and the total protein by 27%. This concentration of CH has been shown to decrease protein synthesis by 95% in muscle cell cultures (Shainberg et al., 1971). One way to determine activity of the Na+-K+-pump is by measurement of ouabain-sensitive uptake of 86Rb, which is recognized by the external K+-binding site of the pump. We examined effects of TM on Na+-K+-pump activity in 4 experiments. In each experiment, Rb uptake in TM-treated myotubes was lower than that in control, untreated cells. The TM-induced decrease in ouabain-sensitive "Rb uptake ranged from 41-92% of control in the individual experiments (mean 66.8%; P < 0.02). The contribution of the Na+-K+-pump to resting membrane potential of cultured chick myotubes has been estimated to be of the order of 15-18 mV (Brodie and Sampson, 1989b). In this study, we sought to determine if the decrease in number and activity of the Na'K+-pump obtained by treatment with TM is also expressed by change in the level of Em.This was done by comparing the mean resting Em of cells treated with TM for 24 h r before and after addition of ouabain (1 mM) with that of control, untreated myobutes of the same culture. The results of 3 experiments of this type are summarized in Figure 1. The mean (? S.E.) Em of control myotubes was -69 (? 1.2) mV, while that of TM-treated cells was -61.5 ( 2 0.8) mV (P < 0.001), and the E, was reduced to 31% of that in control myotubes. We considered the possibility that the ability of TM to decrease ouabain binding might be related to a nonselective, general interference with the li and. To examine this, we incubated myotubes in [ HI-ouabain plus various concentrations of TM, from 0.1 to 10 pg/ml, for 30 min. The effects of increasing concentrations of TM are shown in Figure 2, which shows that TM does not have a nonselective or competitive effect on ouabain binding.
1
40 -.
20 -.
0
i
OUAB
Fib
EP
Fig. 1. Effects of TM on expression of Na'-K' pump activity in cultures skeletal muscle as determined by measurements of L3H1-ouabain binding (OUAB), "Rb-uptake (Rb),and electrogenic pump component of resting membrane potential (E ), All values expressed as % of those in control, untreated sister cells. "he data for ouabain binding are the mean (* S.E.) of triplicate values obtained in 7 separate experiments (n = 21; P < 0.001). The data for Rb-uptake were obtained on triplicate samples in each of 4 experiments (n = 12; P < 0.01). The data for Emwere obtained from measurements on 9 cells in 2 dishes in each of 3 experiments (n = 54; P < 0.001).
RESULTS Na+-K+-pumpexpression in cultured skeletal muscle is readily assessed by the measurements of specific [3H]-ouabain binding, ouabain-sensitive "Rb uptake, and the level of transmembrane resting Em recorded with intracellular microelectrodes. The results of studies on the effects of TM on these 3 parameters are summarized in Figure 1. Ouabain is a specific antagonist of the Na+-K+-pump and binds to the external K+-transport site on the enzyme. This ligand has been used to quantitate the number of Na+-K+-pumpsite in a variety of different cells under different conditions. In this study, we compared the number of [3H]-ouabain binding site in TM-treated cells with that in control myotubes from the same culture. The number of ouabain binding sites in control cells ranged from 166 to 243 fmoledmg protein. These values are in the range reported by others for skeletal muscle in culture (Vandenburgh and Kaufman, 1981; Brodie and Sampson, 1989a). In each of 5 experiments, 24-hr incubation of myotubes in M TM resulted in significant decrease (P < 0.01) in the number of ouabain binding sites. In addition to effects on glycosylation, TM in the concentration used in this study has been reported to inhibit protein synthesis. That the ability of TM to impair expression of the Na+-K+-pumpis not likely to be due entirely to a general effect on protein synthesis is indicated by the following. First, TM treatment for 24 h r had only a slight effect on total protein content of the
8
Comparison between effects of TM and TTX on Na+-K+-pumpexpression The results obtained so far demonstrate that treatment of cultured myobutes with TM, which inhibits N-terminal glycosylation, causes a decrease in number of ouabain-binding sites, a decrease in Na-K pump activity, and a decrease in resting Em. These findings suggest that p-subunit glycosylation is, therefore, important for expression of Na+-K+ pump activity. Another explanation, however, may be that TM-treatment causes a decrease in internal Na+ as a result of the decreased expression of voltage-dependent Na-channels, a n effect known to occur in this a s well as other preparations (Bar-Sagi and Prives, 1983). The level of intracellular Na+ has been shown to be a major determinant of Na+-K+ pump density (Wolitzky and Fambrough, 1986; Brodie and Sampson, 1989a). To examine
TUNICAMYCIN REDUCES Na ' -K ' -PUMP EXPRESSION
643
Na-loading cells treated with TTX for 24 h r caused Em to change by the same amount as that obtained in control cells (Fig. 4). This indicates that even though the number of pump sites is reduced slightly by TTX, this fewer number is capable of responding to Na-loading to the same degree as in control.
