D R Hunter, R A Haworth and H A BerkoffCellular calcium turnover in the perfused rat heart: modulation by caffeine and procaine.

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1982 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.51.3.3631982;51:363-370Circ Res. 

http://circres.ahajournals.org/content/51/3/363World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from by guest on October 18, 2013http://circres.ahajournals.org/Downloaded from

363

Cellular Calcium Turnover in the Perfused Rat Heart:

Modulation by Caffeine and Procaine

Douglas R. Hunter, Robert A. Haworth, and Herbert A. BerkoffFrom the Department of Surgery, Clinical Science Center, University of Wisconsin, Madison, Wisconsin

SUMMARY. Pool A is a rapidly exchangeable cellular pool of Ca++ whose release is triggered by amechanism involving extracellular Ca++ (Hunter et al., 1981). We have now found that pool A israpidly released from the perfused rat heart when 10 mM caffeine is added to the perfusate. Pool Arelease by caffeine was demonstrated during a Ca++ free perfusion. When the perfusate contained2.5 mM Ca++, caffeine induced an immediate contractile failure. The steady state level of pool A(normally 69 ± 10 nmol Ca+ +/g wet wt heart) was also decreased by 60%. Pool A was similarlydepleted in hearts perfused with medium containing 0.2 mM Ca++. Procaine (5 mM) inhibited by 50%the release of pool A triggered by caffeine and inhibited by 86% the release triggered by extracellularCa++. The inhibition of Ca++-induced release of pool A by procaine was partially relieved byexternally stimulating the hearts. External stimulation also decreased the inhibition by procaine ofCa++ uptake by pool A from 83% to 28%. These results are further evidence that pool A is locatedin the sarcoplasmic reticulum and that release of pool A to the myofibriles is triggered by excitation-dependent Ca++ influx. (Ore Res 51: 363-370, 1982)

IN a previous publication, we reported a method formeasuring rapidly exchangeable cellular Ca++ in theperfused, beating rat heart (Hunter et al., 1981). Itentailed labeling the heart with 45Ca++ for a 5-minuteperiod, washing away extracellular label by perfusingthe heart in a cold room for 16 minutes, and finallyreperfusing the heart at 37°C to selectively releasecellular 45Ca++. Approximately 125 nmol of exchange-able cellular Ca+ +/g wet wt of heart was found, andit exchanged rapidly with perfusate Ca++, with a half-time of less than 1 minute. In addition, a fraction(38%) of the measured cellular Ca++ was found torequire extracellular Ca++ for release. We termed thatfraction pool A and concluded that its location maybe the sarcoplasmic reticulum (SR). Furthermore, weacknowledged the possible involvement of pool A inthe contractile mechanism, since its release to themyoplasm may be triggered by influx of extracellularCa++. Such a Ca++-induced Ca++ release mechanismhas been studied in both skeletal and cardiac skinnedmuscle cell preparations (Endo et al., 1970; Fabiatoand Fabiato, 1975), and is the basis for an hypothesisfor excitation-contraction coupling (Endo, 1977; Fa-biato and Fabiato, 1977). A case for the importance ofpool A has therefore been presented. However, todetermine conclusively the exact function and loca-tion of pool A requires more detailed experimentationon how pool A responds to different changes in thefunctional state of the heart. The present work beginsthis task: The modulation of pool A by two potentcardioactive reagents, caffeine and procaine, is de-scribed.

MethodsLabeling of the Heart with 45Ca++

Retired female breeder rats (Sprague-Dawley) were anes-thetized by intraperitoneal injection of thiamylal (100 mg/

kg). The heart was excised, and cannulas were placed in theaorta and pulmonary vein. Incisions were made in thepulmonary artery and the right atrium to ensure adequatedrainage. The basal perfusate contained 118 mM NaCl, 25mM NaHCO3, 4.7 mM KC1,1.2 mM KH2PO4/1.2 mM MgSCX,,11 mM glucose, and 2.5 mM CaCl2. It was gassed with 95%O2 plus 5% CO2, and warmed to 37°C. The pH was 7.4.The heart was perfused for 5 seconds through the pulmo-nary vein cannula to wash out any air bubbles. Then it wasperfused through the aorta. The height of the reservoir was80 cm above the heart, and the flow rate, which varied fordifferent rat hearts, was between 6 and 12 ml/min. Thespontaneous rate of beating was 195 ± 10. After 5 minutesof perfusion, the heart was switched to a low Ca++ perfusatewhich was identical to the basal perfusate, except Ca++ was0.2 mM. After 30 seconds of perfusion with the low Ca++

perfusate, a solution of 1 mM Ca++ containing 45Ca++ (2 X106 ml) was delivered (20 ml/hr) by an infusion pump tothe perfusate 4 cm above the heart. The heart was labeledfor 3 minutes and the effluent perfusate collected. Analiquot of this was taken to measure the specific activity ofthe perfusate 45Ca++. Ventricular pressure was measured bya ventricular balloon connected by a polyethylene catheterto a Statham pressure transducer (Rusy and Coulson, 1973).The balloon was made from a vinyl examination glove andwas inserted through a slit made in the left atrium. It wasfilled with water to 0.15 ml at a filling pressure of 27 mmHg and was secured by tying the balloon catheter to theaortic cannula. For some experiments, the heart was per-fused with the basal perfusate in the working mode accord-ing to the method of Neeley et al. (1967). The height of theatrial reservoir was maintained at 35 cm above the heartand aortic pressure was monitored. The heart rate duringworking mode perfusion averaged 237 ± 18.

