Sarcolemmal Ca influx through L-type Ca channels in ventricular myocytes of a teleost fish. MATTI VORNANEN. Department of Biology, University of Joensuu, ...
Sarcolemmal in ventricular
Ca influx through L-type Ca channels myocytes of a teleost fish
MATTI VORNANEN Department of Biology,
University
of Joensuu,
Vornanen, Matti. Sarcolemmal Ca influx through L-type Ca channels in ventricular myocytes of a teleost fish. Am. J. Physiol. 272 (Regulatory Integrative Comp. Physiol. 41): R1432-R1440, 1997.-The whole cell patch clamp method was used to measure Ca current through L-type Ca channels in enzymatically isolated ventricular myocytes of crucian carp (Carassius carassius L.) heart. Fish were acclimated to 22°C for more than 4 wk, and properties of Ca current were measured at room temperature (21 ? 1°C). Depolarizing voltage steps from -50 mV evoked rapidly activating Ca currents, which exhibited a bell-shaped voltage dependence with peak amplitude at 0 mV The currents were suppressed by nifedipine (5 PM), verapamil(2.5 PM), and Cd2+ (175 PM). The current amplitude was increased by 67.5 2 17.2% (n = 5) in the presence of 1 PM isoproterenol. Steady-state inactivation and activation curves showed half-maximal inactivation at -31.3 t 0.95 mV, with a slope factor of 5.88 t 0.51, and half-maximal activation at -10.6 t 1.65 mV, with a slope factor of 7.84 t 0.54 (n = 9). The overlap of inactivation and activation curves suggests the presence of a small window current, which is maximally 4% of the peak current at -27 mV. The density of L-type Ca current was 6.95 t 0.79 pA/pF at 0 mV (n = 35). A total increment in cellular Ca contributed by L-type Ca current during a 500-ms voltage clamp pulse was calculated from the integral of Ca current and cell volume. The charge transfer through L-type Ca current was 0.325 5 0.023 pC/pF, and the mean cell volume was 1,377 t 44 pm3. The increment in total cellular Ca by Ca influx through L-type Ca channels was calculated to be 39.3 t 2.8 PM. These findings imply that Ca influx through L-type Ca channels can contribute significantly to the activation of contraction in the ventricular myocytes of fish heart.
FIN-80101
Joensuu,
Finland
present in the ventricular myocytes of different fish species (22) but its physiological significance is largely unresolved. Contraction of the fish ventricle is resistant to ryanodine, which suggests that Ca stores in the SR are not directly involved in the activation of contraction (14, 18, 25). Instead, the force of contraction is relatively sensitive to low concentrations of Cd (27), indicating a Cd-sensitive pathway, possibly Ca channels of the SL, as a Ca influx route. In the frog heart, iyhich also lacks an efficient Ca release mechanism in the SR, both Na/Ca exchanger and L-type Ca channels are able to provide Ca for contractile activation, although the latter seems to be the primary physiological pathway for Ca influx (11, 19, 20). Because neither of these sarcolemmal pathways has been characterized or quantified in teleost hearts, the aim of the present work was to study the presence of L-type Ca current in ventricular myocytes of teleost fish, to characterize its properties, and finally, to calculate the contribution of transsarcolemmal Ca influx through L-type Ca channels to the total intracellular Ca. The results show the ventricular myocytes in crucian carp have a robust L-type Ca current, which may contribute a significant proportion of the activator Ca. METHODS
THE CONTRACTIONOF CARDIACmyocyte is initiated and graded by an increase in the concentration of free intracellular Ca. In vertebrate hearts the contractile Ca has a dual origin; part of the activator comes from extracellular space through the sarcolemma (SL) either via Ca channels or an NalCa exchanger, part from the intracellular stores of the sarcoplasmic reticulum (SR) through ryanodine-sensitive Ca channels. The relative importance of extra- and intracellular sources of Ca varies greatly between different classes of vertebrates. In ectothermic vertebrates, like fish and amphibians, extracellular stores are considered to be the primary source of Ca ions (18, 19, 23, 25, 27), whereas in avian and mammalian hearts a small amount of transarcolemma1 Ca influx releases a major part of the activator Ca from the SR via a Ca-induced Ca release mechanism (5, 7, 15, 16, 29). This difference between ectotherms and endotherms in the source of activator Ca has a clear ultrastructural basis in that the cardiac SR is relatively sparsely developed in lower vertebrates compared with birds or mammals (6). Variable amounts of SR are