k
3
E: 1 2 0 1 0
CON
0.1
1
10
TM CONCENTRATION (yg/ml) Fig. 2. Effects of different concentration of TM on ['HI-ouabain binding by cultured myotubes. Values expressed as in Figure 1. Each bar represents the mean f S.E. of triplicate measurements in each of 2 experiments (n = 6).
this possibility, we treated myotubes with tetrodotoxin (TTX), which blocks voltage-dependent Na-channels, thereby decreasing internal Na concentration and down-regulating Na+-K+ pumps (Wolitzky and Fambrough, 1986; Brodie and Sampson, 1989a1, and compared the effects of TTX with those of TM on specific [3H]-ouabain binding and on Em. The comparison between TM and TTX on ouabain binding and Em is shown in Figure 3. Treatment with TTX for 24 h r caused a 24% decrease in ouabain binding compared with that in control, sister cells (Fig. 3A). This effect was not statistically significant (P > 0.05) (see, however, Brodie and Sampson, 1989a). In contrast, TM caused a decrease in specific ouabain binding of 48% (P < 0.001) compared with control myotubes. The relatively large standard error can be attributed to the fact that in one of the 4 experiments of this type, TM reduced the number of ouabain binding sites by 74%, to 26% of control. Similarly, TM caused a greater decrease in Em than did TTX following treatment for 24 h r (Fig. 3B). Thus, Emwas reduced from a mean ( * S.E.) control value of -77.6 mV (+ 1.2) to -72.4 mV (+ 2.3) by treatment with TTX (P > 0.051, but to -61.3 mV (k 0.9) by treatment with TM (P < 0.001). In another series of experiments, we examined effects of TM and TTX on Na+-K+ pump activation by Naloading. This was accomplished by first cooling the cells to 10°C for 30 min and then rewarming them to 37°C. Em was recorded 10-20 min after the cells were rewarmed. In control cells, Na-loading caused Em to increase significantly (P < 0.0001) from -67.7 ( + 4.5) to -74.5 (+ 4.7), whereas in TM-treated cells, this procedure had no effect on Em (before cooling, -57.9 + 3.9 mV; rewarmed, -58.2 ? 4.5 mV; P > 0.4). In contrast,
Effects of TM on [3Hl-ouabainbinding in fibroblasts Other studies regarding inhibition of n-glycosylation by TM on Na+-K' pump expression were done on fibroblasts derived from various sources (Olden et al., 1979; Zamofing et al., 1988,1989).These studies have yielded conflicting results. On the possibility that fibroblasts and skeletal muscle cells may differ with regard to the importance of glycosylation in activity of Na+-K+ pumps, we examined effects of TM on ouabain-binding by preparations of fibroblasts prepared from chick limb muscle. As shown in Figure 5, which summarizes the results obtained in 3 separate experiments, TM-treatment caused ouabain-bindingby fibroblasts to decrease by over 70% compared with control. DISCUSSION TM has been used in many studies to assess the role of core-glycosylation of protein subunits in a number of membrane receptor and transport systems. It has been reported that the concentration of TM used in this study will block glycosylation by 85-90%, with less than a 20% decrease in protein synthesis (Prives and Olden, 1980; Merlie et al., 1982; Waechter et al., 1983; Deas and Erecinska, 1989). This results in impaired expression of ion channels (Waechter et al., 1983; BarSagi and Prives, 1983; Sumikawa et al., 1988; Zona et al., 1990), ACh-receptors (Prives and Olden, 1980; Merlie et al., 1982; Sumikawa and Miledi, 1989), high-affinity amino acid carriers in C6 glioma cells (Deas and Erecinska, 1989), insulin receptors (Ronnet and Lane, 19811, the hexose carrier in mouse 3T3 cells (Kitagawa et al., 1985), and in maintenance of normal high affinity of growth hormone receptors in r a t adipocytes (Szecowa et al., 1990). In all these cases, it has been concluded that N-glycosylation of the appropriate subunit is required for expression of the functional protein. In this study, we found that TM markedly reduced functional expression of the Naf-K+ pump in cultured skeletal muscle as determined by measurements of the s ecific binding of L3H]-ouabain,the specific uptake of "Rb, and by the level of resting Em and its electrogenic component measured with intracellular microelectrodes. Thus, after 24 h r of incubation in TM a t a concentration of 1 p.g/ml, ouabain binding was reduced to 60% of control, Rb uptake decreased by a mean of nearly 35%, resting Em fell by 10-15%, and E, was reduced by almost 70%. In addition to reducing expression of resting pump activity, TM completely eliminated the increase in Em as a result of Na+-K' pump activation by Na-loading. All these findings indicate that the Na+-K+pump does not function normally after treatment with TM. It is possible that the reduced functional activity of the Na+-K+pump after TM treatment is due to reduced pump expression secondary to decreased intracellular Na+ following effects of this inhibitor of glycosylation
644
ALBOIM ET AL.
