- Email: [email protected]
Ca-ATPase activity in bovine parathyroid cells
Ca-ATPase
Activity
Bess F. Dawson-Hughes,
in Bovine Parathyroid
Richard H. Underwood,
and Edward
Cells
M. Brown
Ca-ATPase is thought to function as a calcium extrusion pump that may regulate cytosolic calcium concentration. Because the parathyroid gland is among the few tissues that are directly regulated by extracellular calcium and because cytosolic calcium may be a mediator of the effects of extracellular calcium on parathyroid hormone secretion, we have investigated the presence of this enzyme in homogenates of parathyroid cells. High performance liquid chromatography (HPLC) was used to quantify the formation of ADP from ATP following incubation of ATP with cellular homogenate in a buffer containing ethylenedioxy- (diethylenedinitrilo) tetra acetic acid (EGTA), ouabain, and calcium. Enzyme activity was calcium-dependent, with Ca-ATPase showing two Km (Ca) values, 31 and 853 nM. High affinity Ca-ATPase activity was reduced by the calmodulin inhibitor, trifluoperazine (TFP), with half-maximal inhibition occurring at 7 x lo-’ M. Monovalent cations stimulated high affinity Ca-ATPase activity (K+ > Na+ > Rb+ z Lif) in the presence of calcium. Magnesium (0.8 mM) also stimulated cleavage of ATP. Sodium increased Ca-dependent ATPase activity by 82% but had no significant effect on Mg-stimulated activity. Furthermore, azide, an inhibitor of mitochondrial ATPase(s1, had a significantly greater inhibitory effect on Mg-dependent than on Ca-dependent activity. In summary, a high affinity Ca-ATPase is present in bovine parathyroid cells which has a Km in the range of the cytosolic calcium concentration that is found in other cells. Ca-ATPaselsf may be of importance in regulating the cytosolic calcium concentration and, therefore, hormonal secretion in this cell type.
HERE is a large gradient for calcium across the plasma membrane, with the extracellular free calcium concentration ( 1O-3 M) being about four orders of magnitude higher than that in the cytoplasm (lo-’ mo1).‘S2Although the mechanisms by which this gradient is maintained are not fully understood, several have been identified. Sodium-calcium exchange is a process that couples calcium extrusion to the inward movement of sodium down its electrochemical gradient.3 ATP indirectly supplies the energy for this process through maintenance, by the (Na+-K+)ATPase, of the transmembrane gradient for sodium.3 In addition, calcium-dependent ATPases (CaATPase) have been shown to participate directly in the regulation of cytosolic calcium concentration in the intact squid giant axon4 and in the red blood ce11.2This enzyme has been found in plasma membrane, mitochondria and microsomes of adipocytes,’ pancreatic islets,6 muscle,’ and nerve cells,8 where indirect evidence suggests that it extrudes calcium from the cytosol, both across the plasma membrane and into intracellular compartments.
T
From the Endocrine-Hypertension Unit, Brigham and Women’s Hospital, and Department of Medicine, Harvard Medical School, Boston, Mass. Received for publication February 23, I983. This work was supported by grants from the National Institutes of Health (1 F32 AM06759-01. 2 T32 HLO7236-05. and 1 R01 AM30028-01). Dr. Brown is recipient of NIH-PHS Research Career Development Award, 1K04-AA400627-04. This work was presented in part at the American Federation for Clinical Research Eastern Section. Clinical Research 29.6814. 1981. Address reprint requests to Dr. Bess Dawson-Hughes. EndocrineHypertension Unit, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. Q 1983 by Grune & Stratton, Inc. 0026-0495/83/3209J.IOO3$01.00/0 a74
The parathyroid gland is among the few tissues that are directly regulated by the extracellular calcium concentration. Indirect evidence suggests that cytosolic calcium may be a mediator of the effects of extracellular calcium on parathyroid hormone (PTH) secretion.“” Because Ca-ATPase affects the transmembrane calcium gradient in other cells and because this gradient may play a role in PTH secretion, we investigated the presence of Ca-ATPase(s) in parathyroid cells. In this paper we report several of the properties of Ca-ATPase(s) in dispersed bovine parathyroid cells. MATERIALS
AND
METHODS
Isolated
bovine parathyroid cells were prepared by the method of Brown et al” which entailed mincing and digestion of parathyroid tissue with collagenase and DNAse. After several washings, dispersed cell preparations contained approximately 95% parathyroid cells and 5% erythrocytes. The dispersed parathyroid cells were homogenized by using ten strokes with a manual homogenizer in a solution of 0.25 moles sucrose, 1 mmole EGTA, and 15 mM tris buffer (pH 7.5) at 4 OC and a cell concentration of 10 million/ml. The homogenate was then sedimented for 15 minutes at 27,000 x g at 4 OC. The pellet was resuspended in 20 mmoles tris buffer (pH 7.5) and used directly in the assay. Ca-ATPase activity was quantified by measuring the amount of ADP that is released from ATP. The assay mixture in a final volume of 0.1 ml contained 1 mmole of ATP, 1 mmole of EGTA, 0.1 mmole of ouabain [to inhibit (Na+-K’)-ATPase activity], 20 mM tris buffer (pH 7.5), variable amounts of CaCl,, no added magnesium, and homogenate equivalent to 200,000 parathyroid cells per tube. Assays were carried out for 30 minutes at 37 OC and terminated by placing samples in a boiling water bath for two minutes. Samples were then centrifuged at 10,000 x g for two minutes to remove cellular debris and 15 ~1of supernatant was analyzed by HPLC. The HPLC system contained the following components: Model 5040 Terniary Liquid Chromatograph, Vista 401 Chromatography Data System, Varichrom UV-VIS Spectrophotometric Detector (Varian Instrument Division), Supelcosil LC-18-DB column 15 x 4.6 mmole ID (Supelco, Inc.), and a Guard Column packed with Metabolism, Vol. 32, No. 9 Eeptember), 1993
875
CA-ATPASE ACTIVITY IN BOVINE PARATHYROID CELLS
3ondapak C,,/Corasil (Waters Associates, Inc.). The solvent system was 0.10 M KH,PO, containing 0.025 mole tetra-nbutyl ammonium hydrogen sulfate (pH 7.2) and 13% methanol,” and the flow rate was 2 mls/min. Absorbance was read at 259 rnp and areas under the nucleotide peaks were integrated. Retention times for ATP, ADP, and AMP were established for each experiment by direct comparison with a standard mixture of these nucleotides. In some experiments, the cleavage of ATP was assessed by quantifying the release of 32P from [y-32P]ATP according to the method of Seals et al.13 Results obtained with this method were comparable to those with the HPLC method already described. Because of reduced cost and radiation exposure and the capability of monitoring all the nucleotides, the HPLC method was used in most studies. All samples were assayed in triplicate. The protein concentration in the 27,000 x g pellet was equivalent to about 1 mg/2.5 x 10’ cells.‘4 Although the intraassay coefficient of variation for cleavage is 3% to 4%, interassay variation is greater, probably at least in part due to errors in estimating cell count, upon which the membrane concentration is based. The concentration of free calcium in CaEGTA buffers was calculated according to the equations of Perrin and Sayce.” Results are reported as means k SEM. Statistical probability was evaluated by using students’ t-test as outlined by Bancroft.16 For calculation of threshold concentration (that concentration of additive that causes a significant change from control), we used analysis of variance, obtaining P values from Dunnett’s tables.” ly-“P]ATP was obtained from Amersham/Searle, Arlington Heights, Ill, tetrabutylammonium hydrogen sulfate from Aldrich Chemical Company, Inc., ATP and ADP from Sigma Chemical Company, and AMP from Eastman. RESULTS
In order to quantify enzymatic cleavage by the use of HPLC, the stability of ATP and ADP in the assay were first assessed. When ATP was incubated without
enzyme for 30 minutes at 37 “C, 96.6% appeared on HPLC as ATP, 2.0% as ADP, and 3.3% as AMP. At 1.2 x lo-’ M calcium, there was no change in spontaneous conversion of ATP to ADP or AMP. When ADP was incubated without enzyme, only 5% was spontaneously cleaved to AMP, and thus 95% of the AMP arose directly from ATP. In an additional experiment, calcium caused a slight decrease in conversion of ADP to AMP (from 6% to 4.4%). Since this results in less than a 2% change in apparent conversion of ATP to ADP, this would not significantly affect the subsequent results. When ATP was incubated with homogenate equivalent to 200,000 cells/tube with or without 1.2 x IO-‘/mol free calcium (Fig. l), addition of calcium increased the proportion of ADP from 1.3 to 7.396, whereas the AMP remained relatively constant at about 9%. To determine whether ADP is stable in the presence of enzyme, additional experiments were carried out (not shown) in which standard amounts of enzyme were incubated with ADP. This resulted in conversion of less than 10% of the ADP to AMP. Therefore, less than 10% of the total AMP that is produced arises from ADP, indicating that the area under the ADP peak is a reasonable estimate of the total ADP that is formed. AMP presumably arises directly from ATP by pyrophosphorylation. Subsequent results are expressed as a percent of total nucleotide present as ADP [ADP/(ATP + ADP + AMP) x 1001 as an estimate of the formation of ADP from ATP.
