AMERICAN JOURNAL OF PHYSIOLOGY Vol. 231, No. 4, October 1976. Printed
in U.S.A.
Ischemia in isolated interventricular septa: mechanical events KENNETH I. SHINE, ANNIE M. DOUGLAS, AND NICHOLAS RICCHIUTI Departments of Physiology and Medicine and Los Angeles County Cardiovascular Laboratory, University of California at Los Angeles, Center for the Health Sciences, Los Angeles, California 90024
I., ANNIE M. DOUGLAS, AND NICHOLAS Ischemia in isolated interventricular septa: mechanical events. Am. J. Physiol. 231(4): 12251232. 1976. Isolated blood-perfused rabbit interventricular septa were adapted for studies of global ischemia by enclosure in a constant-humidity nitrogen atmosphere. During ischemia, developed tension (DT) and maximal rate of relaxation (-dP/dt> declined monoexponentially, h = 0.39 min-’ at 37°C and 72 beats/min with a Q10 of 1.4 for DT and a Q10 of 1.9 for -dP/dt. After a 60- to 90-s delay the maximal rate of tension development (+dP/dt) declined at the same rate as DT. Time-to-peak tension (TPT) shortened immediately with ischemia but action potential duration shortened after 60-90 s. Calcium at a concentration of 5 mM slowed the rate of decline of +dPldt to h = 0.26 min. Upon reperfusion after 10 min of ischemia the rates of recovery of DT, +dP/dt, and --dP/dt were similar, h = 0.21-0.23 min-‘, and were not temperature dependent. The magnitude of recovery was lo-17% less at 37°C than 28°C. Potassium at a concentration of 10 mM did not alter the rate of decline of mechanical function, but significantly (P < 0.01) increased the magnitude of mechanical recovery. The results suggest depletion and/or repletion of single compartments as the rate-limiting steps in ischemia and reperfusion. SHINE, RICCHIUTI.
reperfusion;
KENNETH
calcium;
potassium;
rabbit
septa
PRECISE MECHANISMS for loss of force development in ischemic myocardium are unknown (4). Although several satisfactory preparations are available for studies of hypoxia in isolated myocardial muscle (17), the controlled examination of ischemia has been less easily achieved. Ligation of a coronary artery in the in situ heart presents problems of blood flow via collateral coronary and bronchopulmonary vessels, thereby producing variable degrees of ischemia and hypoperfusion in the affected myocardium. Other models have included graded decreases in coronary flow without total ischemia (11). Langendorff- ‘type preparations generally require coronary perfusion rates during the control period which are severalfold greater than physiological levels, with considerable interstitial edema and potentially large heterogeneity of flow during low flow states and upon reperfusion (11). In all these preparations the venous effluent collected may originate in areas of markedly different degrees of perfusion and oxygenation. We have adapted the isolated blood-perfused interTHE
Research
ventricular septal preparation to studies of ischemia so that abrupt total ischemia can be produced with reproducible consequences under a wide variety of conditions. Ischemia and reperfusion at physiological perfusion rates can be accomplished with rapid analysis of mechanical, electrophysiological, metabolic, and ionic events. The same preparation can be used for studies of hypoxemia. Mechanical and electrophysiological results of ischemia in isolated blood-perfused septa are reported here. These data provide the basis for additional observations of metabolic changes and ionic movement in this preparation. METHODS
The experimental preparation was the isolated bloodperfused interventricular septum of 1.5 to 2.0-kg white male New Zealand rabbits according to the technique described by Shine et al. (12, 13). The animals were anticoagulated with 10 mg sodium heparin intravenously, anesthetized with an intravenous injection of pentobarbital, 40 mg/kg, and the heart was removed. The septal artery, a branch of the left coronary artery, was cannulated with a small polyethylene cannula and perfused at a constant flow rate by a Harvard pump. A triangular portion of the septum with the perfusion cannula at its base was dissected and suspended. The two lower corners were held in clamps while the apex of the triangle was attached by a suture to a Statham UC4 transducer (Fig. 1). The transducer recorded only that vector of tension developed along its axis, but the proportion of total force represented by this vector remained constant throughout each experiment. A rabbit septum was accepted for study if 10.0 g or more of developed tension were recorded with a resting tension up to 10 g. Rabbit septa were perfused at 1.0 ml/min (1.5-2.0 ml/g per min). Rabbit septa perfused under these conditions maintained stable tension for 6-8 h at rates of contraction up to 120/min. At the perfusion rates described steady-state tissue water and tissue potassium and calcium contents were achieved within 45 min. The temperature of the perfusate was maintained by passing current through a 500-a power resistor placed around the metal perfusion cannula which attached to the polyethylene cannula. The muscle was placed under a rectangular plastic
1225
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1226
SHINE, TO
RECORDER
/
FIG. 1. Apparatus for studies text for description of technique. tors against artifacts in tension gular septum is shown in position. removed in diagram to facilitate
of ischemia and hypoxemia. See Tennis balls were efficient insulaand dP/dt measurements. TrianAnterior section of hood has been visualization.