B
s E
c,
c
0 0
100 -
-80
v
u-
0 CTRL
0
(3 60
-40
z a I
a
0
TREATMENT
TREATMENT
Fig. 3. Comparison between effects of W X with those of TM on Na'-K +-pumpexpression in cultured skeletal muscle. A: Ouabain binding. Values, representing the mean ? S.E. of measurements on triplicate samples in each of 3 experiments (n = 91, are expressed as % of control values determined from untreated, sister cultures. B: Mem-
brane Potential (mV). The data were obtained from measurements on 9 cells in 2 dishes in each of 3 experiments (n = 54; P < 0.001). Data for both ouabain binding and En,were obtained from cells from the same cultures. Cells were treated with either 2 x 10-7M TTX (back crosshatch) or 1 pgiml TM (solid bars) for 24 hr.
to reduce Na-channel expression (Waechter e t al., 1983; Bar-Sagi and Prives, 1983; Sumikawa et al., 1988;Zona et al., 1990). Studies from this and other laboratories (Brodie and Sampson, 198913; Wolitzky and Fambrough, 1986) have shown that the level of intracellular Na+ plays a major role in regulation of Na+-K+ pump synthesis in skeletal muscle. Thus, we earlier reported in a study of Na+-K+ pump regulation in rat skeletal muscle that reduction of intracellular Naf by treatment with TTX decreased the number of L3H]-ouabain binding sites, ouabain-sensitive "Rb uptake, and the size of the electrogenic pump component of Em (Brodie and Sampson, 1989a) in a manner similar to that reported here with TM. For this reason, we compared effects of TM and TTX on these functional properties. In every case, effects of TM treatment were significantly greater than those of TTX treatment. It should be pointed out that effects of TTX on Em of cultured rat muscle are somewhat different from those of cultured chick myotubes. Thus, in the former Em is reduced significantly by TTX, a n effect entirely attributable to the decrease in E (Brodie and Sampson, 1989a). In contrast neither nor E, was changed significantly by TTX treatment in this study, even though the number of ouabain-binding sites was decreased. This indicates that the reduced number of pump sites in chick muscle was capable of maintaining Em a t control levels. Moreover, these fewer pump sites were capable of responding sufficiently to the challange of Na-loading and cause hyperpolarization of Em to the same level a s control myotubes. This is in contrast to effects of TM which reduced both the number of pump sites and the ability of existing pumps to maintain Em and respond to Na-
loading. Hence, effects of TM on functional Na+-K+ pump activity do not appear to be exclusively secondary to reduced Na-channel expression. TM, in the concentration used in this study, does have a slight inhibitory effect on protein synthesis (Prives and Olden, 1980; Merlie et al., 1982; Waechter et al., 1983; Deas and Erecinska, 1989). To examine the possibility that TM effects on Na'-Kc pump expression might be due to inhibition of protein synthesis, we compared TM with cycloheximide, a well-known inhibitor of protein synthesis, on ouabain binding. TM effects were consistently greater than those of cycloheximide. In addition, TM did not consistently affect the total protein content of the cultures over 24 h r of incubation. Thus, inhibition of protein synthesis can not entirely account for the effects of TM on the various properties measured in this study. Instead, the data indicate that N-glycosylation of the 6-subunit is required for functional expression of the Na+-Kf pump in cultured skeletal muscle. Our findings thus appear to be in marked contrast to some of those reported for the Na+-K+ pump in other preparations. Olden et al. (1979) reported that TM treatment of chick embryo fibroblasts affected neither Na+-K+pump activity nor other membrane-associated processes, such as adenylate cyclase activity and passive glucose uptake. Such treatment did, however, cause defective transport of glucose and amino acids. Similarly, Tamkun and Fambrough (1986), in a study of chick sensory neurons, reported that core glycosylation of the (3-subunit is not important for either intracellular accumulation of the Na+-K+pump or its transport to and assembly in the plasma membrane, or its
k,
TUNICAMYCIN REDUCES Na+-K ' -PUMP EXPRESSION
645
-I
CON CONTROL
TM
m