Animak. Crucian carp were caught in local ponds near the University of Joensuu in midsummer (June, July). In the laboratory these fish were kept in 500-liter tanks of tap water (ground water) at 22°C and with constant aeration and flow of water. The photoperiod was a 12:12-h light/dark cycle. The fish were fed daily with commercial fish food (Ewos). The size of the fish varied between 26 and 80 g (40.3 t 2.5 g, mean 2 SE, n = 22). IsoLution of ventriczdar myocytes. For perfusion of the heart, a blunt syringe needle was inserted through the bulbus arteriosus inside the ventricle. Solutions were then perfused (retrograde) from a height of 50 cm through the ventricle and atrium. The perfusion was started with a nominally Ca-free, low-Na solution (29) having the following composition (mM): 100 NaCl, 10 KCl, 1.2 KH2P04, 4 MgSOd, 50 taurine, 20 glucose, and 10 N-2-hydroxyethylpiperazine-W-2-ethanesulfonic acid (HEPES) at pH 6.9 (KOH). Seven minutes later the perfusion was continued with the same buffer but with added collagenase (1.3 mg/ml; Sigma Type IA), trypsin (1 mg/ml; Sigma Type III), and bovine serum albumin (1 mg/ml; fatty acid free, Sigma). After 30 min of enzyme perfusion, the ventricle was excised, minced with scissors, and put in a Ca-free, low-Na solution. The cells were then separated by trituration through the opening of a Pasteur pipette. They were stored in a refrigerator (5°C) and were used within 6 h of the isolation. Measurement ofCa current. The whole cell configuration of the patch-clamp technique was used for measurement of Ca current (Ica; 10). Dissociated myocytes were placed in a l-ml chamber on the stage of an inverted microscope (Leica DMIL). After they had adhered to the bottom of the chamber
R1432
the American
excitation-contraction
coupling;
single cells; myocyte size
0363-6119/97
$5.00
Copyright
o 1997
Physiological
Society
L-TYPE
CA
CURRENT
(0.5 ml) the cells were superfused with physiological solution at a rate of 2 ml/min. After exposure to physiological concentrations of Ca, the majority of the cells were relaxed, quiescent, and possessed clear cross striations (see Fig. 1). The composition of the external bathing solution was (mM) 130 NaCl, 5.4 CsCl, 1.5 MgS04, 0.4 NaH2P04, 2.0 CaC12, 10 glucose, and 10 HEPES (pH 7.4 with CsOH). In all measurements of Ca current, 1 PM tetrodotoxin (TTX) was included in the bathing solution to block fast Na currents. Patch pipettes (Vitrex microhematocrit tubing, Modulohm A/S) were fabricated with vertical two-stage pullers (Narashige MF83 or L/M-3P-A List-Electronic). The pipettes were filled with isotonic solution composed (mM) of 130 CsCl, 5 MgATP, 15 tetraethylammonium chloride (TEA-Cl), 1 MgClz, 5 oxaloacetate, 5 succinate, 5 ethylene glycol-bis( P-aminoethyl ether)N,N,N’,N’-tetraacetic acid (EGTA), and 10 HEPES (pH 7.2 with CsOH). Most current measurements were made with the use of this high-EGTA solution. In some experiments the EGTA concentration was lowered to 0.02 mM. When this low-EGTA solution was used, it is mentioned in the text. Whole cell currents were measured at room temperature (21 t l°C), which corresponds to the physiological body temperature of the fish. All recordings were made using an Axopatch 1D amplifier (Axon Instruments) equipped with a CV-4 l/l00 headstage. When pipettes were immersed in the physiological solution, the pipette resistances varied between 1 and 4 Ma (average 2.4 ? 0.19). Junction potentials were adjusted to zero before formation of the seal; the small potential difference between the pipette and the bath solution (-2 mV> was not corrected in the results. The immersion depth of the pipette was kept as shallow as possible, and the pipette capacitance (4-5 pF) was compensated for after formation of a gigaohm seal. The patch was ruptured by delivering a short voltage pulse (zap) to the cell, and capacitative transients were eliminated by iteratively adjusting series resistance and cell capacitance compensation circuits. The cell capacitance was read directly from the dial of the Axopatch 1D amplifier. No leakage correction was implemented. Ca currents were sampled at 2 kHz and Na currents at 10 kHz. Current tracings were filtered at 1 or 5 kHz with a four-pole Bessel filter, digitized (TL-1 DMA, Axon Instruments), and stored on the hard disk of the computer for off-line analyses by pClamp 6.01 software (Axon Instruments). The voltage dependence of steady-state inactivation and activation was determined by double-pulse protocol (Fig. 6). Cells were held at -50 mV, and l-s conditioning pulses were applied between -60 and +50 mV, followed by a 500-ms test pulse to 0 mV The voltage dependence of peak conductance of Ca channels was calculated from the equation
where gca is the Ca conductance of the membrane, Ic, is the peak Ca current at a given potential (V), and Vrev is the apparent reversal potential obtained by extrapolation of the ascending portion of the current-voltage (I-V) relation to the zerocurrent axis. Activation voltage dependence [da(V)] was determined as normalized Ca conductance dm(V) = gc,/gmax, where g,, is the maximum value of Ica conductance. Voltage dependence of steady-state inactivation fm(V) was calculated by dividing the amplitude of the test current by the maximal current elicited. Steady-state kinetic parameters were obtained by fitting the data to Boltzmann equations dm(V) = l/(1 fm(V) = 1 - [l/(1
+ exp [(VO.5 - V)lk]] + exp [(VOe5 - V>lk]>]
IN FISH
R1433
HEART