Washout of 45Ca++ from Heart

Immediately after completion of the labeling period, theheart was switched to a perfusate identical to the basalperfusate except that Ca++ was absent, 0.5 mM EGTA(ethylene glycol bis(/?-aminoethyl ether) N,/V'-tetraacetic

364 Circulation Research/Voi. 51, No. 3, September 1982

acid) and 20 ITIM sucrose were present (EGTA wash solu-tion). In addition, the temperature of the perfusate was23°C. The heart was perfused for 5 seconds through thepulmonary vein cannula to wash out any radioactivitytrapped in the heart chamber; then, for the remainder ofthe washout period, it was perfused through the aorta. Atthe appropriate times shown in the text, the perfusate waschanged to one of the Ca++-releasing perfusates. They wereeither the basal perfusate (containing the standard level of2.5 mM Ca++) or the EGTA wash solution containing caf-feine (either 2 or 10 mM). The effluent throughout thewashout period was collected each minute (or each lh min-ute) and the radioactivity was determined by scintillationcounting. The amount of labeled Ca++ released each minutethen was calculated by dividing the radioactivity in eachsample by the specific activity of the perfusate. After per-fusion was completed, the heart was cut in half: One halfwas dried overnight in an oven at 110°C and weighed.Then, by using the correction factor for the average meas-ured value of 5.6 g wet weight of perfused tissue per g dryweight, all data were expressed in terms of g wet weight ofperfused heart. The second half of the heart was used toassay the amount of 45Ca++ remaining in the heart. It washomogenized with 14 ml of basal perfusate and incubatedfor 10 minutes at 30°C with 5 /IM A23187 to release anystored 45Ca++ (Hunter et al., 1979). The homogenate thenwas centrifuged at 1300 g and the radioactivity in thesupernatant extract was measured.

Statistical MethodsThe mean and standard deviation (SD) of measurements

are reported. Significance of the difference between groupswas determined by the unpaired Student's t-test. Each groupcontained a minimum of four rat hearts.

Results

Release of Pool A by Ca++

In the previous paper (Hunter et al., 1981), themethod for measuring pool A was indirect: Afterextracellular 45Ca++ had been released by cold per-fusion, cellular 45Ca++ was released at 37°C witheither a perfusate containing 2.5 mM Ca++ or onewhich contained no Ca++ but contained 0.5 mMEGTA. The difference in Ca++ release induced bythose two solutions was equal to pool A, and wasapproximately 50 nmol Ca++/g wet wt heart. A directmethod for assaying pool A was developed as follows:After the heart was labeled with 45Ca++, extracellularlabel was washed away by perfusing the heart at 23°Cwith the EGTA wash solution (see Methods) for 6minutes. This wash also removed any cellular 45Ca++

which did not require extracellular Ca++ for release(termed pool B in Hunter et al., 1981). Then uponperfusing with a medium containing 2.5 mM Ca++, asharp peak of 45Ca++ release was triggered (Fig. lA).Importantly, the present data were taken from heartslabeled with perfusate [Ca++] of only 0.2 mM. Thatchange increased the accuracy of the measurementbecause the size of pool A in comparison to theconcentration of extracellular Ca++ was found to in-crease as the perfusate [Ca++] was lowered. Theamount of pool A was determined by measuring thesize of the peak and subtracting from it the baseline

o

15

10

5

0

10 12 14 16 18 20

10 12 14 16 18 20+ Caffeine r(lOmM) L

1

8 10 12 14 16Time of Washout (min)

18 20

FIGURE 1. Release of pool A by Ca ++, caffeine, and A23187. Sixgroups of four hearts labeled with >sCa ++ (0.2 min) were perfusedwith the EGTA wash solution. A: The hearts of 1 group wereswitched at 6 minutes (O) and the hearts of a second group wereswitched at 16 minutes (9) to the Ca ++-containing basal perfusate.B: The hearts of two groups were switched at 6 minutes to thecaffeine-containing EGTA wash solutions as shown. After 6 min-utes, the hearts in both groups were switched to the Ca ++-confa/n-ing basal perfusate. C: Beginning at 6 minutes, A23187 (dissolvedin ethanol at 0.5 mg/ml) was infused into the EGTA wash solution4 cm above the hearts of 1 group (O). For a second group of hearts,ethanol alone was infused into the EGTA wash solution ( • ) . After6 minutes, the infusion was stopped and the hearts of both groupswere switched to the EGTA wash solution containing 20 mMcaffeine. Shown are the mean ± SD of K'Ca++ washout from thehearts.