TREATMENT Fig. 4. Effects of TM and TTX on increase in Emas a result of activation of the Na'-K' pump by Na-loading. Values obtained and expressed as in Figure 3B. On the day of recording (24 h r after treatment), control mean Em t- S.E. was first determined (clear bars). Then, the cells were cooled to 4 4 ° C for 30 min following which they were re-warmed to 37°C. Ten minutes after re-warming, mean Em was again recorded (cross-hatched bars). The bars represent values obtained on 9 cells in 2 dishes in each of 3 experiments (n = 54).
TM
TREATMENT Fig. 5. Effects of TM on specific ["HI-ouabain binding by fibroblasts. Values obtained and expressed as in Figure 1. Cultured fibroblasts were treated for 24 hr with TM (1 Fgiml).
suggested that the difference between their findings (and possibly ours) and those of Tamkun and Fambrough (1986) might be explained on the basis of the relatively high degree of inhibition (60%) of protein synthesis in the latter's study. Zamofing et al. sugested transport function. And, Zamofing et al. (1989) re- that this situation may be ''unfavorable'' to assess efported that p-subunit glycosylation is not important for fects on expression of specific proteins. The precise role of the P-subunit, itself, in functional the hydrolytic or transport properties of the pump in toad urinary bladder cells. These studies indicate non- activity of the Na+-Kf pump is also not clear, nor are glycosylated P-subunits are readily assembled to a-sub- the mechanisms underlying defective expression of the units, and the assembled a@ complex is inserted into Na+-K+ pump in TM-treated cells. Nonetheless, sevthe membrane, where its catalytic functions presum- eral possibilities have been raised by Zamofing et al. ably continue unperturbed. Our findings indicate oth- (1988). They found that nonglycosylated @subunits erwise (i.e., that non-glycosylated $3 complexes do not synthesized in the presence of TM were much more function normally). We have no clear explanation for sensitive to trypsin than core-glycosylated or fully glythe differences between our results on cultured skeletal cosylated forms. Thus, as they stated, acquisition of muscle and those reported for the other systems. One core sugars may be important for correct folding of the possibility may be related to a higher level of pump p-chain o r may protect the protein from cellular proactivity in cultured muscle than in preparations of fi- teases. A number of studies have addressed the quesbroblasts because of the spontaneous electrical and me- tion of P-subunit involvement in Na'-K+ pump regulachanical activity (Brodie and Sampson, 1989a). This tion and have presented evidence that the P-subunit Na-leak may alter pump affinity which may be affected may have several important roles. Sabatini et al. (1981) by TM, as mentioned by Zamofing et al. (1989). This is and Hiatt et al. (1984) suggested that the p-subunit is not likely to be the sole or even main factor, however, involved in stabilization and orientation of the a-subbecause we also found that [3H]-ouabain binding by unit during biogenesis of the Na+-K+pump, a concept fibroblasts was also markedly reduced by TM treat- supported by several recent studies (Zamofing et al., 1988; Ackermann and Geering, 1990; Noguchi et al., ment. Our findings are in good agreement with a n earlier 1990). In recent years evidence has accumulated that study of Zamofing et al. (1988))who reported that TM the P-subunit may actually regulate the number of treatment of toad bladder cells (TBM) caused a decrease Naf-K+ pumps transported to the plasma membrane of newly synthesized a- and p-subunits. Treatment of (i.e., that up-regulation of p-subunits is the key mechaTBM with TM for 21 h r caused a decrease of about 35% nism for regulation of Na+-K+pump abundance) (Mcin total cellular content of a-subunit, and this was asso- Donough et al., 19901.Zamofing e t al. (1988) also found ciated with a corresponding reduction Na+-K+ATPase that the amount of a-subunit, which is not glycosylated, activity. These changes are of the same order of magni- was decreased by TM treatment to a similar extent a s tude we report in this study. Zamofing et al. (1988) the P-subunit. They suggested that efficient cellular
646
ALBOIM ET AL.