where Vo.5 is the half-activation or half-inactivation potential, and h is a slope factor. Charge transfer through L-type Ca channels was determined by integration of the inactivating part of Ca current for a 500-ms voltage pulse from -50 to 0 mV. The total increment in intracellular Ca was calculated from the time integral of the current and the myocyte volume according to the equation ACatot = SIC&> dt+FV)-l where z is the equivalent charge of 2 carried by the Ca ions, F is the Faraday constant with a value of 96,500 A* s-l mol-l, I&)/dt is the time integral of the Ca current and V is the volume of the myocyte. The Ca-accessible space was considered to be 78.8% of the whole myocyte volume, which is the electron microscopically determined nonmitochondrial volume of the isolated cells. Determination of cell size. For the analysis of cell volume and SL surface area, a magnified image of the cell was projected through a video camera on the screen of a television monitor. The contours of the cell were drawn on transparencies for later determination of maximum length and average width of the myocyte. Cell width was determined at three levels, in the middle and halfway from the midpoint to the ends. For quantification of external surface area and myocyte volume, cells were considered to be right cylinders with an elliptical cross section. The cell volume (V) was calculated from the equation l
V = lFa*b4 where a and b are the shorter and longer radii of the ellipse, respectively, and Z is the length of the cell. The ratio of minor to major axes of the ellipse was approximated from the cross sections of isolated myocytes as 1:2. Accordingly, the radii of the ellipse could be derived from the measured width (w) of the cell (b = w/2 and a = w/4>. The surface area of the cell (A,) is then the sum of the surface of the cylinder (Al) and its ends (P) of the ellipse was (A,) (A, = Al + 2A2). The perimeter approximated from the equation (24) P=2rI3/G5Tj The surface area of the cylinder
is then
A, = P.1 and the surface area of the ends A, = lFa*b ELectron microscopy. A small aliquot of the myocyte suspension was pipetted into Eppendorf tubes. The cells were fixed by adding 4 vol of 2% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.5) to 1 vol of myocyte suspension. Myocytes were allowed to form a sediment on the bottom of the vial, the solution was drawn out with a Pasteur pipette and replaced with an equal volume of fresh glutaraldehyde fixative. The cells were postfixed in 2% 0~0~ and 0.8% K4Fe(CN)G buffered with 0.1 M Na cacodylate (pH 7.5). Myocyte pellets were embedded in Epon, and thin sections were cut with an ultramicrotome (Ultratome Nova, LKBProdukter) and stained with saturated uranyl acetate and 0.4% lead citrate. Specimens were examined and photographed with a Zeiss 90 electron microscope at an operating voltage of 80 kV. Volume fractions of cellular organelles were obtained from paper prints of longitudinally sectioned myocytes by point counting.
R1434
L-TYPE
Table 1. Morphological Length,
pm
109.8 22.47 Results
are means
Width,
characteristics pm
5.78 t 0.11 2 SE of 104 myocytes
CA
of crucian Area,
pmz
1,584 2 39 from
CURRENT
IN FISH
carp ventricular Volume,
pm3
1,456 2 57
HEART
myocytes Length/Width
19.84 f 0.61
Area/volume,
pm-’
1.15 i 0.02
Fish
Mass,
g
32.5t4.1
3 fish.
Statistics. All data are presented as means + SE, and statistical comparisons were made using a two-tailed Student’s t-test. Graphs were drawn with the use of SigmaPlot software (Jandel Scientific). RESULTS
Myocyte size and morphology. Large numbers of thin, cylindrical, cross-striated and Ca-tolerant cells were obtained when a retrograde perfusion of fish hearts with collagenase and trypsin was used. Forward perfusion through the atrium was equally successful. The total duration of cell isolation with the perfusion technique was -1 h. On the other hand, isolation of cells by incubation of tissue chunks in dissociating solutions took 4-5 h with a very low yield of morphologically
intact cells. The ventricular myocytes of the crucian carp heart were on average 109.8 +- 2.27 pm in length and 5.78 + 0.11 jum in width, giving a length/width ratio of 19.84 + 0.61 (Table 1). In cross sections the myocytes appear somewhat flat, and thus an elliptical shape approximates the cross-sectional area better than a round shape does (Fig. lC>. The length ratio of the minor and major axes of the ellipse was approximated from electron microscope pictures as 1:2. The calculated area of the cell surface was 1,584 t 39 pm2 and cell volume was 1,456 2 57 ,um3, giving a surface-tovolume ratio of 1.15 t 0.02 pm-l (mean + SE of 104 myocytes from 3 fish; 32.5 % 4.1 g). Electron microscopic examination showed that the myocyte cytoplasm is occupied by one nucleus, a few peripherally located
Fig. 1. Ultrastructural characteristics of ventricular myocytes in crucian carp. A: photomicrograph of an enzymatically isolated myocyte of crucian carp heart. Scale bar, 20 pm. B: electron micrograph of a short section of longitudinally cut myocyte showing 1 nucleus, peripherally located myofibrils, mitochondria, and glycogen granules. Original magnification X 7,000. C: electron micrograph of a transversally sectioned ventricular myocyte. Note peripheral myofibrils, centrally located mitochondria, and abundant glycogen. Nonjunctional sarcoplasmic reticulum (arrowheads) is present beneath the sarcolemma. Original microscopic magnification x 12,000. Scale bar, 1 pm in both B and C.