USE OF HPLC TO DEMONSTRATE Co-DEPENDENT ATPose ACTIVITY IN HOMOGENATE OF BOVINE RWXIWROID CELLS -
CALCIUM
+ CALCIUM
c
AMP Fig. 1. HPLC separation of nucleotides from assays done in the absence and presence of 1.2 x lo-’ M free calcium. The assay mixture also contained 1 mM ATP, 1 mM EGTA. 0.1 mM ouabain. 20 mM tris buffer (pH 7.5). and homogenate equivalent to 200,000 bovine parethyroid cells in a final volume of 0.1 ml. Assays were carried out at 37 “C for 30 minutes.
I
I
,
I
1
0
2
I
,
I
,
I
4
6
6
0
2
4
6
8
MIN
MIN
876
DAWSON-HUGHES
ET AL
6V 4-
I OLYIO-*
I lo*
I lo-’
FREE [Co++],
I
I
ItI)
Icr’
M
I O
I 0
I
I
I
I
40
80
120
160
v/s
Effect of Caf+ on Ca-ATPase activity in homogenate of bovine parathyroid cells. IA) Results of a single representative Fig. 2. experiment done in triplicate by using the method of Seals to quantify release of =P from [y-PP]ATP. The blank, representing spontaneous hydrolysis of P from ATP, was 0.64% + 0.07 (SEMI and has been subtracted. (B) Hofstee plot of the data from Fig. 2(A) in which V = nmoles Plhrltube and v/s = pmol P/hrltube/[ca+‘]. Two distinct slopes, -8 (r = 0.91) and -738 (r = -0.921, are apparent, indicating the presence of high and low affinity Ca-ATPase activities with KM ICa) values of 8 nM and 738 n&l, respectively. The maan Km (Cal values of three experiments, each done in triplicate are 31 + 12 nM and 883 r 114 nM.
Calcium-dependent activity was determined by subtracting the percent of total nucleotide present as ADP with EGTA alone from that obtained with chelator plus calcium. Similarly, magnesium-stimulated activity was that increment that occurred with the addition of magnesium to the incubate containing EGTA. The activity of Ca-ATPase was linear with time for up to 60 minutes and was directly proportional to the conCo -ATPose
T-
Mg - ATPose
T
ant
MAzkk
mmm
m ‘omm
Fig. 3. Effect of sodium and aside on Ca- and Mg-ATPase activities. The effect of sodium on Ca- and Mg-ATPase activity was determined by measuring cleavage in the presence of either 1.2 x lo-’ M calculated free calcium or 8 x 1Om4M added magnesium, respectively, with and without (control) the addition of 48 mM NaCI. The effects of azide were determined similarly. However. since sodium azide was used, an amount of sodium equivalent to that in the aside preparation was included in the control. Although aside inhibited both Mg- and Ca-dependent activity, the degree of inhibition of Mg-dependent activity was significantly greater (P < 0.001). Sodium significantly enhanced Ca-ATPase activity (P < 0.001) but had no significant effect on Mg-ATPase activity. Each bar represents the mean of three experiments, each done in triplicate. Control bars represent 8.7 r 1.6% of ADP formed for Ca and 31 + 4% for Mg.
centration of homogenate added. The rate of cleavage of ATP was sixfold slower at 4 “C and ATPase activity was destroyed by boiling. The dispersed parathyroid cells that were used in the assay contain 5% to 10% red blood cells.” Since red blood cells contain Ca-ATPase,* we assessed their contribution to ATP cleavage in our assay. When the predicted number of bovine red blood cells were prepared and assayed as above in the presence of 1.2 x lo-’ M free calcium, there was only 0.23% calciumdependent cleavage of ATP, which is negligible. In the presence of ten times the number of red blood cells and a higher free calcium concentration (6 x lo-’ M), only 0.7% of ATP was cleaved. Thus, the presence of red blood cells in the parathyroid cell preparation had a negligible influence on the results. A calcium activation curve for Ca-ATPase in homogenates of bovine parathyroid cells is shown in Fig. 2(A). With increasing calcium concentrations, there was a progressive increase in ATPase activity, which was biphasic with respect to calcium concentration, with near maximal activity at 10m4 M free calcium. Figure 2(B) shows a Hofstee plot of the data in Fig. 2(A). Two distinct slopes are apparent, indicating the presence of high and low affinity Ca-ATPase activities. In this experiment, their Km(Ca) values are 8 nM and 738 nmol, respectively. The mean Km(Ca) values
*Although the Km(Ca) for the high affinity activity in Fig. 2 (8 nmoles) is lower than the values obtained by HPLC (average 42 nmoles), in a total of three such experiments that were carried out by using the r*P method, the km (Ca) for the high affinity site averaged 25 nmoles. Thus, the two methods appear to yield comparable results.
CA-ATPASE ACTIVITY IN BOVINE PARATHYROID CELLS
877
of three experiments (two using HPLC, one using 32P), each done in triplicate, were 31 + 12 nM and 853 -r114 nmol.* Magnesium also activated ATPase in this assay. To determine whether calcium and magnesium stimulate the same or different enzymes, we studied the effect of several agents on cleavage in the presence of either calcium or magnesium (Fig. 3). When 40 nmol of sodium was added to the incubation mixture containing 1.2 x lo-’ M free calcium, enzyme activity significantly increased by 82% (P < 0.001); whereas in the presence of 8 x 10m4 mol magnesium, which gave nearly maximal stimulation, activity increased by only 1l%, a value not significantly different from control. Similar results were obtained at 100 pM magnesium which caused approximately 60% of maximal activity (.not shown). Conversely, although 20 mmol sodium azide, a mitochondrial ATPase inhibitor, reduced activity that was stimulated by both divalent cations, the inhibition of Mg-dependent activity was significantly greater than that of Ca-dependent activity (P < 0.001). Control samples for the azide experiments contained 20 mmol of sodium chloride and either 1.2 x 10 ‘/mol free calcium or 8 x 1O-4 M magnesium. When both magnesium and calcium were added to the incubation mixture, the cleavage of ATP (14.8 k 1.2%) was less than the sum of that seen in the presence of calcium (3.7 i 0.3%) or magnesium I. 1%) alone. To maximize the effects of
0
L
lo+
’ ’ ICY5 lo-4
[TRIFLU~PERA~INE..