45Ca++ washout which occurred with the EGTA washsolution. An average value of 23 ± 5 nmol Ca++/gwet wt heart for pool A was obtained. This amount ofCa++ is considerably smaller than the 50 nmol/greported in the previous paper, which is consistentwith the fact that these hearts were Ca++ deficient.When this direct method for pool A measurementwas used on hearts labeled with a perfusate [Ca++] of2.5 mM, a value for pool A of 69 ± 10 nmol/g wet wtheart was obtained (Fig. 2).

The success of this direct method for measuringpool A was dependent on minimizing the damagingeffects of the "Ca++ paradox" (Zimmerman and Huls-mann, 1966). This was accomplished by using 23°Cinstead of 37°C for the washout temperature (Diger-ness et al., 1980) and by having 20 mM sucrose present

Hunter et ai./Ca++ Turnover in Perfused Rat Heart

60

0

c'e

nmoL

leas

ed (

a>

CVJ

Om

50

40

30

20

10

—

-

-

• 1(2.5

1

i2 +

ImM)

5 6 7 8 9 10 IITime of Washout (min)

FIGURE 2. Effect of caffeine on the steady state /eve/ of pool A.Using a retrograde perfusion through the aorta, three groups offour hearts were labeled with AiCa** (2.5 mM) for 3 minutes. Thelabeling perfusate was either: the basal perfusate ( • ) , the basalperfusate containing 10 mM caffeine (O), or the basal perfusatecontaining 2 mM caffeine (A). After labeling, the hearts wereperfused for 6 minutes with the ECTA wash solution, at whichtime they were switched to the Ca**-containing basal perfusate.Shown are the mean ± SD of 45Ca ** washout from the hearts.

365

in the EGTA wash solution. The results from a rep-resentative experiment are given in Figure 3A. Tenseconds after switching the heart to the Ca++-contain-ing basal perfusate, ventricular contractile activityresumed fitfully and continued for approximately 40seconds at which time spontaneous excitation waslost. Interestingly, 45Ca++ release (pool A) lagged sig-nificantly behind the transient contractile activity, asrelease continued for well over 1 minute. Importantly,coronary flow remained good throughout. For sevenexperiments, it averaged 7 + 2 ml/min for the 6-minute period immediately following readdition ofCa++. In addition, contractile activity which was lostfollowing readdition of Ca++ at 23°C, readily returnedif the perfusion temperature was raised again to 37°C.Therefore, damage caused by the "Ca++ paradox"was not excessive. However, it was important to useas brief a period of EGTA wash as possible beforereading Ca++. If, instead of 6 minutes, a 16-minutewashout period with EGTA was used, the resultingrelease peak of pool A was flattened out considerablyand not all the available Ca++ was released (Fig. 1A).Only 12 ± 3 nmol of Ca+ +/g was released. In addition,after completion of the washout period, 9 ± 2 nmol/g of 45Ca++ remained in the hearts in which Ca++ wasadded back after 16 minutes, as compared to 5 ± 1nmol/g remaining in hearts in which Ca++ was addedback after 6 minutes.

Release of Pool A by Caffeine

Caffeine was found to be as effective as Ca++ forreleasing pool A. After a 6-minute perfusion periodwith EGTA wash solution, a group of four hearts wasswitched to the EGTA wash solution containing 10mM caffeine. An extremely rapid release of 45Ca++

occurred (Fig. IB). The amount of Ca++ in that peakwas 18 ± 4 nmol/g wet wt heart, which is not signifi-cantly different (P > 0.1) from the 23 ± 5 nmol/greleased by 2.5 mM Ca++ (Fig. 1A). That caffeine andCa++ released the same pool of 45Ca++ (pool A) wasdemonstrated by the fact that readdition of Ca++ at12 minutes to the hearts treated with 10 mM caffeine,

FIGURE 3. Contracri/e activity during release ofpool A byby Ca ** and caffeine. A ventricular bal-loon was placed in two hearts as described inMethods. They were labeled for 3 minutes with4bCa * * during perfusion with the perfusate con-taining 0.2 mM Ca **. Pool A release was inducedin one heart by Ca** (A) and by caffeine (B) inthe other, by the methods described in the legendto Figure J. Ventricular pressure was monitoredas described in Methods.

Washout Time (min)

366 Circulation Research/Voi. 51, No. 3, September 1982

resulted in just 1 ± 1 nmol/g of additional 45Ca++

release. If a lower concentration of caffeine was used,only part of pool A was released. Two mM caffeinereleased 11 ± 3 nmol Ca++/g wet wt heart. That valueis significantly different from the 23 ± 5 nmolCa++/g released by 2.5 mM Ca++ (P < 0.01). If then,Ca++ was readded at 12 minutes, an additional 7 ± 2nmol Ca++/g was released from the 2 mM caffeine-treated hearts. The release of pool A by caffeine isaccompanied by a ventricular contraction. For a rep-resentative experiment, shown in Figure 3B, the con-traction occurred just 5 seconds after switching theheart to the perfusate containing 10 mM caffeine.45Ca++ release (pool A), measured on the same heart,also began soon after the addition of caffeine, andcontinued for an additional 30 seconds after the re-laxation period was complete (Fig. 3B). Significantly,the contraction induced by caffeine occurred in theabsence of extracellular Ca++ and even when pool Ahad been partially depleted during the labeling periodby perfusion with a medium containing only 0.2 mMCa++

