J Mol

Cell Cardiol

E@ects

23, 1223-1230

(1991)

of Monoclonal Pump ATPase

Yoshihiro

Kim-

Antibody against Phospholamban of Cardiac Sarcoplasmic Reticulum

Makoto In* Tatsuya Sasaki

The First Department of Medicine

Masaaki Kadomn, Yosh+aki and Michihiko Tadat

and Department of Pathophysiology, Medicine, Osaka, Japan

(Received 18 June 1990, accepted in revisedform

on Calcium

Kijima,

Osaka University School of

16 October 1990)

Y. KIMURA, M. INUI, M. KADOMA, Y. KIJIMA, T. SASAKI AND M. TADA. Effects of Monoclonal Antibody against Phospholamban on Calcium Pump ATPase of Cardiac Sarcoplasmic Reticulum. Journal of Molecular and Cellular Cardiology (1991) 23,1223-1230. A monoclonal antibody against phospholamban has been reported to increasr Gas+ uptake by cardiac sarcoplasmic reticulum. We compared the effect of this antibody on Ca2+ pump ATPase activity of cardiac sarcoplasmic reticulum vesicles to the effect of CAMP-dependent phosphorylation of phospholamban. The antibody markedly stimulated the Ca 2’-dependent ATPase activity in parallel to the increase in Caa + uptake by cardiac sarcoplasmic reticulum. When the Ca 2+-dependent profile of the ATPase activity was compared, the K, was shifted from 1.24 to 0.62 PM by the antibody, whereas CAMP-dependent phosphorylation of phospholamban shifted the K, to 0.84 PM. When cardiac sarcoplasmic reticulum vesicles were treated with both CAMP-dependent protein kinase and the antibody, the stimulation was the same as that by the antibody. The with the antibody alone. Thus, the Ca’+ pump ATPase seems to be fully activated stoichiometry between Ca*+ uptake and ATPase rate was around 1 and no significant change was observed by reticulum the treatment with the antibody. Therefore, the stimulation of Ca*+ uptake of cardiac sarcoplasmic by the antibody occurred by the stimulation of Ca2+ pump ATPase, not by other mechanisms such as channel activity of phospholamban. These results indicate that the binding of the antibody to phospholamban produces essentially the same mode of action on Ca2+ pump ATPase as that of phospholamban phosphorylation. The antibody and phospholamban phosphorylation appear to release the inhibitory action of phospholamban on Ca2+ pump ATPase, resulting in the stimulation of Ca2+ pump. KEY

WORDS:

Sarcoplasmic

reticulum;

Ca”+

pump

ATPase;

Introduction Cardiac sarcoplasmic reticulum (SR) has a unique regulatory system of Ca2+ transport, in that the Ca’ + pump ATPase is under the control of another cardiac SR* protein phospholamban (Tada and Katz, 1982; Tada and Inui, 1983). Phosphorylation of phospholamban catalyzed by CAMP-dependent protein kinase markedly stimulates the ATPase activity of the Ca2+ pump, resulting in an increase in Ca2+ uptake by cardiac SR (Tada et al., 1974). Phospholamban is also phosphorylated by calmodulin-dependent protein kinase (Le Peuch et al., 1979) and by protein kinase C (Movsesian et al., 1984), increasing Ca2+ uptake by cardiac SR. The mechanism

Phospholamban;

CAMP;

Monoclonal

antibody.

of phospholamban regulation has been studied kinetically (Tada et al., 1979, 1980). Phospholamban was purified (Inui et al., 1985), and its primary structure was determined by cDNA cloning and sequencing (Fujii et al., 1987). Recently, a direct protein-protein interaction between Ca2+ pump ATPase and phospholamban was demonstrated by a cross-linking experiment (James et al., 1989). Although reconstitution of the Ca2 ’ pump ATPase-phospholamban system was also tried, the effect of phospholamban in the reconstituted system was different from that of native cardiac SR vesicles (Kim et al., 1990). The monoclonal antibody against phospho-

t Please address all correspondence to: Michihiko Tada, The First Department of Medicine, Osaka University School of Medicine, l-1-50 Fukushima, Fukushima-ku, Osaka 553, Japan. *Abbreviations: SR, sarcoplasmic reticulum; cDNA, complementary DNA; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid; CAMP-PK, CAMP-dependent protein kinase; mAb, monoclonal antibody. 0022-2828/91/