L-TYPE
CA CURRENT
myofibrils, a number of centrally located mitochondria, and massive stores of glycogen (Fig. 1). T tubules are totally absent but SR profiles are evident around the myofibrils and especially under the surface SL. Rough estimates of mitochondrial and myofibrillar volume densities were obtained by point counting the cellular constituents in longitudinal sections of seven myocytes. The average mitochondrial and myofibrillar volume densities were 21.2 2 1.4 and 38.3 t 3.4%, respectively. Large areas of the cytoplasm were occupied by glycogen granules (30.2 t_ 3.8%), as reported previously (26). Characterization of Ca current in fish ventricular myocytes. The ventricular myocytes of the crucian carp heart are suitable for patch-clamp experiments because they are relatively small with a mean cell capacitance of only 25.2 t 0.9 pF (n = 62). Recordings of Ica were made in K-free external solution and using a pipette solution containing Cs and TEA to suppress interfering K currents. No outward current tail was observed at the end of the voltage pulse, demonstrating that dialysis with Cs and TEA had blocked any outward currents-activated during the pulse (e.g., Figs. 4A and 6A). Ca-activated currents were prevented by chelating intracellular Ca with 5 mM EGTA. Because experiments were conducted in Na-containing solution, it was necessary to exclude any inward currents that might flow through Na channels. The ability of TTX to block the Na currents was studied in six myocytes. Depolarizing clamp steps from the holding potential of -80 mV elicited large Na currents, which were completely blocked by 1 PM TTX in all six cells studied (Fig. 2). This shows that, like frog cardiac myocytes but unlike mammalian myocytes, the Na channels of fish ventricu-
A
nA -2
-4
L
5 ms
- zm -80 -120
1
-80
I
-60
I
-40
I
-20
I
I
I
I
I
0
20
40
60
80
mV Fig. 2. Tetrodotoxin (TTX) suppresses voltage-activated Na+ current (INa) in ventricular myocytes of crucian carp heart. A: superimposed traces in absence and presence of 1 PM TTX. Currents were elicited by lo-mV increment voltage steps (50 ms, 0.1 Hz) from a holding potential of -80 down to -70 mV and up to +60 mV. B: currentvoltage relationships of peak I Na in absence and presence of 1 PM TTX.
IN FISH
R1435
HEART 11
-6 -80
-60
-40
-20
0
20
40
60
80
mV Fig. 3. Current-voltage relations of peak inward Ca current (1~~) elicited from the holding potential of -80 mV. In 10 of 12 ventricular myocytes (e) the first clearly visible inward current was evident at the pulse potential of - 30 mV and peaked at 0,mV. In the remaining 2 cells ( n ) an additional low-threshold Ica current with peak amplitude at -40 mV was present. Experiments were conducted in the presence of 1 jzM TT’X.
lar cells are highly sensitive to TTX (12). In all subsequent experiments, 1 JJM TTX was included in the external solution. Next, the Ca currents were elicited, in the presence of TTX, from the holding potential of -80 mV In 10 of 12 cells, Ca current appeared first at the pulse potential of -40 or -30 mV and reached peak amplitude at around 0 mV, suggesting that only L-type Ca channels were present in these cells. On the other hand, in 2 of 12 cells, a smaller current component was present at more negative potentials, which reached an early maximum at -40 mV (Fig. 3). Because small T-type Ca currents seem to be present in some ventricular myocytes of crucian carp heart, all subsequent experiments were conducted from the holding potential of -50 mV (in the presence of 1 PM TTX) to inactivate possible T-type Ca channels. L-type Ca currents were elicited from the holding potential of - 50 mV to test potentials between -40 and +60 mV for 500 ms at a rate of 0.1 Hz. Under these experimental conditions, rapidly activating inward currents appeared during the depolarizing voltage steps; at 0 mV the peak current was reached within 6 ms from the start of the clamp step. The threshold for lca was about -30 mV, and the amplitude increased with successive depolarizations up to 0 mV, at which Ic, was maximal. At positive to 0 mV Ica decreased, and only a small inward current was present at +50 mV The current was almost completely suppressed by 5 PM nifedipine (96.7 5 2.3%, n = 8) (Fig. 4), 2.5 PM verapamil (98.6 t 1.7%, n = 4), and 175 PM CdC12 (98.5 t 0.9%, n = 4). A maximally effective concentration of isoprenaline (1 PM) (25) increased the amplitude of Ica with concomitant shift of the I-V relationship by 10 mV in a negative direction (Fig. 5). The peak Ca2+ current at 0 mV and its time integral increased from 169 to 281 pA(67.5 t 17.2%; P < 0.05) and 7.34 to 13.07 pC (72.1 t 20.9%, n = 5; P < 0.05), respectively. Steady-state activation and inactivation of Ica. Voltage dependence of activation and inactivation of Ica were determined by the double-pulse protocol. The data were fitted with the Boltzmann function, and halfmaximal voltage (Vo5) . and slope factor (k) were ob-
R1436
L-TYPE
A
I
CA
CURRENT
-40
-20
I200 ms
I
-40
I
-20
t
I
0
20
HEART
fast component comprised 62 2 22% of the total current at this potential. Both time constants were voltage dependent. The slow component showed a clear minimum at 0 mV and increased steeply both at more negative and more positive membrane potentials. The time constant of the fast component was shortest at -10 mV and increased at more positive voltages. At a more physiological level of Ca buffering (20 PM EGTA), the time constants for fast and slow components were 24 1 + 4.15 and 96.3 2 11.5 ms, respectively (n = 7). The somewhat slower inactivation rate of the fast component in the presence of low EGTA may be explained by weaker Ca-dependent inactivation of Ic, as a consequence of 20% smaller current density under these experimental conditions. Contribution of Ca current to total cel&r Cu. The amplitude of L-type Ca current and its time integral were measured in 35 myocytes for a voltage step from - 50 to 0 mV (Table 2). Only the inactivating component of the Ca current was measured. Because the noninactivating window current is practically zero at 0 mV, the time integral should be representative of the total charge transfer through L-type Ca channels at this potential. The increment in total cellular Ca was determined from the time integral of current and myocyte volume. Cell volume was calculated for each myocyte from the capacitive surface area and morphometrically determined surface-to-volume ratio of 1.15. Membrane capacity was converted to surface area by
, +40 mV
HP=-50 mV -J
100 pA
IN FISH
I
I
40
60
A
mV
100 pA
Fig. 4. Voltage-dependence of L-type Ca current and its block by 5 PM nifedipine. A: superimposed current traces in absence and presence of 5 PM nifedipine at 5 different membrane potentials. B: mean current-voltage relationship of nifedipine-sensitive Ica in ventricular myocytes of crucian carp heart (n = 19). HP, holding potential.