Insensitivity of Pool A to A23187

Infusion of an ethanolic solution of A23187 at 15jug/min into the EGTA wash solution of 45Ca++ la-beled hearts was begun at 6 minutes. The averageflow rate was 11 ± 2 ml/min, and the average perfu-sate concentration of A23187 therefore was 2.5 ± 0.4JUM. That concentration of A23187 produced a slow(<2-fold) stimulation of 45Ca++ release when com-pared to control hearts treated with the ethanol alone(Fig. 1C). A total of 8 ± 3 nmol Ca++/g wet wt heartwas released in 6 minutes by A23187. Addition ofcaffeine (at 12 minutes) to the hearts then induced arapid release of an additional 10 ± 1 nmol Ca++/gfrom the A23187-treated hearts and 12 ± 1 nmolCa++/g from the ethanol-treated control hearts. WhenA23187 (at 15 jug/min) was infused into the basalperfusate of the working rat heart, a slowly developingcontractile failure occurred. It took 2.4 ± 0.6 min forthe peak aortic pressure to fall by 50%.

Effect of Caffeine on the Steady State Level ofPool A

Addition of 10 mM caffeine to the basal perfusateof the working rat heart induced a rapid contractilefailure (Fig. 4A). A 50% loss of peak aortic pressuretook just 3 ± 1 sec (n = 4). Contractile failure wasalso rapidly induced by 2 mM caffeine (Fig. 4B). Therea 50% loss of peak aortic pressure took 5 ± 1 sec (n= 4). Perfusion of those two groups of caffeine-treatedhearts was continued with a retrograde flow throughthe aorta. For 3 minutes they were labeled with 45Ca++

in the presence of caffeine and the steady state levelof pool A was measured. Pool A was released by Ca++

after a 6-minute washout period with the caffeine-free EGTA wash solution (Fig. 2). For the hearts whichhad 10 mM caffeine in the labeling perfusate, pool Acontained 28 ± 10 nmol Ca++/g wet wt heart. Forthose which had 2 mM caffeine present, pool A con-

tained 28 ± 12 nmol Ca++/g. Both values are signifi-cantly different (P < 0.01) than the 69 ± 10 nmolCa++/g found in pool A of control hearts.

Inhibition of Pool A Release by Procaine

Procaine did not induce pool A release. Instead, itinhibited the release induced by Ca++ (Fig. 5) and bycaffeine (Fig. 6). With 5 mM procaine present, Ca++

released just 3 ± 2 nmol 45Ca++/g, while caffeinereleased 8 ± 2 nmol 45Ca++/g. Each value is signifi-cantly different (P < 0.01) than its non-procaine-treated control: Ca++ released 21 ± 4 nmol 45Ca++/g(Fig. 5) and caffeine released 16 ± 4 nmol 45Ca++/g(Fig. 6). The blockage by procaine of pool A releasewas reversible. At 12 minutes, the basal perfusate(without procaine) was perfused through the hearts.An additional 18 ± 2 nmol 45Ca++/g then was releasedfrom the Ca++ plus procaine-treated hearts (Fig. 5)and 11 ± 3 nmol 45Ca++/g was released from thecaffeine plus procaine-treated hearts (Fig. 6).

Role of Excitation in the Release of Pool A ByCa++

Concentrations of procaine as low as 0.5 mM stillcaused considerable inhibition of the Ca++-inducedrelease of pool A, but no significant inhibition of thecaffeine-induced release. Procaine (0.5 mM) producedan immediate blockage of the action potential whengiven to hearts during the 6-minute EGTA washperiod. The inhibitory effect of procaine on excitationis well known (Blinks et al., 1972; Josephson andSperelakis, 1976; Kohlhardt et al., 1972; Thorens,1971; Weidmann, 1955). When the 0.5 mM procaine-treated hearts were switched to the Ca++-containingbasal perfusate, spontaneous excitation transiently re-sumed, accompanied by weak ventricular contractionsand a slow release of pool A (data not shown). Incomparison, the action potential of hearts, treatedwith high levels of procaine (5 mM) did not recoverwhen Ca++ was added, and little if any release of poolA occurred (Fig. 5). If, however, the hearts treatedwith 5 mM procaine were externally stimulated afterCa++ addition, release of pool A took place (Fig. 5):14 ± 2 nmol 45Ca++/g was released in 6 minutes, andonly 7 ± 2 nmol 45Ca++/g was released following theremoval of procaine. Importantly, stimulation in theabsence of Ca++ caused no release of pool A evenwhen procaine was not present in the perfusate. Nei-ther was there an effect of stimulation on the caffeine-induced release of pool A from procaine-treatedhearts.