hood through which nitrogen was passed at 6 liters/min. The nitrogen was bubbled through a heated water column (temperature humidification well) so as to maintain constant temperature and h .umidity around the suspended muscle. A thermistor located close to the muscle surface activated the humidification-well power controller (RFL Industries, no. 7872) which passed current through high-resistance wire wrapped around the water column. The well water temperature was continuously adjusted to maintain muscle surface temperature constant. Simultaneously a thermistor probe inserted in the .muscle monitored muscle core temperature. Sealable ports located on both sides of the hood permitted manipulation of the muscle within the hood. Within the apparatus muscle temperat ure was stable to rtO.25”C throughout each experiment including the period of no perfusion. Ischemia was produced by switching off the Harvard pump driving the perfusion syringes. With nitrogen flow of 6 literslmin the oxygen content of the environment was reduced to less than 1 ~01% (8 mmHg) by Van Slyke determination. Cessation of perfusion produced almost immediate deep cyanosis of the muscle surface which persisted throughout the period of no perfusion as continual evidence for ischemia of the surface of the muscle. Reperfusion was accomplished by starting the perfusion pump. With the seital tissue mounted it was possible to record a number of parameters continuously and simultaneously: a) frequency of contraction with the rate controlled by external stimulation; b) isometric tension, maximal rate of tension development (positive dP/&), maximal rate of relaxation (negative dP/&), time-topeak tension (TPT); c) the temperature of the muscle (*0.25”C) by means of the inserted thermistor needle; and d) environmental temperature under the dome by an inserted thermometer. The preparation is also suita-
DOUGLAS,
AND
RICCHIUTI
ble for measurements of isotopic activity in the muscle by a closely apposed G-M probe and for collection of effluent to determine radioisotopic or chemical content. Experiments in the present study were conducted at 28 or 37 t 0.25”C with heart rates of 72 beats/min, except as described subsequently. Action potentials were recorded by a flexibly mounted microelectrode as described by Woodbury and Brady (18) with the tip less than 0.5 ,um filled with 3 M KCl. A microelectrode preamplifier was placed through the dome on a balland-socket hinge which allowed manipulation before, during, and after ischemia without perturbing the dome environment (Fig. 1). Action potentials were recorded on a type 564 Tektronix storage oscilloscope and on a Brush 440 recorder at paper speeds of 125 mm/s. Results reported are only for muscles in which the microelectrode remained in the same cell for the duration of measurement. Standard solutions for perfusion contained (in mM): NaCl, 142; KCl, 4-10; MgClz, 1.0; NaH,PO,, 0.435; dextrose, 5; sodium pyruvate, 2; CaCl,, 0.5-5.0. After equilibration of the aqueous solution at 98% O,-2% CO, at 24°C red blood cells were added to a final hematocrit of 20%. As previously described (13), the buffer capacity of the red blood cells was adequate to maintain a stable pH throughout perfusion without addition of bicarbonate at 28°C. For experiments at 37°C 12 mM NaHCO, was added and the NaCl was reduced to 130 mM in order to maintain a final pH at 7.35-7.40 at both temperatures. Perfusate Pco2 was 24-26 mmHg by Van Slyke determinations. The blood perfusate contained 0.08 ml 0, per milliliter with effluent venous content of 0.04 ml 0, per milliliter at a heart rate of 160 beats/min. Homogeneity of perfusion has been previously demonstrated in these preparations and changes in the quantity of muscle perfused observed by following the washing out of 42K (1). Potassium contents calculated from 42K tissue uptake and 42K effluent analysis gave identical results and agreed with values obtained from flame photometry. A change in the amount of muscle perfused after labelling with 42K produced an abrupt rise or fall in 42K counts appearing in the efluent (1). To obtain catechol-depleted septa rabbits were injected intravenously with reserpine 3 mg/kg, 36 and 18 h prior to sacrifice. A dosage of 1 .O mglkg has been shown to produce almost complete depletion of tissue catecholamines in rabbit atria within 16-22 h (2). All values have been expressed as means t standard errors of the means. Statistical significance of a difference between two means was determined by the Student test and/or by the paired-t test. Experimental mechanical data during ischemia were fitted to give the least-squares fit on a 360/91 computer in APL language (conversational nonlinear curvefitting program by C. Clausen). During reperfusion, the experimental mechanical data were similarly fitted except in experiments with perfusates containing 5.0 mM Ca or 10 mM K (Fig. 8). Because of the shape of these curves, only the initial 6 min of reperfusion were analyzed. In all experiments the +dP/& data at 1 min were not fitted because of the delay in decline of that parameter (Fig. 3 and 4).