accumulation of the a-subunit, the site of ion transport and hydrolytic function, depends on accumulation of the P-subunit. And, Horowitz et al. (1990) reported that P-subunit mRNA decreases and the P-subunit protein virtually not expressed in hypothyroid skeletal muscle. Taormino and Fambrough (1990) found that veratridine up-regulation of Na+-K+pumps in chick skeletal muscle is correlated with up-regulation of P-subunit mRNA. Finally, Lescale-Matys et al. (1990) reported that low K+-induced increases in Naf-Kf pumps in LLC-PKlIC41 cells occurs as a result of a differential increase in p- hut not in a-subunit mRNA. It has, thus, become obvious that the P-subunit of the Na'-K' pump is a primary determinant of its normal and regulated expression. Our data on ouabain binding are consistent with this concept. In addition, the differences in effects of TM and TTX, both of which decrease pump number, may indicate further that post-translational glycosylation of the @-subunit must occur for the ciP subunit complexes to function normally after insertion into the membrane. ACKNOWLEDGMENTS This work was supported by the Ben and Effie Raber Neuroscience Research Fund, the Charles Krown Health Sciences Research Fund, and the Otto Meyerhoff Center. S.R.S. in the incumbent of the Louis Fisher Chair in Cellular Pathology. LITERATURE CITED Ackerman, U., and Geering, K. (1990) Mutual dependence of Na-KATPase 01- and 6-subunits for correct posttranslational processing and intracellular transport. FEBS Lett., 269t105-108. Bar-Sagi, D., and Prives, J . (1983) Tunicamycin inhibits the expression of surface Na ' channels in cultured muscle cells. J. Cell. Physiol., 114.77-81. Brodie, C., and Sampson, S.R. (1985) Contribution of electrogenic sodium-potassium ATPase to resting membrane potential of cultured rat skeletal myotubes. Brain Res., 347t28-35. Brodie, C., and Sampson, S.R. (1987) Nerve growth factor supports growth of rat skeletal myotubes in culture. Brain Res., 435t393397. Brodie, C., and Sampson, S.R. (1989a)Regulation of the sodium-potassium pump in cultured rat skeletal myotubes by intracellular sodium ions. J. Cell. Physiol., 140t131-137. Brodie, C., and Sampson, S.R. (1989b) Characterization of resting membrane potential and its electrogenic pump component in cultured chick myotubes. Int. J Dev. Neurosci. 7t165-172. Brodie, C., Bak, A., Shainberg, A,, and Sampson, S.R. (1987) Role of Na-K ATPase in regulation of resting membrane potential of cultured rat skeletal myotubes. J . Cell. Physiol., 130t191-198. Deas, J., and Erecinska, M. (1989) Effect of tunicamycin, a n inhibitor of protein glycosylation, on the high-affinity transport of acidic amino acid neurotransmitters in C6 gliona cells. Brain Res., 483t84-90. Hiatt, A., McDonough A.A., and Edelman, I.S. (1984) Assembly of the (Na +-K')-adenosine triphosphatase. Post-translational membrane integration of the a subunit. J . Biol. Chem., 259t2629-2635. Horowitz, B., Hensley, C.B., Quintero, M., Azuma, K.K., Putnam, D., and McDonough, A.A. (1990) Differential regulation of Na,K-ATPase a l , a 2 and 6 subunit mRNA and protein levels by thyroid hormone. J. Biol. Chem., 265t14308-14314. Kitagawa, K., Nishima, H., and Iwashima, A. (1985) Effect of tunicamycin on hexose transport in mouse embryo fibroblast Swiss 3T3 cells. Biochim. Biophys. Acta, 821t67-71.