111223 +08

$03.00/O

0

1991 Academic

Press Limited

Y. Kimura

1224

lamban (mAbA1) has been reported to increase Ca* + uptake by cardiac SR (Suzuki and Wang, 1986). In the present study, we examined the mechanism of the stimulation of Ca* ’ uptake by mAbA1, compared to that of CAMP-dependent phosphorylation of phospholamban. We demonstrated that the stimulation of Ca* + uptake by mAbA1 is brought about by the stimulation of Ca2+ Pump ATPase in the same manner as phospholamban phosphorylation, but not by the modulaof Ca*+ channel activity of tion phospholamban which was observed in planar bilayers with purified phospholamban (Kovacs et al., 1988).

Materials

and Methods Materials

45CaC12 (I 11.1 M~/~mol) was purchased from New England Nuclear (Boston, MA), Miilipore filters (type GS, 0.22 pm) were from Nihon MiIIipore Kogyo (Yonezawa, Japan). Catalytic subunit of bovine cardiac CAMP dependent protein kinase, ATP, and phosphoenoIpyruvate were from Sigma (St Louis, MO). Pyruvate kinase from swine myocardium was purchased from Oriental Yeast Co. (Osaka, Japan). EGTA was from Dojin Chemicals (Tokyo, Japan). Hydroxylapatite was obtained from Bio Rad (Richmond, CA). Celglosser-H, a serum-free medium for hybridomas, was purchased from Sumitomo Pharmaceuticals (Osaka, Japan). Cardiac SR vesicles were prepared from dog heart according to the method of Chamberlain et al, (1983). In some experiments, cardiac microsomeswithout the final step of finear sucrose-dextran gradient centrifugation were used. Hybridomas producing mAbA1 (Al hybridomas) were a generousgift of Dr J. H. Wang of the University of Calgary. Al hybridomas were grown in Celglosser-H. Subcloned hybridoma cellswere injected intraperitoneally into Pristane-treated ~alb~c mice to induce ascites. Purification of antibody was performed employing hydroxylapatite column chromatography according to the methods described by Stanker et al. ( 1985). Ascites fluid was applied to the column after ammonium sulfate fractionation. Purified IgG fraction

et PI.

was concentrated by Centriprep 10 and Centricon 10 (Amicon, Danvers, MA). Purity and titer of the antibody was checked by enzymelinked immunosorbent assay and Western immunoblottitlg.

Pre-treatment

of cardiac microsomes

Phosphorylation of cardiac SR or microsomes was carried out at 25°C for 3 min in the reaction mixture containing 1 mg/ml SR or microsomes, 0.1 mg/ml catalytic subunit of CAMP dependent protein kinase, 20 mM Trismaleate-KOH (pH 6.8), 100 mM KGI, I mM MgCl,, and 1 mM ATP. The reaction was started by adding ATP. For the mAbA1 treatment, cardiac SR or microsomes (1 mg/ml) were incubated on ice for 15 min with 2 mg/ml of mAbA1 in 20 mM Tris-maleate-KOH (pH 6.8) and 100 mM KGI. In the combined treatment with CAMP-dependent protein kinase and mAbA1, cardiac SR or microsomeswere first phosphorylated by catalytic subunit of CAMP-dependent protein kinase, then incubated with mAbA1. After the re-treatments, aliquots were analyzed for CaP+ uptake activity and ATPase activity.

Assay for A TPase a~ti~i~

The Ca*+-dependent ATPase activity was determined as reported previously (Tada et al., 1983). Twenty-five micrograms of cardiac SR or microsomeswere added to give a final concentration of 50 fig/ml to the reaction mixture containing 20 mM Tris-maleateKOH, pH 6.8, IO0 mM KCl, 0.5 rnM MgCl,, 2.5 rnM oxalate-KOH, 5 mM NaN,, 0.1 mM ATP, 0.1 to 25 ,UMionized Ca2+ (GalEGTA buffer containing 0.5 mM CaCI,, and various amounts of EGTA), and ATP regenerating systemconsisting of 2.5 mM phosphoenolpyruvate and 20 IU/ml pyruvate kinase. The sample was incubated at 25°C for 3 min, and the reaction was stopped by adding 1.7 ml of the solution containing 0.3 m&i 2,4dinitrophenylhydrazine and 0.35 N HCl. The ATPase activity was determined by measuring the amounts of pyruvate liberated, as determined photometrically by the procedure of Reynard et al. (1961). The time course of the ATPase reaction was linear for at least 5 min.