A
lo-3
, M
Effect of trifluoperazine (TFP) on Cs-ATPase activity in the presence of 1.2 x 1 O-’ M free calcium as described in Fig. 1. Control represents 5.7 k 0.6 of ADP formed. Half-maximal inhibition occurred at 7 x lOm6 M TFP.
NO+ Fib+
[CATION],
mu
Fig. 5. Effect of monovalent cations on high affinity CaATPase. Ca-ATPase activity was measured as described in Fig. 1 in the presence of 1.2 x lo-’ M free calcium and increasing concentrations of K+, Na+. Rb+, and Li+. Each point represents the mean of three to five experiments, each done in triplicate. Control represents 3.6 f 0.2% of ADP formed. The indicates significance (P < 0.01) above control. l
calcium on ATPase activity, therefore, most experiments were carried out without added magnesium. Calmodulin has been found in bovine parathyroid cells.” To examine the possibility that this activator protein influences Ca-ATPase activity, we studied the effect of TFP, a known inhibitor of calmodulin, on Ca-ATPase activity (Fig. 4). Increasing concentrations of TFP caused a progressive inhibition of high affinity Ca-ATPase activity, with half-maximal inhibition occurring at 7 x 10s5 M TFP. In two experiments, 10e4 M TFP likewise caused 82% inhibition of low affinity Ca-ATPase (at lOa M calcium). The addition of exogenous calmodulin (100 ng) had no effect on low or high affinity Ca-ATPase activity. The effect of monovalent cations on Ca-ATPase activity in the presence of 1.2 x 1O--’ M free calcium is shown in Fig. 5. Although potassium ions were most effective in stimulating Ca-ATPase activity, a significant increase in activity (P < 0.01) was also observed with each concentration of added sodium. Higher concentrations of rubidium (80 and 120 mmol) and 40 mmol of lithium also significantly enhanced CaATPase activity (P < 0.01). In the absence of added calcium (in assay buffer containing EGTA), there was no increase in basal activity with the addition of any of the monovalent cations in the concentration range of 0 to 120 mmol. DISCUSSION
Ca-ATPase activity has been reported in many different cell types and localized to the plasma mem-
878
and endoplasmic reticubrane, mitochondria, lum. 5-6*‘9-20 In most of these tissues, Ca-ATPase(s) have been found to have two sites of differing affinities for calcium. A physiologically relevant Ca-ATPase that functions to extrude calcium from the cytosol should be activated by a calcium concentration that is close to that of the cytosol (10-7/mol). The high affinity Ca-ATPase(s) described here has a Km(Ca) in this range (3 x lo-’ M), as do those recently described in rat adipocyte,’ and in mouse and rat pancreatic islet plasma membranes692’ values for Km(Ca) range from 7 x IO-* to 1.4 x 10m7M. The 27,000 x g pellet of the homogenate undoubtedly contains a mixture of plasma membranes, mitochondria, and endoplasmic reticulum. However, it is likely that this preparation contains a substantial proportion of plasma membranes since (1) plasma membranes are pelleted at 15,000 x g,22 (2) faster centrifugation (78,000 x g) is required to pellet endoplasmic reticulum,22 and (3) azide, a mitochondrial ATPase inhibitor, caused only 30% inhibition of CaATPase activity (Fig. 3). Furthermore, Ca-ATPase is traditionally a magnesium-requiring enzyme with an optimal magnesium concentration of 4 to 5 x 10e3 mM for sarcoplasmic reticulum Ca-ATPasez3 but only 3 to 5 x 10m6M for plasma membrane Ca-ATPase.’ In the latter studies, in the absence of exogenous magnesium, assay medium and tissue contained about 6 x lO-‘j mol magnesium from endogenous sources, which was sufficient to activate plasma membrane Ca-ATPase. Thus, using a similar assay medium without added magnesium, we suspect that enough endogenous magnesium is present to meet the presumed requirement of plasma membrane Ca-ATPase but not that of endoplasmic reticulum Ca-ATPase in bovine parathyroid cells. The membrane preparation used in this study contained very active Mg-ATPase(s). Our available data do not allow us to establish whether Ca- and Mgdependent ATPase are separate enzymic entities. Sodium and azide, however, had differential effects on these two enzymic activities. Sodium significantly stimulated cleavage in the presence of calcium (P < 0.001) but not in the presence of magnesium. Duggan similarly found no effect of 20 mmol sodium on Mg-ATPase activity and a slightly inhibitory effect at higher concentrations.24 Furthermore, although the mitochondrial ATPase inhibitor, sodium azide, inhibited ATPase cleavage in the presence of each divalent cation, the degree of inhibition was significantly greater in the presence of magnesium than in calcium (P < 0.001). On theother hand, we have found that the effects of calcium and magnesium on ATP cleavage are not additive. Vilhardt and Hope made a similar observation in bovine neural lobe and yet were able to
DAWSDN-HUGHES
ET AL
demonstrate the presence of two enzymes on disc gel electrophoresis, one activated by calcium alone and the other by either calcium or magnesium.25 Goz also found that the stimulatory effects of magnesium and calcium on ATPase activity in microsomes from bovine adrenal medulla were not additive and demonstrated that the addition of magnesium to the assay had an inhibitory effect on calcium-dependent activity.26 The reverse has been shown in other systems, wherein calcium inhibits magnesium-dependent enzymes such as adenylate cyclase27 and protein kinase.” Thus, the finding that the effects of calcium and magnesium on ATP cleavage are not additive may reflect complex interactions of these cations and ATPase(s). To facilitate characterization of Ca-ATPase, we carried out most of our experiments in the absence of added magnesium, as a number of others have done.5*6*25 The magnesium concentration to which this enzyme is exposed in the parathyroid cell is unknown. Although it is possible that the two cations stimulate a single enzyme in bovine parathyroid cells, our data are also consistent with those of workers who have directly demonstrated the presence of two separate enzymes. The ubiquitous calcium-binding protein, calmodulin, activates a variety of cellular enzymes in a calcium-dependent fashion, such as myosin light-chain kinase29 and cyclic nucleotide phosphodiesterase.30 We have recently found that bovine’* and human3’ parathyroid cells contain calmodulin and that the activity of this protein is inhibited by the phenothiazine trifluoperazine (TFP), with an IC,o of approximately 3 x 1O-‘/mol. Furthermore, calmodulin has been shown to activate Ca-ATPase in red blood cells, and this activity has been half-maximally inhibited by 1.35 x 10m4M TFP.32 The concentration of TFP that is required to inhibit half-maximally high affinity Ca-ATPase in this study (7 x lo-‘/mol) is similar to those already cited. Moreover, maximal inhibition of bovine parathyroid Ca-ATPase by TFP (60%) is similar to that previously observed in Ca-ATPase from red blood cells (70%).32 These results must be interpreted with caution since phenothiazines are membrane-active agents that might affect many cellular processes. Although exogenous calmodulin did not stimulate activity, we have no assurance that the membranes were depleted of endogenous calmodulin during preparation of cellular particmates for enzyme assay. Stimulation of Ca-ATPase by monovalent cations and their order of effectiveness in our studies (K’ > Na+ > Rb’ > Li+) are in agreement with reports in other cell types.7*33,34 The mechanism(s) by which this stimulation occurs is not fully established. Duggan reported that potassium increased both Ca-ATPase activity and the rate of calcium uptake in isolated
CA-ATPASE
ACTIVITY
IN BOVINE PARATHYROID
a79
CELLS
muscle microsomes without any overall change in the efficiency of the latter process7 Ca-ATPase from rabbit sarcoplasmic reticulum was less sensitive to tryptic digestion in the presence of greater than 50 mM K+, Nat, or Lif, suggesting that monovalent cations may stabilize the conformation of Ca-ATPase.35 Thus, cations may interact directly with Ca-ATPase to stimulate its activity. Our understanding of the factors that control PTH release remains incomplete. Indirect evidence suggests that changes in cytosolic calcium concentration may play an important role in regulating the secretory process with an inhibition of secretion at high cytosolic calcium concentrations.“” Recent studies implicate sodium-calcium exchange as one mechanism that contributes to maintaining a normal cytosolic calcium concentration in this cell type.36v37The high affinity
Ca-ATPase described in the present studies may also regulate cytosolic calcium in the parathyroid cell by removing calcium from the cytosol. Our studies do not suggest any unique properties of the enzyme(s) in this cell type that might directly link changes in extracellular calcium to alterations in enzyme activity and, consequently, cytosolic calcium. It is possible, therefore, that it plays a role in the maintenance of the resting cytosolic calcium concentration with other transport systems for calcium contributing to changes in cytosolic calcium. ACKNOWLEDGMENTS We would like to thank Dr. John Gergely for his assistance in calculating free calcium concentrations with Ca-EGTA buffers and Mrs. Nancy Orgill and Miss Diane Rioux for typing the manuscript.