Release of Pool A by Ca++ Potentiated by High K+

Inhibition of the action potential by elevating theperfusate [K+] from 5.9 to 25.9 mM caused little if anyinhibition of pool A release by Ca++ (Fig. 7). In fact,high K+ was found to partially relieve the blockageby procaine of pool A release by Ca++ (Fig. 7). Ca++

released 16 ± 5 nmol 45Ca/g when procaine was 5 mMand [K+] was 25.9 mM. That is significantly different(P < 0.01) than the 3 ± 2 nmol 45Ca++/g released

Hunter et a/./Ca++ Turnover in Perfused Rat Heart 367

FIGURE 4. Effect of caffeine on aortic pressure. Atthe point indicated by the arrows, the hearts wereswitched from the basal perfusate to either thebasal perfusate containing 10 HIM caffeine (A) orthe basal perfusate containing 2 mM caffeine (B).

when procaine was 5 mM and [K+] was 5.9>mM (Fig.5). However, when [K+] was 25.9 mM, the rate ofrelease of pool A from procaine-treated hearts wasconsiderably slower than the release rate from non-treated hearts.

Inhibition by Procaine of Ca++ Uptake into Pool A

Procaine (1.5 mM) added to the perfusate of workingrat hearts produced a fairly rapid heart failure: Therewas an initial drop in peak aortic pressure of 40 ± 3%

14

c"EoEc<D</>O

|

+CM

OOm

I 0

8

0

tCa2 +

(2.5mM)

8 10 12 14Washout Time (min)

16

FIGURE 5. Inhibition by procaine of pool A release by Ca+*. Threegroups of four hearts, labeled with ibCa + + (0.2 mM), were perfusedwith the EGTA wash solution. One group (0) was switched at 6minutes to the basal perfusate. The remaining two groups of heartswere switched to the EGTA wash solution containing 5 mM procaineat 4 minutes, and then switched again at 6 minutes to the basalperfusate containing 5 mM procaine. One group (A) was externallystimulated between 6 and 10 minutes at 2 Hz, with 10 V and witha pulse wide of 50 msec. At 12 minutes, both procaine-treatedgroups of hearts were then switched to the basal perfusate. Shownare the mean ± SD of tsCa ++ washout from the hearts.

RemoveProcaineand AddCa2+(2.5mM)

8 i i iWashout Time (min)

FIGURE 6. Inhibition by procaine of pool A release by caffeine. Twogroups of four hearts, labeled with 45Ca** (0.2 mM), were perfusedwith the EGTA wash solution. One group (9) was switched to theEGTA wash solution containing 10 mM caffeine at 6 minutes. Thesecond group of hearts (O) was switched to the EGTA wash solutioncontaining 5 mM procaine at 4 minutes, and was switched again at6 minutes to the EGTA wash solution containing 10 mM caffeineplus 5 mM procaine. Both groups of hearts then were switched tothe basal perfusate at 12 minutes. Shown are the mean ± SD of ASCawashout from the hearts.

368 Circulation Research/Vol. 51, No. 3, September 1982

8 10 12 14Washout Time (min)

16

FIGURE 7. Effect of K* on pool A release by Ca*+. Two groups offour hearts, labeled with 45Ca *+ (0.2 mM), were perfused with theECTA wash solution. At 4 minutes, one group (0) was switched toa high K+, ECTA wash solution (the EGTA wash solution with 20mmol/liter NaCl replaced with 20 mmol/liter KCI). Then, at 6minutes, it was switched to a high K+, basal perfusate (the basalperfusate with 20 mmol/liter NaCl replaced with 20 mmol/literKCI). The other group of hearts (O) was switched at 4 minutes tothe high K+, EGTA wash solution containing 5 mM procaine. At 6minutes they were then switched to the high K+, basal perfusatecontaining 5 mM procaine. Shown are the mean ± so of t5Ca ++

washout from the hearts.

by 11 ± 2 sec followed by a total block of ventricularexcitation at 57 ± 6 sec. Ca++ uptake by the procainepoisoned hearts then was measured and compared tocontrol hearts: A retrograde perfusion through theaorta was carried out and the 45Ca++ uptake during a30-second labeling period was measured. The perfu-sate [Ca++] was 2.5 mM during labeling. To measurepool A 45Ca++ uptake, the standard release assay forpool A by Ca++ was performed on the hearts (Fig. 8).The control hearts contained 53 ± 19 nmol 45Ca++/gin pool A, while the procaine-treated hearts had only9 ± 3 nmol 45Ca++/g in pool A. Those values aresignificantly different with P < 0.01. 45Ca++ uptakeby a third group of hearts was also measured. Theywere also poisoned with 1.5 MM procaine, but duringthe labeling period, they were externally stimulated.Contractile activity resumed upon stimulation andalso the entry of perfusate 45Ca++ into pool A wasgreatly increased to 38 ± 9 nmol 45Ca++/g (Fig. 8).That value is significantly different (P < 0.01) fromthe 9 ± 3 nmol 45Ca++/g for pool A uptake by non-stimulated, procaine-treated hearts, but it may not besignificantly different (P > 0.1) from the uptake of 53± 19 nmol 45Ca++/g by control hearts.