tained for each cell (n = 9). For voltage-dependent activation of L-type Ca current, the average VOa5was -10 .6 -+ 165 . mV, and the average k was 7.84 5 0.54 mV. Corresponding values for steady-state inactivation were -31.3 t 0.95 and 5.88 t 0.51 mV, respectively. The relief of inactivation at positive prepulse voltages suggests that part of the inactivation is Ca dependent (3). Inactivat’ ion and activation curves overlapped, suggesting the presence of a small window current (Fig. 6). The window current was maximal at -27 mV, where it constituted slightly ~4% of the peak inward Ca current. At 0 mV the window current was negligible. Kinetics of inactivation of &. The time course of decay of Ica during depolarization was studied by fitting the current change between the inward peak and the current level 500 ms after depolarization by doubleexponential function. Figure 7 gives examples of the time course of decay of L-type Ca current at -2O,O, and +20 mV as well as average results from seven cells for a voltage range of -30 to +30 mV. At all potentials the double-exponential function fits the data very well. The time constants for fast and slow components at 0 mV were 17.3 ? 2.2 and 93.4 + _ 5.5 ms, respectively The
v\
Iso 1 pM
5 PC
control Is0 200
ms
B O-2 s hn 3
-4-6-8 -
-60
-40
-20
0
20
40
60
80
mV Fig. 5. Amplification of L-type Ca current by crucian carp ventricular myocytes. A: original an increase in amplitude of Ca current and its (n = 5) current-voltage relation of lca before (Iso; 1 PM) application.
isoproterenol(1 PM) in current traces showing time integral. B: mean and after isoproterenol
L-TYPE
CA
CURRENT
HP= -50 mV
‘a
I/\
I v
‘C
‘b
k
R1437
HEART
most notable exception (13), SR seems to contribute little to the activation of contraction under normal physiological conditions (27). The total absence of postextrasystolic potentiation and ryanodine insensitiveness of contraction are indications of low functional activity of SR in ventricular muscle of crucian carp. Suppression of force development by low concentrations of Cd (100,300 PM) suggests that sarcolemmal Ca channels play a central role in control of contraction in the ventricular myocytes of fish heart (27). The influence of Ca current could be either a direct contribution
A
100PAI
IN FISH
a
500 ms
I3
A
.,: -
1.0
0.6
zQ 25 -200
0.4
-300
0.8
3
z,=19.1 2,=211.1
-20 mv I
I
I
B 0.2
-90
-60
-30
0
mV
30
60
-90
-60
-30
0
30
60
mV
o- *. . zn -100- : .. $ -200 - :j
Fig. 6. Steady-state activation and inactivation of lca in ventricular myocytes of crucian carp heart. A: conditioning pulses of l-s duration were delivered from the holding potential of -50 mV to voltages between - 70 and +60 mV, and amplitude of the lca of the conditioning pulse and the following test pulses (a-c) to 0 mV was measured. B: inactivation and activation curves as means of 9 myocytes (Left) and window current (right; % of peak 1~~) calculated as product of inactivation and activation parameters for a voltage range of -70 to +60 mV
using a value of 62.8 pm2/pF (specific membrane capacitance of 1.59 pF/cm2), which was obtained by dividing the average area of cell surface (1,584 pm2) by the average capacitance of the myocytes (25.2 pF, n = 62). Peak Ca current density was 6.95 t_ 0.79 pA/pF, and integrated current for a 500-ms pulse was 8.69 -+ 0.71 PC, which makes 0.325 t 0.023 pC/pF (Table 2). Ca influx through L-type Ca channels increased total cellular Ca by 39.3 t 2.8 JXM (n = 35) and in the presence of 1 PM isoproterenol as much as 55.6 t 8.8 PM (n = 5). With the low-EGTA pipette solution the density and charge transfer of the nifedipine (5 PM)sensitive current were 20% (5.51 t 1.1 pA/pF, n = 7) and 10% lower (0.295 t 0.021 pC/pF, n = 7), respectively, than with the high-EGTA pipette solution. The contribution of L-type Ca current to total myoplasmic Ca with low-EGTA buffer was 35.6 t 2.54 PM, which is 10% less than in the presence of 5 mM EGTA buffer. DISCUSSION
In the heart of crucian carp and the cardiac muscle of many other teleost species, the tuna heart being the
3=17.5 Ts=90.3 OmV I
1
-300
I
I
x,=22.6 zs=l 05.8 +20 mV
I
I
I
0
200
400
600
ms
n 200 150 z 100
01 -40
I -20
I
I
I
0
20
40
mV Fig. 7. Inactivation rate of L-type lca in ventricular myocytes of crucian carp heart. A, B, and C show original traces of current (dots) at -20, 0, and +20 mV, respectively, and their computer fits to the biexponential equation (continuous lines). D illustrates mean results (n = 7) for time constants of slow (+T~; top) and fast (Tf; bottom) components of Ica.