REFERENCES 1. Portzehl H. Caldwell PC, Ruegg JC: The dependence of contraction and relaxation of muscle fibers from the crab maia squinado on the external concentration of free calcium ions. Biochim Biophys Acta 79581-591, 1964 2. Schatzmann HJ: Active calcium transport ATPase in human red cells. Curr Top Membr 1975
and Ca2+-activated Transp 6:125-168,
of the dopamine-responsive adenylate cyclase of bovine parathyroid cells and its relationship to parathyroid hormone secretion. Endocrinology 107:1776-1781, 1980 15. Perrin DD, Sayce JG: Computer calculation of equilibrium concentrations in mixtures of metal ions and complexing agents. Talanta 14:833-842, 1967 16. Bancroft H: Introduction & Row Publishers, 1957, p 74
to biostatistics.
New York, Harper
3. Blaustein MP: The interrelationship between sodium and calcium fluxes across cell membranes. Rev Physiol Biochem Pharmacol70:33-82. 1974
17. Dunnett CW: New tables control. Biometrics 10:482, 1964
4. DiPolo 274:39&39
18. Brown EM: Calcium-regulated phosphodiesterase parathyroid cells. Endocrinology 107:1998-2003, 1980
R: Ca pump driven
by ATP in squid axons.
Nature
I, 1978
5. Pershadsingh HA, McDonald JM: A high affinity calciumstimulated magnesium-dependent adenosine triphosphatase in rat adipocyte plasma membranes. J Biol Chem 255:4087-4093, 1980
19. Quist erythrocytes.
for multiple
comparisons
with a in bovine
EE, Roufogalis BD: Calcium transport in human Arch Biochem Biophys 168:240-25 1, 1975
6. Formby B, Capito K, Egeberg J, et al: Ca-activated ATPase activity in subcellular fractions of mouse pancreatic islets. Am J Physiol 230:441-448, 1976
20. Scharff 0: Ca2+ activation of membrane-bound (Ca” + Mg”)-dependent ATPase from human erythrocytes prepared in the presence or absence of Ca*+. Biochem Biophys Acta 443:206-218, 1976
7. Duggan PF: Calcium uptake and associated adenosine triphosphatase activity in fragmented sarcoplasmic reticulum. J Biol Chem 252:162&1627, 1977
21. Pershadsingh HA, McDaniel ML, Landt M, et al: Ca’+activated ATPase and ATP-dependent calmodulin-stimulated Ca*+ transport in islet cell plasma membrane. Nature 288:490-495, 1980
3. Baker PF, McNaughton cium efflux from squid giant 144. 1976
22. Maciel RMB, Ozawa Y, Chopra IJ: Subcellular localization of thyroxine and reverse triiodothyronine outer ring monodeiodinating activities. Endocrinology 104:365-371, 1979
PA: Kinetics and energetics of calaxons. J Physiol (London) 259:103-
9. Habener JF, Stevens TD, Ravazzola M, et al: Effects of calcium ionophores on the synthesis and release of parathyroid hormone. Endocrinology 101:1524-I 537, 1977
23. Hasselbach W: Sarcoplasmic membrane ATPases, in Boyer PD (ed): The Enzymes, Vol X. New York, Academic Press, 1974, pp 43 1467
IO. Brown EM, Gardner DG, Aurbach GD: Effects of calcium ionophore A23187 on dispersed bovine parathyroid cells. Endocrinology 106:133-138,198O
24. Duggan PF: Calcium-independent adenosine triphosphatase activity in frog muscle microsomes. Life Sci 7: 1265-l 27 1, 1968
Il. Brown EM, Hurwitz S, Aurbach isolated 1976
bovine
parathyroid
cells.
GD: Preparation of viable Endocrinology 99:1582-1588,
Hoffman NE, Liao JC: Reversed phase high performance liquid chromatographic separations of nucleotides in the presence of solvophobic ions. Anal Biochem 49:223 l-2234, 1977 i2.
3. Seals JR, McDonald JM, Bruns D, et al: A sensitive and precise isotopic assay of ATPase activity. Ann Biochem 90:785-795, 1978 14. Attie MF, Brown EM, Gardner
DG, et al: Characterization
25. Vilhardt H, Hope DB: Adenosine triphosphatase the neural lobe of the bovine pituitary gland. Biochem 190,1974
activity in J 143:181-
26. Goz B: Properties of a microsomal adenosine triphosphatase from the adrenal medulla. Bicchem Pharmacol 16:593-596, 1967 27. Matsuzaki S, Dumont JE: Effect of calcium ion on horse parathyroid gland adenyl cyclase. Biochim Biophys Acta 284:227234, 1972 28. Pines M, Hurwitz S: Cyclic-AMP activates and calcium inhibits protein kinase activity in avian parathyroid glands. FEBS Lett 133:27-30, 1981
880
29. Nairn AC, Perry SV: Calmodulin and myosin light-chain kinase of rabbit fast skeletal muscle. Biochem J 179:89-97, 1979 30. Dedman JR, Potter J, Jackson RL, et al: Physicochemical properties of rat testis Ca*+-dependent regulator protein of cyclic nucleotide phosphodiesterase. Relationship of Ca*+-binding, conformational changes and phosphodiesterase activity. J Biol Chem 252:8415-8422,1977 3 1. Brown EM, Dawson-Hughes BF, Wilson RE, et al: Calmodulin in dispersed human parathyroid cells. J Clin Endocrinol Metab 53:1064-1071,1981 32. Roufogalis BD: Phenothiazine antagonism of calmodulin: A structurally-nonspecific interaction. Biochem Biophys Res Commun 98:607-613,1981 33. Scharff 0: Stimulating effects of monovalent cations on
DAWSON-HUGHES
ET AL
activator-associated states of Ca’-ATPase in human erythrocytes. Biochim Biophys Acta 512:309-317, 1978 34. Jones LR, Besch Jr HR, Watanabe AM: Monovalent cation stimulation of Ca*+ uptake by cardiac membrane vesicles. J Biol Chem 252:3315-3323,1977 35. Louis CF, Buonaffina R, Binks B: Effect of trypsin on the proteins of skeletal muscle sarcoplasmic reticulum. Arch Biochem Biophys 161:83-92, 1974 36. Brown EM, Adragna N, Gardner DG: Effect of potassium on PTH secretion from dispersed bovine parathyroid cells. J Clin Endocrinol Metab 53: 1304-I 306, 198 1 37. Rothstein M, Morrissey J, Slatopolsky E, et al: The role of Na/Ca exchange on PTH secretion. Endocrinology 111:225-230, 1982
Activity
Bess F. Dawson-Hughes,
in Bovine Parathyroid
Richard H. Underwood,
and Edward
Cells
M. Brown
Ca-ATPase is thought to function as a calcium extrusion pump that may regulate cytosolic calcium concentration. Because the parathyroid gland is among the few tissues that are directly regulated by extracellular calcium and because cytosolic calcium may be a mediator of the effects of extracellular calcium on parathyroid hormone secretion, we have investigated the presence of this enzyme in homogenates of parathyroid cells. High performance liquid chromatography (HPLC) was used to quantify the formation of ADP from ATP following incubation of ATP with cellular homogenate in a buffer containing ethylenedioxy- (diethylenedinitrilo) tetra acetic acid (EGTA), ouabain, and calcium. Enzyme activity was calcium-dependent, with Ca-ATPase showing two Km (Ca) values, 31 and 853 nM. High affinity Ca-ATPase activity was reduced by the calmodulin inhibitor, trifluoperazine (TFP), with half-maximal inhibition occurring at 7 x lo-’ M. Monovalent cations stimulated high affinity Ca-ATPase activity (K+ > Na+ > Rb+ z Lif) in the presence of calcium. Magnesium (0.8 mM) also stimulated cleavage of ATP. Sodium increased Ca-dependent ATPase activity by 82% but had no significant effect on Mg-stimulated activity. Furthermore, azide, an inhibitor of mitochondrial ATPase(s1, had a significantly greater inhibitory effect on Mg-dependent than on Ca-dependent activity. In summary, a high affinity Ca-ATPase is present in bovine parathyroid cells which has a Km in the range of the cytosolic calcium concentration that is found in other cells. Ca-ATPaselsf may be of importance in regulating the cytosolic calcium concentration and, therefore, hormonal secretion in this cell type.