Discussion

In our previous report, we postulated that pool Awas located in the SR (Hunter et al., 1981). The

finding presented here that caffeine stimulated andprocaine inhibited release of pool A supports thatproposal. Caffeine (5-10 mM) has been shown toenhance the rate of Ca++ release from SR isolatedfrom both skeletal muscle (Johnson and Inesi, 1969;Katz et al., 1977) and cardiac muscle (Pretorius et al.,1969; Blayney et al., 1978). Also, caffeine has beenpostulated to decrease the amount of Ca++ in the SRof cardiac muscle (Blinks et al., 1972; Henderson etal., 1974; Chapman and Leoty, 1976). In addition, thelocal anesthetics, procaine and tetracaine, have beenpostulated to inhibit the release of Ca++ from SR ofskeletal muscle (Aimers and Best, 1976; Caputo et al.,1979; Liittgau and Oetliker, 1968) and cardiac muscle(Blinks et al., 1972; Chapman and Miller, 1974).

Ca++ influx through the sarcolemma is enhancedby excitation (Chapman and Ellis, 1977; Niedergerke,1963). The physiological mechanism of pool A releaseappears to involve that process: External stimulationpromoted Ca++-dependent pool A release from heartswhose spontaneous action potential had been blockedby procaine. It is known that excitation enhances Ca++

6 7 8 9Washout Time (min)

10

FIGURE 8. Inhibition by procaine of Ca** uptake by pool A. Usinga retrograde perfusion through the aorta, three groups of fourhearts were labeled with "Ca** (2.5 mM) for 30 seconds. Thelabeling perfusate was the basal perfusate (9) or the basal perfusatecontaining 5 mM procaine (O, A). One group (A) was externallystimulated (2 Hz, with 10 V with a 50-msec pulse width) during thelabeling period. After labeling, the hearts were perfused for 6minutes with the ECTA wash solution, at which time they wereswitched to the basal perfusate. Shown are the mean ± so of **Ca +*washout from the hearts.

Hunter et a/./Ca++ Turnover in Perfused Rat Heart 369

influx by depolarizing the sarcolemma (Beeler andReuter, 1970). Elevating extracellular K+ also depolar-izes the sarcolemma and promotes Ca++ influx (Hodg-kin and Horowicz, 1960; Chapman, 1973). As ex-pected high K+ also stimulated Ca++-dependent poolA release.

The site of action of caffeine is probably directlyon the SR (pool A) (Endo, 1977). However, procainemost likely has multiple sites of inhibition. It blockedexcitation probably by interacting with the sarcolem-mal Na++ channel (Deitmer and Ellis, 1980; Josephsonand Sperelakis, 1976; Kohlhardt et al., 1972; Thorens,1971; Weidmann, 1955). A second site of action forprocaine is most likely the SR (Aimers and Best, 1976;Caputo et al., 1979). The evidence that procaine di-rectly effects pool A is the following: (1) procaineinhibited the release of pool A by caffeine, indepen-dent of external stimulation; (2) the inhibition ofCa++-induced release of pool A by procaine was onlypartially relieved by external stimulation; and (3)procaine partially inhibited the release of pool A byCa++ even when excitation had been blocked by highperfusate [K+]. An alternative explanation for (2) and(3) is that there is a third site of procaine inhibition,which is the sarcolemmal Ca++ channel (Josephsonand Sperelakis, 1976; Thorens, 1971).

Based on the following lines of evidence, it appearslikely that pool A is supplying Ca++ for the contractileproteins during excitation. (1) Pool A is of sufficientsize (69 ± 10 nmol/g wet wt heart) to completelysaturate the myofibrils (Solaro and Briggs, 1974); (2)It is an active pool in that it turns over rapidly with ahalf-time of <1 minute. Seventy-seven percent ofpool A (53 ± 19 nmol/g) was labeled by a 30-secondperfusion period. (3) The release of pool A probablyinvolves triggering by Ca++, which is a mechanismpostulated to occur during excitation-contraction cou-pling (Endo, 1977; Fabiato and Fabiato, 1978). ThatCa++-induced Ca++ release is the normal mechanismfor pool A release rather than Ca++-Ca++ exchange(Takakuwa and Kanazawa, 1981) is supported by thefact that caffeine releases pool A as effectively as doesCa++: the release triggered by caffeine cannot resultfrom Ca++-Ca++ exchange, but must involve net Ca++

release. (4) Pool A can be rapidly mobilized to pro-duce a contraction: release of pool A, triggered byeither Ca++ or caffeine, is associated with contractileactivity. Significantly, the caffeine-induced contrac-tion at 23°C occurred in the absence of extracellularCa++, and within a time span of just 5 seconds aftercaffeine was added to the whole heart. (5) Conditionswhich decrease the size of pool A by some 60% inducecardiac failure. Those conditions are adding caffeine(2 or 10 mMJ to the perfusate or lowering the perfusateCa++ from 2.5 to 0.2 mM. Relevant to this point is thestudy of Chapman and Leoty (1976). They demon-strated that caffeine potently depressed the strengthof contraction of trabeculae isolated from mammalianhearts and postulated that caffeine depleted a pool ofactivator Ca++ located in the SR. Refilling of that poolwas rapid {\}h = 30 sec) and critically dependent onthe extracellular [Ca++]. The authors concluded, how-

ever, that the activator pool was refilled by Ca++

contained in a second intracellular compartmentrather than Ca++ from the extracellular space. Ourresults suggest that their activator pool of Ca++ ismost likely pool A and that it is refilled by extracel-lular Ca++ rather than a second internal Ca++ store.