R1438
L-TYPE
2. Characterization
Table
Cell,
pF
26.4 t 0.88 Results
are means
Icm PA
of Ica in ventricular Ica Density,
182 +: 20 t SE from
pA/pF
6.95 _+ 0.79 35 myocytes.
&,
Ca current;
CA
CURRENT
myocytes
IN FISH
of crucian
HEART
carp heart
SIcmPC
PC/PF
8.69 t 0.71 Vcell, cell volume;
of extracellular Ca to the myofilaments or an indirect one through the maintenance of an action potential plateau, thus providing transsarcolemmal Ca influx by other voltage-dependent pathways, such as Na/Ca exchangers. To clarify the direct contribution of Ica to contractile activation, we have estimated the total increment in intracellular Ca via L-type Ca channels. To this end, charge transfer through Ca channels as well as volume of the myocytes were determined. Our analysis shows that the ventricular myocytes of a teleost heart possess a robust L-type Ica, whereas T-type Ica, if present, seems to be relatively rare and small in amplitude. Maylie and Morad (17) have shown that in elasmobranch (dogfish Squalus acanthias) ventricular cells the density of T-type lca is equal to that of L-type current. Because only one elasmobranch and one teleost species have been studied thus far, it remains to be shown how representative these findings are for other members of these vertebrate classes. The density of L-type Ica of the crucian carp. heart is slightly lower than the Ca current of shark ventricular myocytes (measured at 5 mM extracellular Ca) (17) but somewhat greater than the L-type current of frog atria1 or ventricular myocytes (2,3). The kinetics and pharmacological properties of the crucian carp Ica are similar to those in other vertebrates. The current is blocked by dihydropyridines (nifedipine) and phenylalkylamines (verapamil) as well as by inorganic divalent cations (Cd). The amplitude of the current is increased by P-receptor agonist (isoproterenol), and, typical for the P-receptor response, the peak current is shifted slightly to more negative voltages (17). Steady-state inactivation and activation curves are very similar to those recorded in the myocytes of frog heart (2) and suggest the presence of persistent window current at slightly negative voltages. The window lca probably contributes to the maintenance of the long plateau phase of cardiac action potential in crucian carp (27). Compared with mammalian ventricular myocytes, fish ventricular cells are almost the same length but less than one-fourth the width. The narrow shape of heart cells in fish has two important consequences: it increases the surface-to-volume ratio of the myocytes and decreases diffusion distance from SL to the core of the cell. The relative sarcolemmal surface area of fish myocytes is almost four times larger than the external surface area of adult mammalian myocytes and still more than twofold when the T tubular surface of mammalian cells is included (6). On the other hand, surface-to-volume ratio of frog ventricular myocytes (1.19) (6) is close to that of crucian carp myocytes. Therefore, it seems that efficient sarcolemmal Ca influx in ectothermic vertebrates is attained by a high surfaceto-volume ratio of myocytes without prominent in-
0.325 ACatOt,
change
Vcell,
t 0.023 in total
Pm3
1,377 t 44
ACatot9
PM
39.3 + 2.8
Ca.
creases in Ca current density Because Ca diffusion inside cells is greatly attenuated by an abundance of Ca buffering compounds (l), the distance and location of the myofibrils with regard to the Ca source is crucial for activation of contraction. In crucian carp myocytes, myofibrils form a distinct cortical layer beneath the SL, whereas the nucleus, mitochondria, and glycogen are restricted to the center of the cell. The close proximity of the myofibrils and SL in conjunction with relatively slow force development would make an extracellular source of Ca an effective means of initiating contraction in the cardiac myocytes of fish. The access of activator Ca to myofilaments may be limited by the SR, which in several myocytes forms a clearly discernable network between SL and the myofibrils, and thus may be capable of sequestering part of the sarcolemmal Ca influx. Cyclopiazonic acid (20 PM), an inhibitor of SR Ca-ATPase, does not change the force or time course of contraction in the crucian carp ventricle (unpublished observations), which suggests that under normal physiological conditions SR is not a strong Ca buffer. In crucian carp myocytes the total Ca flux into the CytOSOl via Ica was calculated to be 39.3 PM, which is about four times more than in the ventricular myocytes of adult mammals (4, 21). This value is, however, only slightly lower than that estimated for ventricular myocytes of newborn rats, which like fish myocytes are supposed to be strongly dependent