HERE is a large gradient for calcium across the plasma membrane, with the extracellular free calcium concentration ( 1O-3 M) being about four orders of magnitude higher than that in the cytoplasm (lo-’ mo1).‘S2Although the mechanisms by which this gradient is maintained are not fully understood, several have been identified. Sodium-calcium exchange is a process that couples calcium extrusion to the inward movement of sodium down its electrochemical gradient.3 ATP indirectly supplies the energy for this process through maintenance, by the (Na+-K+)ATPase, of the transmembrane gradient for sodium.3 In addition, calcium-dependent ATPases (CaATPase) have been shown to participate directly in the regulation of cytosolic calcium concentration in the intact squid giant axon4 and in the red blood ce11.2This enzyme has been found in plasma membrane, mitochondria and microsomes of adipocytes,’ pancreatic islets,6 muscle,’ and nerve cells,8 where indirect evidence suggests that it extrudes calcium from the cytosol, both across the plasma membrane and into intracellular compartments.
T
From the Endocrine-Hypertension Unit, Brigham and Women’s Hospital, and Department of Medicine, Harvard Medical School, Boston, Mass. Received for publication February 23, I983. This work was supported by grants from the National Institutes of Health (1 F32 AM06759-01. 2 T32 HLO7236-05. and 1 R01 AM30028-01). Dr. Brown is recipient of NIH-PHS Research Career Development Award, 1K04-AA400627-04. This work was presented in part at the American Federation for Clinical Research Eastern Section. Clinical Research 29.6814. 1981. Address reprint requests to Dr. Bess Dawson-Hughes. EndocrineHypertension Unit, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. Q 1983 by Grune & Stratton, Inc. 0026-0495/83/3209J.IOO3$01.00/0 a74
The parathyroid gland is among the few tissues that are directly regulated by the extracellular calcium concentration. Indirect evidence suggests that cytosolic calcium may be a mediator of the effects of extracellular calcium on parathyroid hormone (PTH) secretion.“” Because Ca-ATPase affects the transmembrane calcium gradient in other cells and because this gradient may play a role in PTH secretion, we investigated the presence of Ca-ATPase(s) in parathyroid cells. In this paper we report several of the properties of Ca-ATPase(s) in dispersed bovine parathyroid cells. MATERIALS
AND
METHODS
Isolated
bovine parathyroid cells were prepared by the method of Brown et al” which entailed mincing and digestion of parathyroid tissue with collagenase and DNAse. After several washings, dispersed cell preparations contained approximately 95% parathyroid cells and 5% erythrocytes. The dispersed parathyroid cells were homogenized by using ten strokes with a manual homogenizer in a solution of 0.25 moles sucrose, 1 mmole EGTA, and 15 mM tris buffer (pH 7.5) at 4 OC and a cell concentration of 10 million/ml. The homogenate was then sedimented for 15 minutes at 27,000 x g at 4 OC. The pellet was resuspended in 20 mmoles tris buffer (pH 7.5) and used directly in the assay. Ca-ATPase activity was quantified by measuring the amount of ADP that is released from ATP. The assay mixture in a final volume of 0.1 ml contained 1 mmole of ATP, 1 mmole of EGTA, 0.1 mmole of ouabain [to inhibit (Na+-K’)-ATPase activity], 20 mM tris buffer (pH 7.5), variable amounts of CaCl,, no added magnesium, and homogenate equivalent to 200,000 parathyroid cells per tube. Assays were carried out for 30 minutes at 37 OC and terminated by placing samples in a boiling water bath for two minutes. Samples were then centrifuged at 10,000 x g for two minutes to remove cellular debris and 15 ~1of supernatant was analyzed by HPLC. The HPLC system contained the following components: Model 5040 Terniary Liquid Chromatograph, Vista 401 Chromatography Data System, Varichrom UV-VIS Spectrophotometric Detector (Varian Instrument Division), Supelcosil LC-18-DB column 15 x 4.6 mmole ID (Supelco, Inc.), and a Guard Column packed with Metabolism, Vol. 32, No. 9 Eeptember), 1993
875
CA-ATPASE ACTIVITY IN BOVINE PARATHYROID CELLS
3ondapak C,,/Corasil (Waters Associates, Inc.). The solvent system was 0.10 M KH,PO, containing 0.025 mole tetra-nbutyl ammonium hydrogen sulfate (pH 7.2) and 13% methanol,” and the flow rate was 2 mls/min. Absorbance was read at 259 rnp and areas under the nucleotide peaks were integrated. Retention times for ATP, ADP, and AMP were established for each experiment by direct comparison with a standard mixture of these nucleotides. In some experiments, the cleavage of ATP was assessed by quantifying the release of 32P from [y-32P]ATP according to the method of Seals et al.13 Results obtained with this method were comparable to those with the HPLC method already described. Because of reduced cost and radiation exposure and the capability of monitoring all the nucleotides, the HPLC method was used in most studies. All samples were assayed in triplicate. The protein concentration in the 27,000 x g pellet was equivalent to about 1 mg/2.5 x 10’ cells.‘4 Although the intraassay coefficient of variation for cleavage is 3% to 4%, interassay variation is greater, probably at least in part due to errors in estimating cell count, upon which the membrane concentration is based. The concentration of free calcium in CaEGTA buffers was calculated according to the equations of Perrin and Sayce.” Results are reported as means k SEM. Statistical probability was evaluated by using students’ t-test as outlined by Bancroft.16 For calculation of threshold concentration (that concentration of additive that causes a significant change from control), we used analysis of variance, obtaining P values from Dunnett’s tables.” ly-“P]ATP was obtained from Amersham/Searle, Arlington Heights, Ill, tetrabutylammonium hydrogen sulfate from Aldrich Chemical Company, Inc., ATP and ADP from Sigma Chemical Company, and AMP from Eastman. RESULTS
In order to quantify enzymatic cleavage by the use of HPLC, the stability of ATP and ADP in the assay were first assessed. When ATP was incubated without
enzyme for 30 minutes at 37 “C, 96.6% appeared on HPLC as ATP, 2.0% as ADP, and 3.3% as AMP. At 1.2 x lo-’ M calcium, there was no change in spontaneous conversion of ATP to ADP or AMP. When ADP was incubated without enzyme, only 5% was spontaneously cleaved to AMP, and thus 95% of the AMP arose directly from ATP. In an additional experiment, calcium caused a slight decrease in conversion of ADP to AMP (from 6% to 4.4%). Since this results in less than a 2% change in apparent conversion of ATP to ADP, this would not significantly affect the subsequent results. When ATP was incubated with homogenate equivalent to 200,000 cells/tube with or without 1.2 x IO-‘/mol free calcium (Fig. l), addition of calcium increased the proportion of ADP from 1.3 to 7.396, whereas the AMP remained relatively constant at about 9%. To determine whether ADP is stable in the presence of enzyme, additional experiments were carried out (not shown) in which standard amounts of enzyme were incubated with ADP. This resulted in conversion of less than 10% of the ADP to AMP. Therefore, less than 10% of the total AMP that is produced arises from ADP, indicating that the area under the ADP peak is a reasonable estimate of the total ADP that is formed. AMP presumably arises directly from ATP by pyrophosphorylation. Subsequent results are expressed as a percent of total nucleotide present as ADP [ADP/(ATP + ADP + AMP) x 1001 as an estimate of the formation of ADP from ATP.