We have reported here only on the effects of caf-feine on the rapidly exchangeable pool A Ca++. Noattempt was made to measure how caffeine affectscellular pools of Ca++ which may turn over with muchslower half times (Guthrie and Nayler, 1967; Shineand Langer, 1971; Jandt et al., 1975). We were suc-cessful in studying pool A because it could be releasedselectively by either Ca++ or caffeine. Of some sur-prise is the fact that A23187 was only weakly activein releasing pool A. The concentration of A23187 weused (2.5 /HM) does rapidly release Ca++ from isolatedSR and mitochondria (Scarpa et al., 1972; Reed andLardy, 1972). However, A23187 required a consider-able incubation period to produce failure of the work-ing rat heart: It took 2.4 ± 0.6 min for a 50% loss inpeak aortic pressure to occur, while—for compari-son—caffeine induced a 50% loss in just 3 ± 1 sec.

Caffeine-releasable cellular Ca++ appears to bepresent in many types of excitable tissue. Not onlydoes caffeine mobilize cellular Ca++ in cardiac andskeletal muscle (Fabiato and Fabiato, 1975; Endo etal., 1970), but it does so also in smooth muscle (Itohet al., 1981; Haensler et al., 1981) and in non-muscletissue as demonstrated with sympathetic neurones(Kuba, 1980). If the caffeine-sensitive Ca++ pool inthe rat heart (pool A) is SR, then a Ca++ compartmentlike SR may be present in other excitable tissuesensitive to caffeine (Kuba, 1980). In smooth muscle,noradrenaline is active in releasing cellular Ca++ (vanBreemen and Siegel, 1980; Casteels and Droogman,1981). Casteels and Droogman (1981) estimated thesize of the noradrenaline releasable Ca++ by a methodsimilar to the one reported here for measuring poolA. Significantly, it is identical in size to pool A.

This work was supported by a grant from the Surgical Associatesof the University of Wisconsin Hospitals; also by a Crant-in-Aidfrom the American Heart Association and with hinds contributedin part by the Wisconsin Affiliate.

Address for reprints: Douglas R. Hunter, Department of Sur-gery, Clinical Science Center, University of Wisconsin, Madison,Wisconsin 53792.

Received March 18, 1982; accepted for publication June 18,1982.

ReferencesAimers W, Best PM (1976) Effects of tetracaine on displacement

currents and contraction of frog skeletal muscle. J Physiol (Lond)262: 583-611

Beeler BW, Reuter H (1970) The relation between membranepotential, membrane currents and activation of contraction inventricular myocardial fibres. J Physiol (Lond) 207: 211-229

Blayney L, Thomas H, Muir J, Henderson A (1978) Action ofcaffeine on calcium transport by isolated fractions of myofibrils,mitochondria, and sarcoplasmic reticulum from rabbit heart. CircRes 43: 520-526

Blinks JR, Olson CB, Jewell BR, Braveny P (1972) Influence of

370 Circulation Research/Vo/. 51, No. 3, September 1982

caffeine and other methylxanthines on mechanical properties ofisolated mammalian heart muscle. Circ Res 30: 367-392

Caputo C, Vergara J, Bezanilla F (1979) Local anaesthetics inhibittension development and Nile blue fluorescence signals in frogmuscle fibres. Nature 277: 400-402

Casteels R, Droogmans G (1981) Exchange characteristics of thenoradrenatine-sensitive store in vascular smooth muscle cells ofrabbit ear artery. J Physiol (Lond) 317: 263-279

Chapman RA (1973) The ionic dependence of the strength andspontaneous relaxation of the potassium contracture induced inthe heart of the frog Rana pipiens. J Physiol (Lond) 231: 209-232

Chapman RA, Ellis D (1977) Uptake and loss of manganese fromperfused frog ventricles. J Physiol (Lond) 272: 355-366

Chapman RA, Leoty C (1976) The time-dependent and dose-de-pendent effects of caffeine on the contraction of the ferret heart.J Physiol (Lond) 256: 287-314

Chapman RA, Miller DJ (1974) The effects of caffeine on thecontraction of the frog heart. J Physiol (Lond) 242: 589-613

Deitmer JW, Ellis D (1980) The intracellular sodium activity ofsheep heart purkinje fibres: effects of local anaesthetics andtetrodotoxin. J Physiol (Lond) 300: 269-282

Digerness SB, Shragge BW, Blackstone EH, Conti VR (1980) Tem-perature dependence of the calcium paradox in the isolatedworking rat heart: discrepancy between functional recovery andenzyme release. J Mol Cell Cardiol 12: 511-517

Endo M (1977) Calcium release from the sarcoplasmic reticulum.Physiol Rev 57: 71-108

Endo M, Tanaka M, Ogawa Y (1970) Calcium induced release ofcalcium from the sarcoplasmic reticulum of skinned skeletalmuscle fibres. Nature 228: 34-36