Phospholutlbnn Basal ATPase activity was measured in reaction mixtures containing 5 mu EGTA without added Ca2+. In the latter condition, free Mg and Mg-ATP concentrations are calculated to be 2.27 x 10e4~ and 5.95 x ~O-‘IU, respectively, while free Mg and Mg-ATP concentrations are calculated to be 2.3 1 to 2.25 x low4 M and 5.96 to 6.08 x IO-’ M respectively, in the presence of Ca/EGTA buffer. The Ca2+dependent ATPase activity was determined as the difference between the total ATPase activity and the basal ATPase activity. Assay for calcium uptake Oxalate-facilitated 4sCa2+ uptake by cardiac SR or microsomes was determined under conditions essentially the same as those for ATPase assay. Twenty-five micrograms of SR or microsomes were added to give a final concentration of 50 pg/ml to the reaction mixture containing 20 mM Tris-maleate-KOH (pH 6.8), 100 mM KCl, 0.5 mM MgCl,, 2.5 mM oxalate-KOH, 5 mM NaNs, 0.1 mu ATP, 0.1 buffer, the to 25 PM ionized Ca 2+ (Ca/EGTA same as that for ATPase assay), 2.5 mM phosphoenolpyruvate, and 20 IU/ml pyruvate kinase. The reaction was started by adding the SR or microsomes. The sample was incubated at 25°C for 3 min and then the reaction was terminated by adding 2 ml of the stop solution which contained 20 mM Tris-maleate-KOH (pH 6.8), 100 mM KCI, 10 mM MgCI,, 3 mM EGTA. The sample was immediately passed through a Millipore filter (type GS, 0.22 pm), and the filter was washed four times with 2 ml of the stop solution. After drying, the Millipore filter was liquefied with 1 ml of acetone, 0.1 ml of which was transferred into solid scintillator, Ready-CapTM (Beckman Fullerton, CA), dried completely, and counted for In some experiments, Ca*+ radioactivity. uptake was determined by counting the filtrate without adding the stop solution (see caption, Table 2). Miscellaneous methods Protein concentrations were measured by the method of Lowry et al. (1951) with bovine serum albumin as standard. Ionized Ca2+ concentrations at pH 6.8 were calculated with apparent association constants of

I225

Reglhtion

9.93 X l@/M, 2.38 x lo/M, 3.99 x 103/M, 9.00 x 103/~, 1.00 x 103/~ and 3.55 x 102/~ for calcium-EGTA, magnesium-EGTA, magnesium-ATP, calciumcalcium-ATP, oxalate and magnesium-oxalate, respectively, by the computer programs described by Fabiato and Fabiato (1979). Results Effects of nAbA

on Ca’+-dependent activip

ATPase

To examine the effects of mAbA1 on the Ca2+-de P endent ATPase activity of cardiac SR, we determined the optimal condition for the pre-treatment of cardiac microsomes with mAbA 1. When cardiac microsomes were preincubated on ice with mAbA1 at the weight ratio of 2 to the microsomes, stimulation of Ca2 +-dependent ATPase activity was observed. The extent of the stimulation increased as pre-incubation time increased up to 10 min (Fig. 1, inset). Incubations longer

! / II(III 0

0. I

[mAbAl]/[CSR]

I

I I, !A--

:

I

L

(wh)

FIGURE 1. The effects of mAbAl on the Gas+dependent ATPase activity of cardiac microsomes; 25 pg of microsomes were pre-incubated on ice for 15 min with mAbA1 at the designated weight ratio [mAbAa]/ [CSR]) in 25 ~1 of mixture containing 20 mn Trismaieate-KOH, pH 6.8, and 100 mn KCI, and then the mixture was subjected to the ATPase assay as described in Materials and Methods. The Ca*+-dependent ATPase activity was determined at the Gas+ ion concentration of 1.2 L(M. Data are means of three different experiments. Inset: 25 pg of microsomes were pre-incubated with 50 pg of mAbA1 on ice for the designated duration in 25 ~1 of the medium containing 20 rnM Tris-maleate-KOH pH 6.8 and 100 my KCl, and then the ATPase assay was carried out at the Ca*+ concentration of 1.2 pm as described in Materials and Methods. Data are means of three different experiments.