USE OF HPLC TO DEMONSTRATE Co-DEPENDENT ATPose ACTIVITY IN HOMOGENATE OF BOVINE RWXIWROID CELLS -
CALCIUM
+ CALCIUM
c
AMP Fig. 1. HPLC separation of nucleotides from assays done in the absence and presence of 1.2 x lo-’ M free calcium. The assay mixture also contained 1 mM ATP, 1 mM EGTA. 0.1 mM ouabain. 20 mM tris buffer (pH 7.5). and homogenate equivalent to 200,000 bovine parethyroid cells in a final volume of 0.1 ml. Assays were carried out at 37 “C for 30 minutes.
I
I
,
I
1
0
2
I
,
I
,
I
4
6
6
0
2
4
6
8
MIN
MIN
876
DAWSON-HUGHES
ET AL
6V 4-
I OLYIO-*
I lo*
I lo-’
FREE [Co++],
I
I
ItI)
Icr’
M
I O
I 0
I
I
I
I
40
80
120
160
v/s
Effect of Caf+ on Ca-ATPase activity in homogenate of bovine parathyroid cells. IA) Results of a single representative Fig. 2. experiment done in triplicate by using the method of Seals to quantify release of =P from [y-PP]ATP. The blank, representing spontaneous hydrolysis of P from ATP, was 0.64% + 0.07 (SEMI and has been subtracted. (B) Hofstee plot of the data from Fig. 2(A) in which V = nmoles Plhrltube and v/s = pmol P/hrltube/[ca+‘]. Two distinct slopes, -8 (r = 0.91) and -738 (r = -0.921, are apparent, indicating the presence of high and low affinity Ca-ATPase activities with KM ICa) values of 8 nM and 738 n&l, respectively. The maan Km (Cal values of three experiments, each done in triplicate are 31 + 12 nM and 883 r 114 nM.
Calcium-dependent activity was determined by subtracting the percent of total nucleotide present as ADP with EGTA alone from that obtained with chelator plus calcium. Similarly, magnesium-stimulated activity was that increment that occurred with the addition of magnesium to the incubate containing EGTA. The activity of Ca-ATPase was linear with time for up to 60 minutes and was directly proportional to the conCo -ATPose
T-
Mg - ATPose
T
ant
MAzkk
mmm
m ‘omm
Fig. 3. Effect of sodium and aside on Ca- and Mg-ATPase activities. The effect of sodium on Ca- and Mg-ATPase activity was determined by measuring cleavage in the presence of either 1.2 x lo-’ M calculated free calcium or 8 x 1Om4M added magnesium, respectively, with and without (control) the addition of 48 mM NaCI. The effects of azide were determined similarly. However. since sodium azide was used, an amount of sodium equivalent to that in the aside preparation was included in the control. Although aside inhibited both Mg- and Ca-dependent activity, the degree of inhibition of Mg-dependent activity was significantly greater (P < 0.001). Sodium significantly enhanced Ca-ATPase activity (P < 0.001) but had no significant effect on Mg-ATPase activity. Each bar represents the mean of three experiments, each done in triplicate. Control bars represent 8.7 r 1.6% of ADP formed for Ca and 31 + 4% for Mg.
centration of homogenate added. The rate of cleavage of ATP was sixfold slower at 4 “C and ATPase activity was destroyed by boiling. The dispersed parathyroid cells that were used in the assay contain 5% to 10% red blood cells.” Since red blood cells contain Ca-ATPase,* we assessed their contribution to ATP cleavage in our assay. When the predicted number of bovine red blood cells were prepared and assayed as above in the presence of 1.2 x lo-’ M free calcium, there was only 0.23% calciumdependent cleavage of ATP, which is negligible. In the presence of ten times the number of red blood cells and a higher free calcium concentration (6 x lo-’ M), only 0.7% of ATP was cleaved. Thus, the presence of red blood cells in the parathyroid cell preparation had a negligible influence on the results. A calcium activation curve for Ca-ATPase in homogenates of bovine parathyroid cells is shown in Fig. 2(A). With increasing calcium concentrations, there was a progressive increase in ATPase activity, which was biphasic with respect to calcium concentration, with near maximal activity at 10m4 M free calcium. Figure 2(B) shows a Hofstee plot of the data in Fig. 2(A). Two distinct slopes are apparent, indicating the presence of high and low affinity Ca-ATPase activities. In this experiment, their Km(Ca) values are 8 nM and 738 nmol, respectively. The mean Km(Ca) values
*Although the Km(Ca) for the high affinity activity in Fig. 2 (8 nmoles) is lower than the values obtained by HPLC (average 42 nmoles), in a total of three such experiments that were carried out by using the r*P method, the km (Ca) for the high affinity site averaged 25 nmoles. Thus, the two methods appear to yield comparable results.
CA-ATPASE ACTIVITY IN BOVINE PARATHYROID CELLS
877
of three experiments (two using HPLC, one using 32P), each done in triplicate, were 31 + 12 nM and 853 -r114 nmol.* Magnesium also activated ATPase in this assay. To determine whether calcium and magnesium stimulate the same or different enzymes, we studied the effect of several agents on cleavage in the presence of either calcium or magnesium (Fig. 3). When 40 nmol of sodium was added to the incubation mixture containing 1.2 x lo-’ M free calcium, enzyme activity significantly increased by 82% (P < 0.001); whereas in the presence of 8 x 10m4 mol magnesium, which gave nearly maximal stimulation, activity increased by only 1l%, a value not significantly different from control. Similar results were obtained at 100 pM magnesium which caused approximately 60% of maximal activity (.not shown). Conversely, although 20 mmol sodium azide, a mitochondrial ATPase inhibitor, reduced activity that was stimulated by both divalent cations, the inhibition of Mg-dependent activity was significantly greater than that of Ca-dependent activity (P < 0.001). Control samples for the azide experiments contained 20 mmol of sodium chloride and either 1.2 x 10 ‘/mol free calcium or 8 x 1O-4 M magnesium. When both magnesium and calcium were added to the incubation mixture, the cleavage of ATP (14.8 k 1.2%) was less than the sum of that seen in the presence of calcium (3.7 i 0.3%) or magnesium I. 1%) alone. To maximize the effects of
0
L
lo+
’ ’ ICY5 lo-4
[TRIFLU~PERA~INE..