on transsarcolemma1 influx of activator Ca (28). Patch-clamp measurement of Ica involves intracellular perfusion with artificial solutions and stimulation with simplified voltage waveforms, which raises the question of the physiological relevance of the results. The duration of the voltage step was selected to correspond to the duration of cardiac action potential of the warm-acclimated crucian carp, and therefore the measured Ca influx should approximate the in vivo conditions. On the other hand, the high intracellular Ca buffering, which was necessary to inhibit interfering Ca-activated currents, could cause overestimation of lca due to an unphysiologically steep electrochemical gradient for Ca influx or absence of Ca-dependent inactivation of .Zca. In fact, at more physiological Ca buffering (20 PM EGTA), the peak density of I ca was reduced by 20%. However, at the same time the inactivation rate of Ica also declined so that the integrated Ca current was only 10% less than in the presence of high EGTA. Ca-dependent inactivation of 1ca is directly related to current density (3), which could explain this finding. That a significant amount of Ca-dependent inactivation is present with 5 mM EGTAin the pipette solution is shown by the relief of steady-state inactivation at positive prepulse voltages (Fig. 6). On the other hand, the amplitude of lca could be underestimated by the use of a relatively
L-TYPE
CA
CURRENT
depolarized holding potential; in the ventricular cells of the frog heart a drop of holding potential from -80 to -40 mV causes a -50% decrease in Ic, amplitude (2). In the ventricular myocytes of crucian carp the current density was very similar, whether elicited from -50 mV or from the more physiological holding potential of -80 mV (Figs. 3 and 4). This agrees well with the steady-state inactivation of Ica, which shows that at -50 mV >95% of the channels are available for activation (Fig. 6). These findings suggest that the experimental conditions used do not cause any serious under- or overestimation of &,. The contribution of any Ca source to contractile activation is strongly dependent on the Ca-buffering capacity of the cell. If Fabiato’s (9) calculations for Ca-buffering capacity of mammalian ventricular cells are applied to fish myocytes (supposing a dissociation constant of 0.5 x lo6 M for troponin C), 39.3 PM total Ca would raise the cytosolic free Ca concentration from 0.079 to 0.79 PM (change of 0.71 PM). According to the myofibrillar tension-intracellular Ca concentration relationship of Yue et al. (30) (with a Hill coefficient of 4), this amount of cytosolic free Ca, if in steady state with myofibrils, would produce 86% of the maximal tension that myofibrils are able to generate. With the use of the lower value of 35.6 PM for the total cellular Ca, 68% of the maximal tetanic force would still be generated. AIthough it is unlikely that force and intracellular Ca concentration reach steady state during a normal twitch, these calculations imply that Ic, can make a significant contribution to contractile Ca in fish myocytes. In the rabbit, which among the laboratory mammals studied is most strongly dependent on extracellular Ca for contractile activation, the total cytosolic Ca flux during a steady-state twitch is calculated to be 47 PM; of this, -23% comes through L-type Ca channels (8). In ventricular myocytes of crucian carp, the Ca influx through L-type Ca channels is -3.5 times bigger than in ventricular myocytes of rabbits, and therefore it would be anticipated that Ica contributes a much higher percentage of activator Ca in fish than in mammalian ventricular cells. Although they clearly indicate the importance of Ica in transsarcolemmal Ca influx in fish cardiac myocytes, the quantitative estimates should be interpreted with caution because it is probable that intracellular Ca buffering differs in fish and mammalian cardiac cells. Furthermore, there may be species differences not only between mammalian species but also between different teleost species with regard to the extent to which Ic, contributes to contractile Ca. In conclusion, the present results suggest a prominent sarcolemmal Ca influx through L-type Ca channels in ventricular myocytes of crucian carp heart. Sarcolemmal Ca entry as a source of activator Ca is favored by 1) the small diameter of fish myocytes (which increases the sarcolemmal surface-to-volume ratio of the cells and decreases diffusion distance) and 2) the peripheral location of the myofibrils. This study was supported by Academy of Finland. Address for reprint requests: Dept. of Biology, Univ. PO Box 111, FIN-80101 Joensuu, Finland. Received
11 June
1996: accented
in final
form
10 December
of Joensuu, 1996.