Fabiato A, Fabiato F (1975) Dependence of the contractile activationof skinned cardiac cells on the sarcomere length. Nature 256: 54-56

Fabiato A, Fabiato F (1977) Calcium release from the sarcoplasmicreticulum. Circ Res 40: 119-129

Guthrie JR, Nayler WG (1967) Interaction between caffeine andadenosine on calcium exchangeability in mammalian atria. ArchIntern Pharmacol Ther 170: 249-255

Haeusler G, Richards, JG, Thorens S (1981) Noradrenaline contrac-tions in rabbit mesenteric arteries skinned with saponin. J Physiol(Lond) 321: 537-556

Henderson AH, Brutsaert DL, Forman R, Sonnenblick EH (1974)Influence of caffeine on force development and force-frequencyrelations in cat and rat heart muscle. Cardiovasc Res 8: 162-172

Hodgkin AL, Horowicz P (1960) Potassium contracrures in singlemuscle fibres. J Physiol (Lond) 153: 386-403

Hunter DR, Komai H, Haworth RA, Jackson MD, Berkoff HA(1979) Comparison of Ca2+, Sr2"1", and Mn2+ fluxes in mitochon-dria of the perfused rat heart. Circ Res 47: 721-727

Hunter DR, Haworth RA, Berkoff HA (1981) Measurement ofrapidly exchangeable cellular calcium in the perfused beating ratheart. Proc Natl Acad Sci USA 78: 5665-5668

Itoh T, Kuriyama H, Suzuki H (1981) Excitation-contraction cou-pling in smooth muscle cells of the guinea-pig mesenteric artery,j Physiol (Lond) 321: 513-535

Jandt H, Porzig H, Reuter H, Stucki JW (1975) The effect ofsubstances releasing intracellular calcium ions on sodium-depen-

dent calcium efflux from guinea-pig auricles. J Physiol (Lond)246: 229-253

Johnson PN, Inesi G (1969) The effect of methyxanthines and localanesthetics on fragmented sarcoplasmic reticulum. J PharmacolExp Ther 196: 308-314

Josephson I, Sperelakis N (1976) Local anesthetic blockage of Ca2+-mediated action potentials in cardiac muscle. Eur J Pharmacol40: 201-208

Katz AM, Repke DI, Hasselbach W (1977) Dependence of iono-phore- and caffeine-induced calcium release from sarcoplasmicreticulum vesicles on external and internal calcium ion concen-trations. J Biol Chem 252: 1938-1949

Kohlhardt M, Bauer B, Krause H, Fleckenstein A (1972) Differen-tiation of the transmembrane Na and Ca channels in mammaliancardiac fibres by the use of specific inhibitors Pfluegers Arch335: 309-322

Liittgau HC, Oetliker H (1968) The action of caffeine on theactivation of the contractile mechanism in striated muscle fibres.J Physiol (Lond) 194: 51-74

Neeley JR, Liebermeister H, Battersby EJ, Morgan HE (1967) Effectof pressure development on oxygen consumption by isolated ratheart. Am J Physiol 212: 804-814

Ochi R, Hashimoto K (1978) The effect of procaine on the passiveelectrical properties of guinea-pig ventricle muscle. PfluegersArch 378: 1-7

Pretorius PJ, Pohl WG, Smithen CS, Inesi G (1969) Structural andfunctional characterization of dog heart microsomes. Circ Res 25:487-499

Reed PW, Lardy HA (1972) A23187: A divalent cation ionophore.J Biol Chem 247: 6970-6977

Rusy BF, Coulson RL (1973) Energy consumption in the isolatedrabbit heart. Anesthesiology 39: 428-434

Scarpa A, Baldassare J, Inesi G (1972) The effect of calciumionophores on fragmented sarcoplasmic reticulum. J Gen Physiol60: 735-749

Shine KI, Langer GA (1971) Caffeine effects upon contraction andcalcium exchange in rabbit myocardium. J Mol Cell Cardiol 3:255-270

Solaro RJ, Briggs FN (1974) In Recent Advances in Studies onCardiac Structure and Metabolism, Vol 4, edited by NS Dhalla.Baltimore, University Park Press, pp 359-374

Takakuwa Y, Kanazawa T (1981) Reaction mechanism of (Ca2+,Mg2+)-ATPase of sarcoplasmic reticulum. J Biol Chem 256: 2696-2700

Thorens S (1971) Electrical and mechanical effects of procaine onmammalian heart muscle. Pfluegers Arch 324: 56-66

van Breemen C, Siegel B (1980) The mechanism of a-adrenergicactivation of the dog coronary artery. Circ Res 46: 426-429

Weidmann S (1955) Effects of calcium ions and local anaestheticson electrical properties of Purkinje fibres. J Physiol (Lond) 129:568-582

Zimmerman ANE, Hulsmann WC (1966) Paradoxical influence ofcalicum ions on the permeability of the cell membrane of theisolated rat heart. Nature 211: 646-647

INDEX TERMS: Calcium metabolism • Caffeine • Procaine •Excitation-contraction coupling • Sarcoplasmic reticulum