A
lo-3
, M
Effect of trifluoperazine (TFP) on Cs-ATPase activity in the presence of 1.2 x 1 O-’ M free calcium as described in Fig. 1. Control represents 5.7 k 0.6 of ADP formed. Half-maximal inhibition occurred at 7 x lOm6 M TFP.
NO+ Fib+
[CATION],
mu
Fig. 5. Effect of monovalent cations on high affinity CaATPase. Ca-ATPase activity was measured as described in Fig. 1 in the presence of 1.2 x lo-’ M free calcium and increasing concentrations of K+, Na+. Rb+, and Li+. Each point represents the mean of three to five experiments, each done in triplicate. Control represents 3.6 f 0.2% of ADP formed. The indicates significance (P < 0.01) above control. l
calcium on ATPase activity, therefore, most experiments were carried out without added magnesium. Calmodulin has been found in bovine parathyroid cells.” To examine the possibility that this activator protein influences Ca-ATPase activity, we studied the effect of TFP, a known inhibitor of calmodulin, on Ca-ATPase activity (Fig. 4). Increasing concentrations of TFP caused a progressive inhibition of high affinity Ca-ATPase activity, with half-maximal inhibition occurring at 7 x 10s5 M TFP. In two experiments, 10e4 M TFP likewise caused 82% inhibition of low affinity Ca-ATPase (at lOa M calcium). The addition of exogenous calmodulin (100 ng) had no effect on low or high affinity Ca-ATPase activity. The effect of monovalent cations on Ca-ATPase activity in the presence of 1.2 x 1O--’ M free calcium is shown in Fig. 5. Although potassium ions were most effective in stimulating Ca-ATPase activity, a significant increase in activity (P < 0.01) was also observed with each concentration of added sodium. Higher concentrations of rubidium (80 and 120 mmol) and 40 mmol of lithium also significantly enhanced CaATPase activity (P < 0.01). In the absence of added calcium (in assay buffer containing EGTA), there was no increase in basal activity with the addition of any of the monovalent cations in the concentration range of 0 to 120 mmol. DISCUSSION
Ca-ATPase activity has been reported in many different cell types and localized to the plasma mem-
878
and endoplasmic reticubrane, mitochondria, lum. 5-6*‘9-20 In most of these tissues, Ca-ATPase(s) have been found to have two sites of differing affinities for calcium. A physiologically relevant Ca-ATPase that functions to extrude calcium from the cytosol should be activated by a calcium concentration that is close to that of the cytosol (10-7/mol). The high affinity Ca-ATPase(s) described here has a Km(Ca) in this range (3 x lo-’ M), as do those recently described in rat adipocyte,’ and in mouse and rat pancreatic islet plasma membranes692’ values for Km(Ca) range from 7 x IO-* to 1.4 x 10m7M. The 27,000 x g pellet of the homogenate undoubtedly contains a mixture of plasma membranes, mitochondria, and endoplasmic reticulum. However, it is likely that this preparation contains a substantial proportion of plasma membranes since (1) plasma membranes are pelleted at 15,000 x g,22 (2) faster centrifugation (78,000 x g) is required to pellet endoplasmic reticulum,22 and (3) azide, a mitochondrial ATPase inhibitor, caused only 30% inhibition of CaATPase activity (Fig. 3). Furthermore, Ca-ATPase is traditionally a magnesium-requiring enzyme with an optimal magnesium concentration of 4 to 5 x 10e3 mM for sarcoplasmic reticulum Ca-ATPasez3 but only 3 to 5 x 10m6M for plasma membrane Ca-ATPase.’ In the latter studies, in the absence of exogenous magnesium, assay medium and tissue contained about 6 x lO-‘j mol magnesium from endogenous sources, which was sufficient to activate plasma membrane Ca-ATPase. Thus, using a similar assay medium without added magnesium, we suspect that enough endogenous magnesium is present to meet the presumed requirement of plasma membrane Ca-ATPase but not that of endoplasmic reticulum Ca-ATPase in bovine parathyroid cells. The membrane preparation used in this study contained very active Mg-ATPase(s). Our available data do not allow us to establish whether Ca- and Mgdependent ATPase are separate enzymic entities. Sodium and azide, however, had differential effects on these two enzymic activities. Sodium significantly stimulated cleavage in the presence of calcium (P < 0.001) but not in the presence of magnesium. Duggan similarly found no effect of 20 mmol sodium on Mg-ATPase activity and a slightly inhibitory effect at higher concentrations.24 Furthermore, although the mitochondrial ATPase inhibitor, sodium azide, inhibited ATPase cleavage in the presence of each divalent cation, the degree of inhibition was significantly greater in the presence of magnesium than in calcium (P < 0.001). On theother hand, we have found that the effects of calcium and magnesium on ATP cleavage are not additive. Vilhardt and Hope made a similar observation in bovine neural lobe and yet were able to
DAWSDN-HUGHES
ET AL
demonstrate the presence of two enzymes on disc gel electrophoresis, one activated by calcium alone and the other by either calcium or magnesium.25 Goz also found that the stimulatory effects of magnesium and calcium on ATPase activity in microsomes from bovine adrenal medulla were not additive and demonstrated that the addition of magnesium to the assay had an inhibitory effect on calcium-dependent activity.26 The reverse has been shown in other systems, wherein calcium inhibits magnesium-dependent enzymes such as adenylate cyclase27 and protein kinase.” Thus, the finding that the effects of calcium and magnesium on ATP cleavage are not additive may reflect complex interactions of these cations and ATPase(s). To facilitate characterization of Ca-ATPase, we carried out most of our experiments in the absence of added magnesium, as a number of others have done.5*6*25 The magnesium concentration to which this enzyme is exposed in the parathyroid cell is unknown. Although it is possible that the two cations stimulate a single enzyme in bovine parathyroid cells, our data are also consistent with those of workers who have directly demonstrated the presence of two separate enzymes. The ubiquitous calcium-binding protein, calmodulin, activates a variety of cellular enzymes in a calcium-dependent fashion, such as myosin light-chain kinase29 and cyclic nucleotide phosphodiesterase.30 We have recently found that bovine’* and human3’ parathyroid cells contain calmodulin and that the activity of this protein is inhibited by the phenothiazine trifluoperazine (TFP), with an IC,o of approximately 3 x 1O-‘/mol. Furthermore, calmodulin has been shown to activate Ca-ATPase in red blood cells, and this activity has been half-maximally inhibited by 1.35 x 10m4M TFP.32 The concentration of TFP that is required to inhibit half-maximally high affinity Ca-ATPase in this study (7 x lo-‘/mol) is similar to those already cited. Moreover, maximal inhibition of bovine parathyroid Ca-ATPase by TFP (60%) is similar to that previously observed in Ca-ATPase from red blood cells (70%).32 These results must be interpreted with caution since phenothiazines are membrane-active agents that might affect many cellular processes. Although exogenous calmodulin did not stimulate activity, we have no assurance that the membranes were depleted of endogenous calmodulin during preparation of cellular particmates for enzyme assay. Stimulation of Ca-ATPase by monovalent cations and their order of effectiveness in our studies (K’ > Na+ > Rb’ > Li+) are in agreement with reports in other cell types.7*33,34 The mechanism(s) by which this stimulation occurs is not fully established. Duggan reported that potassium increased both Ca-ATPase activity and the rate of calcium uptake in isolated
CA-ATPASE
ACTIVITY
IN BOVINE PARATHYROID
a79
CELLS
muscle microsomes without any overall change in the efficiency of the latter process7 Ca-ATPase from rabbit sarcoplasmic reticulum was less sensitive to tryptic digestion in the presence of greater than 50 mM K+, Nat, or Lif, suggesting that monovalent cations may stabilize the conformation of Ca-ATPase.35 Thus, cations may interact directly with Ca-ATPase to stimulate its activity. Our understanding of the factors that control PTH release remains incomplete. Indirect evidence suggests that changes in cytosolic calcium concentration may play an important role in regulating the secretory process with an inhibition of secretion at high cytosolic calcium concentrations.“” Recent studies implicate sodium-calcium exchange as one mechanism that contributes to maintaining a normal cytosolic calcium concentration in this cell type.36v37The high affinity
Ca-ATPase described in the present studies may also regulate cytosolic calcium in the parathyroid cell by removing calcium from the cytosol. Our studies do not suggest any unique properties of the enzyme(s) in this cell type that might directly link changes in extracellular calcium to alterations in enzyme activity and, consequently, cytosolic calcium. It is possible, therefore, that it plays a role in the maintenance of the resting cytosolic calcium concentration with other transport systems for calcium contributing to changes in cytosolic calcium. ACKNOWLEDGMENTS We would like to thank Dr. John Gergely for his assistance in calculating free calcium concentrations with Ca-EGTA buffers and Mrs. Nancy Orgill and Miss Diane Rioux for typing the manuscript.