IN FISH
HEART
R1439
REFERENCES 1. Albritton, N. L., T. Meyer, and L. Stryer. Range of messenger action of calcium ion and inositol 1,4,5trisphosphate. Science 258: 1812-1815,1992. 2. Alvarez, J. L., and G. Vassort. Properties of the low threshold Ca current in single frog atria1 cardiomyocytes. A comparison with the high threshold Ca current. J. Gen. PhysioZ. 100: 519-545,1992. 3. Argibay, J. A., R. Fischmeister, and H. C. Hartzell. Inactivation, reactivation and pacing dependence of calcium current in frog cardiocytes: correlation with current density. J. PhysioZ. (Land.) 401: 201-226,1988. 4. Berlin, J. R., J. W. M. Bassani, and D. M. Bers. Intrinsic cy-tosolic calcium buffering properties of single rat cardiac myocytes. Biophys. J. 67: 1775~1787,1994. 5. Bers, D. M. Ca influx and sarcoplasmic reticulum Ca release in cardiac muscle activation during postrest recovery Am. J. PhysioZ. 248 (Heart Circ. Physiol. 17): H366-H381,1985. 6. Bossen, E. H., and J. R. Sommer. Comparative stereology of the lizard and frog myocardium. Tissue CeZZ 16: 173-178,1984. M. B., H. Cheng, and W. J. Lederer. The control of 7. Cannell, calcium release in heart muscle. Science 268: 1045-1049,1995. L. M., J. W. M. Bassani, and D. M. Bers. Steady8. Delbridge, state twitch Ca2+ fluxes and cytosolic Ca2+ buffering in rabbit ventricular myocytes. Am. J. PhysioZ. 270 (CeZZ PhysioZ. 39): C192-C199,1996. A. Calcium-induced release of calcium from the cardiac 9. Fabiato, sarcoplasmic reticulum. Am. J. Physiol. 245 (CeZZ Physiol. 14): Cl-C14,1983. O., A. Marty, E. Neher, B. Sakmann, and F. J. 10. Hamill, S&worth. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. PfZiigers Arch. 391: 85-100,198l. M., and G. Vassort. Sodium-calcium exchange in 11. Horackova, regulation of cardiac contractility. Evidence for an electrogenic, voltage-dependent mechanism. J. Gen. PhysioZ. 73: 403-424, 1979. J. R., and W. Giles. Ionic currents in single isolated 12. Hume, bullfrog atria1 cells. J. Gen. Physiol. 81: 153-194,1983. J. E., A. P. Farrell, G. F. Tibbits, and R. W. Brill. 13. Keen, Cardiac physiology in tunas. II. Effect of ryanodine, calcium, and adrenaline on force-frequency relationships in atria1 strips from skipjack tuna, Katsuwonuspelamis. Can. J. ZooZ. 70: 1211-1217, 1992. J. E., D.-M. Vianzon, A. P. Farrell, and G. F. Tibbits. 14. Keen, Effect of temperature and temperature acclimation on the ryanodine sensitivity of the trout myocardium. J. Comp. PhysioZ. B Biochem. Syst. Environ. PhysioZ. 164: 438-443,1994. O., A. J. Levi, and J. H. B. Bridge. Relation 15. Kohmoto, between reverse sodium-calcium exchange and sarcoplasmic reticulum calcium release in guinea pig ventricular cells. Circ. Res. 74: 550-554,1994. J. R., P. S. Shacklock, C. W. Balke, and W. G. 16. Lopez-Lopez, Wier. Local calcium transients triggered by single L-type calcium channel currents in cardiac cells. Science 268: 1042-1045, 1995. J. G., and M. Morad. Evaluation of T-type and L-type 17. Maylie, Ca2+ currents in shark ventricular myocytes. Am. J. PhysioZ. 269 (Heart Circ. PhysioZ. 38): H1695-H1703,1995. T., and H. Gesser. Sarcoplasmic reticulum 18. Moller-Nielsen, and excitation-contraction coupling at 20 and 10°C in rainbow trout myocardium. J. Comp. PhysioZ. B Biochem. Syst. Environ. Physiol. 162: 526-534,1992. M., Y. E. Goldman, and D. R. Trentham. Rapid 19. Morad, photochemical inactivation of Ca2+-antagonists shows that Ca2+ entry directly activates contraction in frog heart. Nature 304: 635-638,198l. M., G. C. R. Ellis-Davies, J. H. Kaplan, and M. 20. Nabauer, Morad. Modulation of Ca2+ channel selectivity and cardiac contraction by photorelease of Ca2+. Am. J. Physiol. 256 (Heart Circ. PhysioZ. 25): H916-H920,1989.
R1440
L-TYPE
CA
CURRENT
21. Negretti,
IN FISH
N., A. Varro, and D. A. Eisner. Estimate of net calcium fluxes and sarcoplasmic reticulum calcium content during systole in rat ventricular myocytes. J. Physiol. (Lond.) 486:
26.
581-591,1995.
27.
Santer, R. M. Morphology and innervation of the fish heart. Adu. Anat. EmbryoZ. CeZZ BioZ. 89: l-102,1985. 23. Shepherd, N., and F. Kavaler. Direct control of contraction force of single frog atria1 cells by extracellular ions. Am. J. Physiol. 25l(CeZZ PhysioZ. 20): C653-C661,1986. 24. Spiegel, M. R. MathematicaZ Handbook. New York: McGrawHill, 1968, p. 1-271. 25. Tibbits, G. F., L. Hove-Madsen, and D. M. Bers. Calcium transport and the regulation of cardiac contractility in teleosts-a comparison with higher vertebrates. Can. J. ZooZ. 69: 2014-2019,199l. 22.
28. 29.
30.
HEART
Vornanen, M. Seasonal adaptation of crucian carp (Carassius carassius L.) heart: glycogen stores and lactate dehydrogenase activity. Can. J. ZooZ. 72: 433-442,1994. Vornanen, M. Effects of extracellular calcium on the contractility of warm-and cold-acclimated crucian carp heart. J. Comp. PhysioZ. B Biochem. Syst. Enuiron. Physiol. 166: l-11,1996. Vornanen, M. Contribution of Ca current to total cellular Ca in postnatally developing rat heart. Cardiouasc. Res. 32: 400410,1996. Vornanen, M., N. Shepherd, and G. Isenberg. Tensionvoltage relations of single myocytes reflect Ca release triggered by Na/Ca exchange at 35°C but not 23OC. Am. J. PhysioZ. 267 (CeZZ Physiol. 36): C623-C632,1994. Yue, D. T., E. Marban, and W. G. Wier. Relationship between force and intracellular [Ca2+] in tetanized mammalian heart muscle. J. Gen. PhysioZ. 87: 223-242,1986.