Lescale-Matys,L., Hensley, C.B., Crnkovic-Markovic, R., Putnam, D., and McDonough, A.A. (1990)Low K+ increases Na-K-ATPase abundance in LLC-PK,/CI, cells by differentially increasing p, and not 01, subunit mRNA. J. Biol. Chem., 265t17935-17940. Lowry, O.H., Rosenbrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurement with the folin phenol reagent. J . Biol. Chem., 193t265-2 75. McDonough, A.A., Geering, K., and Farley, R.A. (1990) The sodium pump needs its p subunit. FASEB J., 4:159€L1605. Merlie, J.P., Sebbane, R., Tzartos, S.,and Lindstrom, J. (1982) Inhibition of glycosylation with tunicamycin blocks assembly of newly synthesized acetylcholine receptor subunits in muscle cells. J . Biol. Chem., 257t2694-2701. Noguchi, S., Higashi, K., and Kamamura, M. (1990) A possible role of the p subunit of (Na-K)-ATPasein facilitating correct assembly of the cu-subunitinto the membrane. J . Biol. Chem., 265t15991-15995. Olden, K., Pratt, R.M., Jaworski, C., and Yamada, K.M. (1979) Evidence for role of glycoprotein carbohydrates in membrane transport: Specific inhibition by tunicamycin. Proc Natl. Acad. Sci. U.S.A. 76t791-795. Prives, Y., and Olden, K. (1980) Carbohydrate requirement for expression and stability of acetylcholine receptor on the surface of embryonic muscle cells in culture. Proc. Natl. Acad. Sci. U.S.A. 77,52635267. Ronnett, G.V., and Lane, M.D. (1981) Post-translational glycosylation-induced activation of aglycoinsulin receptor accumulated during tunicamycin treatment. J. Biol. Chem., 256t47044707. Sabatini, D., Colman, D., Sabban, E., Sherman, J., Morimoto, T., Kreibich, G., and Adesnik, M. (1981)Mechanisms for the incorporation of proteins into the plasma membrane. Cold Springs Harb. Symp. Quant. Biol., 46t807-818. Shainberg, A., Yagil, G., and Yaffe, D. (1971)Alterations ofenzymatic activities during muscle differentiation in vitro. Dev. Biol., 25:l-29. Sumikawa, K., and Miledi, R. (1989) Assembly and N-glycosylation of all ACh receptor subunits are required for their efficient insertion into plasma membranes. Mol. Brain Res., 5t183-192. Sumikawa, K., Parker, I., and Miledi, R. (1988) Effect of tunicamycin on the expression of functional brain neurotransmitter receptors and voltage-operated channels in Xenopus oocytes. Mol. Brain Res., 4t191-199. Szecowka, J., Tai, L.-H., and Goodman, H.M. (1990) Effects of tunicamycin on growth hormone binding in rat adipocytes. Endocrinology, 126t1834-1841. Tamkun, M.M., and Fambrough, D.M. (1986) The (Na'-K+)-ATPase of chick sensory neurons. Studies on biosynthesis and intracellular transport. J . Biol. Chem., 261t1009-1019. Taormino, J.P., and Fambrough, D.M. (1990) Pre-translational regulation of the (Na' -K+J-ATPasein response to demand for ion transport in cultured chicken skeletal muscle. J. Biol. Chem., 265t41164123. Vandenburgh, H.H., and Kaufman, S. (1981) Stretch-induced growth of skeletal myotubes correlated with activation of the sodium pump. J . Cell. Physiol., 109t205-214. Waechter, C.J., Schmidt, J.W., and Catterall, W.A. (1983) Glycosylation is required for maintenance of functional sodium channels in neuroblastoma cells. J . Biol. Chem., 2585117-5123. Wolitzky, B.A., and Fambrough, D.M. (1986) Regulation of the (Na+K+)-ATPasein cultured chick skeletal muscle. Regulation of expression by the demand for ion transport. J. Biol. Chem., 261t99909999. Zamofing, D., Rossier, B.C., and Geering, K. (1988)Role of the Na-KATPase P-subunits in the cellular accumulation and maturation of the enzyme as assessed by glycosylation inhibitors. J . Membr. Biol., 104t69-79. Zamofing, D., Rossier, B.C., and Geering, K. (1989) Inhibition of N-glycosylation affects transepithelial Na' but not Na+-K+-ATPase transport. Am. J. Physiol., 256tC95€LC966. Zona, C., Eusebi, F., and Miledi, R. (1990)Glycosylation is required for maintenance of functional voltage-activated channels in growing neocortical neurons of the rat. Proc. R. SOC.Lond., B239t119-127.