1226

Y. Kimuta

FIGURE 2. The effects of CAMP-dependent phosphorylation and mAbA1 on Ca ‘+-dependent ATPase activity of cardiac microsomes. Cardiac microsomes (25 pg) pretreated with 2.5 ng of catalytic subunit of CAMPdependent protein kinase (0, n = 6), 50 pg of mAbA1 (B, a = 5), or both (Cl, n = 5), in 25 ~1 of medium as described in Materials and Methods. Then the mixture was subjected to the ATPase assay. In the control experiment (0, n = 9), cardiac microsomes were directly subjected to the ATPase assay. Data are means of different experiments.

et al.

stimulation of ATPase activity was observed in microsomes pre-treated with mAbA1. resulting in a further shift of the curve to the left. As shown in Table 1, a decrease in hca (Ca*+ concentration to attain half maximal ATPase activity) from 1.24 to 0.62 PM was observed for the pre-treatment with mAbA1, whereas Kc, decreased from 1.24 to 0.84 pM upon CAMP-dependent phosphorylation. When cardiac microsomes were first phosphorylated by catalytic subunit of CAMPdependent protein kinase and then incubated with mAbA1, the ATPase activity and its Ca* +-dependence w asalmost the sameasthat of the microsomes pre-treated with mAbA1 alone (Fig. 2). Effects of mAbA1 on Ca*+ uptake

It has been reported that Ca*+ uptake by cardiac microsomesis enhanced by addition of mAbA1 (Suzuki and Wang, 1986). To characterize this stimulation by mAbA1, we determined the Ca* + -dependent p rofile of Ca*+ than 10 min did not give any further signifi- uptake by cardiac microsomesunder the same cant changesin the ATPase activity. Stimula- conditions as the ATPase activity assayexcept tion of ATPase activity was alsodependent on for the presence of 45Ca2+ (Fig. 3). Essenthe amount of antibody added (Fig. 1). The tially, the sameresults were obtained as those of the ATPase activity (Fig. 2). Kc, of Ca*’ maximal effect was observed at antibodymicrosomes ratio (w/w) of higher than 2. uptake decreased from 1.7 to 0.87 PM for According to theseobservations, we chosethe mAbA1, and from I .7 to 1.3 PM for CAMPweight ratio of antibody to cardiac microsomes dependent phosphorylation. The microsomes of 2 and a pre-incubation time of 15 min on ice treated with mAbA1 after phosphorylation by as the pre-treatment condition in the follow- the catalytic subunit of protein kinase showed ing experiments. Under theseconditions, antibody had no effect on the Ca*+ pump ATPase of fast-twitch skeletal muscle SR (data not TABLE 1. Effects of CAMP-dependent phosphoryshown), indicating that the antibody doesnot lation of phospholambanand mAbA1 have any effects on our assaysystem for Ca*+ on K, of Ca*+-dependentATPaseacuptake and Ca *+-dependent ATPase activity. tivity in cardiac microsomes Effects of mAbA1 on the Ca*‘-dependent ATPase activity

projle

of

The Ca* +-dependent profile of ATPase activity was compared in cardiac microsomespretreated with mAbA1 or phosphorylated by CAMP-dependent protine kinase. As reported previously (Tada et al. 1979, 1983), the microsomespre-treated with a catalytic subunit of CAMP-dependent protein kinase exhibited stimulation of Ca*+-dependent ATPase activity especially at lower Ca*+ concentrations, shifting the curve to the left (Fig. 2). Further

Conditions

n

Control CAMP-PK mAbA 1 CAMP-PK

19 12 9 9

‘K, ATPase reciprocal Data ‘P < dP