REFERENCES 1. Portzehl H. Caldwell PC, Ruegg JC: The dependence of contraction and relaxation of muscle fibers from the crab maia squinado on the external concentration of free calcium ions. Biochim Biophys Acta 79581-591, 1964 2. Schatzmann HJ: Active calcium transport ATPase in human red cells. Curr Top Membr 1975
and Ca2+-activated Transp 6:125-168,
of the dopamine-responsive adenylate cyclase of bovine parathyroid cells and its relationship to parathyroid hormone secretion. Endocrinology 107:1776-1781, 1980 15. Perrin DD, Sayce JG: Computer calculation of equilibrium concentrations in mixtures of metal ions and complexing agents. Talanta 14:833-842, 1967 16. Bancroft H: Introduction & Row Publishers, 1957, p 74
to biostatistics.
New York, Harper
3. Blaustein MP: The interrelationship between sodium and calcium fluxes across cell membranes. Rev Physiol Biochem Pharmacol70:33-82. 1974
17. Dunnett CW: New tables control. Biometrics 10:482, 1964
4. DiPolo 274:39&39
18. Brown EM: Calcium-regulated phosphodiesterase parathyroid cells. Endocrinology 107:1998-2003, 1980
R: Ca pump driven
by ATP in squid axons.
Nature
I, 1978
5. Pershadsingh HA, McDonald JM: A high affinity calciumstimulated magnesium-dependent adenosine triphosphatase in rat adipocyte plasma membranes. J Biol Chem 255:4087-4093, 1980
19. Quist erythrocytes.
for multiple
comparisons
with a in bovine
EE, Roufogalis BD: Calcium transport in human Arch Biochem Biophys 168:240-25 1, 1975
6. Formby B, Capito K, Egeberg J, et al: Ca-activated ATPase activity in subcellular fractions of mouse pancreatic islets. Am J Physiol 230:441-448, 1976
20. Scharff 0: Ca2+ activation of membrane-bound (Ca” + Mg”)-dependent ATPase from human erythrocytes prepared in the presence or absence of Ca*+. Biochem Biophys Acta 443:206-218, 1976
7. Duggan PF: Calcium uptake and associated adenosine triphosphatase activity in fragmented sarcoplasmic reticulum. J Biol Chem 252:162&1627, 1977
21. Pershadsingh HA, McDaniel ML, Landt M, et al: Ca’+activated ATPase and ATP-dependent calmodulin-stimulated Ca*+ transport in islet cell plasma membrane. Nature 288:490-495, 1980
3. Baker PF, McNaughton cium efflux from squid giant 144. 1976
22. Maciel RMB, Ozawa Y, Chopra IJ: Subcellular localization of thyroxine and reverse triiodothyronine outer ring monodeiodinating activities. Endocrinology 104:365-371, 1979
PA: Kinetics and energetics of calaxons. J Physiol (London) 259:103-
9. Habener JF, Stevens TD, Ravazzola M, et al: Effects of calcium ionophores on the synthesis and release of parathyroid hormone. Endocrinology 101:1524-I 537, 1977
23. Hasselbach W: Sarcoplasmic membrane ATPases, in Boyer PD (ed): The Enzymes, Vol X. New York, Academic Press, 1974, pp 43 1467
IO. Brown EM, Gardner DG, Aurbach GD: Effects of calcium ionophore A23187 on dispersed bovine parathyroid cells. Endocrinology 106:133-138,198O
24. Duggan PF: Calcium-independent adenosine triphosphatase activity in frog muscle microsomes. Life Sci 7: 1265-l 27 1, 1968
Il. Brown EM, Hurwitz S, Aurbach isolated 1976
bovine
parathyroid
cells.
GD: Preparation of viable Endocrinology 99:1582-1588,
Hoffman NE, Liao JC: Reversed phase high performance liquid chromatographic separations of nucleotides in the presence of solvophobic ions. Anal Biochem 49:223 l-2234, 1977 i2.
3. Seals JR, McDonald JM, Bruns D, et al: A sensitive and precise isotopic assay of ATPase activity. Ann Biochem 90:785-795, 1978 14. Attie MF, Brown EM, Gardner
DG, et al: Characterization
25. Vilhardt H, Hope DB: Adenosine triphosphatase the neural lobe of the bovine pituitary gland. Biochem 190,1974
activity in J 143:181-
26. Goz B: Properties of a microsomal adenosine triphosphatase from the adrenal medulla. Bicchem Pharmacol 16:593-596, 1967 27. Matsuzaki S, Dumont JE: Effect of calcium ion on horse parathyroid gland adenyl cyclase. Biochim Biophys Acta 284:227234, 1972 28. Pines M, Hurwitz S: Cyclic-AMP activates and calcium inhibits protein kinase activity in avian parathyroid glands. FEBS Lett 133:27-30, 1981
880
29. Nairn AC, Perry SV: Calmodulin and myosin light-chain kinase of rabbit fast skeletal muscle. Biochem J 179:89-97, 1979 30. Dedman JR, Potter J, Jackson RL, et al: Physicochemical properties of rat testis Ca*+-dependent regulator protein of cyclic nucleotide phosphodiesterase. Relationship of Ca*+-binding, conformational changes and phosphodiesterase activity. J Biol Chem 252:8415-8422,1977 3 1. Brown EM, Dawson-Hughes BF, Wilson RE, et al: Calmodulin in dispersed human parathyroid cells. J Clin Endocrinol Metab 53:1064-1071,1981 32. Roufogalis BD: Phenothiazine antagonism of calmodulin: A structurally-nonspecific interaction. Biochem Biophys Res Commun 98:607-613,1981 33. Scharff 0: Stimulating effects of monovalent cations on
DAWSON-HUGHES
ET AL
activator-associated states of Ca’-ATPase in human erythrocytes. Biochim Biophys Acta 512:309-317, 1978 34. Jones LR, Besch Jr HR, Watanabe AM: Monovalent cation stimulation of Ca*+ uptake by cardiac membrane vesicles. J Biol Chem 252:3315-3323,1977 35. Louis CF, Buonaffina R, Binks B: Effect of trypsin on the proteins of skeletal muscle sarcoplasmic reticulum. Arch Biochem Biophys 161:83-92, 1974 36. Brown EM, Adragna N, Gardner DG: Effect of potassium on PTH secretion from dispersed bovine parathyroid cells. J Clin Endocrinol Metab 53: 1304-I 306, 198 1 37. Rothstein M, Morrissey J, Slatopolsky E, et al: The role of Na/Ca exchange on PTH secretion. Endocrinology 111:225-230, 1982