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1227
SEPTA
RESULTS
Sixteen septa were subjected to ischemia, eight at 37°C and eight at 28°C. The mechanical responses are demonstrated in Fig. 2A. In the upper panel a septum perfused at 37°C and 72 beatslmin is shown. Cessation of perfusion produced an abrupt and immediate decline in developed tension which fell to zero tension. Resting tension showed a decline of about 0.2 g concomitant with the loss of perfusion, then rose by 0.4 g after 10 min of ischemia. The decline in dP/& was not symmetrical. The maximal rate of relaxation, negative dP/&, declined as rapidly as did developed tension. The maximal rate of tension development, positive dP/&, showed a lmin delay before its decline began. Thereafter the rate of decline was rapid. During the first 4 min of reperfusion, tension recovered rapidly; then, after a transient decline, it gradually recovered to 87% of its preischemia values. The recovery of dP/& was symmetrical. The time course of the response to ischemia is further documented at faster paper speed, under the same conditions in the lower panels on the left (Fig. 2B). Here the asymmetrical decline of dP/&, with a delay in fall of the maximal rate of tension development, is shown during the first 5 min of ischemia. The contribution of catecholamine release to the mechanical events was evaluated in three septa from reserpinized rabbits and in three septa perfused with propranolol, lo-” M, prior to ischemia. The declines in dP/ dt and tension were similar in all six septa to those previously described. In one of the reserpinized septa, shown in Fig. 2C, there was a small transient decline in maximal rate of tension development for 20 s after cessation of perfusion. Despite this transient variation in tension and dP/dt, the decline in maximal rate of tension development lagged 1.2 min behind the declines
in maximal rate of relaxation and developed tension. In Fig. 3 the mean values for developed tension, maximal rates of relaxation, and maximal rates of tension development are shown as a percentage of preischemic control values in eight muscles made ischemic at 37°C. During 10 min of total ischemia developed tension declined to 2.9 t 0.6% of control tension. The decline was linear on a semilogarithmic plot with a rate constant (A) of 0.39 min. Upon reperfusion recovery was slower, h = 0.21 min-‘, with recovery to 73.3 t 3% of preischemic values during 15 min of reperfusion. The data were fit to the curves shown by the iterative computer technique. The best-fit equations are shown. The maximal rate of relaxation (-dP/&) also declined linearly on a semilogarithmic plot at 37°C until this rate had fallen to 10% of control values. With the exception of this terminal slowing, the maximal rate of relaxation declined with a rate constant not significantly different from that for the decline of developed tension. The rate constant for recovery was identical to that for developed tension. In comparison with developed tension and maximal rate of relaxation, the maximal rate of tension development (+dP/dt) declined more slowly during the first-2 min of ischemia. Thereafter the decline was at the same rate as that observed for the other two parameters. Upon reperfusion no lag in recovery was observed. The rate constant for decline was identical to those observed for developed tension and maximal negative dP/dt, when the 1-min value was excluded from the computer fit. The rate constant for recovery was identical to those during recovery of the other two parameters. The responses of eight muscles made ischemic for 10 min at 28°C are shown in Fig. 4. Developed tension again declined along a single exponential to 4.9 t 1.2% of control values with X = 0.29 min. This decline was y=8?
1OOK
I
( I -e-‘.**+)
/y=93t
I-e-0.2’+)
I
10 mln
6 “-5U-J E 0 t-&
c
25
i
0 I
1 min
I
I
10 min
I
FIG. 2. Mechanical responses to ischemia. A: onset of ischemia noted with 1st arrow and reperfusion with 2nd arrow. Delay in ,decline of maximal rate of tension development (+dP/dt) in contrast to immediate decline of maximal rate of relaxation and developed tension is shown. Recovery of dP/dt was symmetrical. No significant change in resting tension occurred. B: onset of ischemia is shown at faster paper speed. C: same phenomenon is shown in reserpinized septum. One-gram decline in resting tension occurred upon cessation of perfusion with transient fluctuation of maximal rate of tension development. Delay in decline of this parameter is shown. All experiments shown were at 37°C and 72 beatslmin.
! 11 0
37” c 72 6fM 1.5mM Co 4.0mM K Developed - dP/dt
/ \ ‘..I hl i
y=
l()()p37+
I 5
. . . . . . . + d,P/dj
'\\
\
fenslon
\
I 10
I -1’ 15
20
25
MINUTES FIG. 3. Mechanical function during and after ischemia at 37°C. Developed tension, maximal rate of relaxation (-dP/dt), and maximal rate of tension development (+dP/dt) as percentages of preischemia values are plotted against time on a semilogarithmic scale for 8 muscles maintained at 37°C. Vertical bars indicate * SE. Computer-fitted curves and their equations are shown. Declines and recoveries fit monoexponential functions closely.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1228
SHINE,
significantly (P < 0.001) slower than that at 37OC and would indicate a Qlo of 1.4 for this process. Upon reperfusion recovery occurred at the same rate (h = 0.23 min-l) as that observed at 37°C but the magnitude of the recovery during 15 min of reperfusion was greater, to 79 t 4% 4% of of control control values. values. The maximal rate of tension development again showed an initial delay in decline and then fell parallel to the decline in developed tension, h = 0.29 min+, and recovery of this parameter also para paralleled that lleled th .at of developed tension during the reperfusion period. The maximal rate of relaxation was the most sensitive parameter to the decrease in temperature. Although it declined along a single exponential, the rate constant, h, decreased from 0.39 min-’ at 37OCto 0.23
28°C 7251% l.5mM Co 4.0mM K - --A Developed -0 - dP/df -....-m+ dP/df
tension
Tension gme ,,[o .i..A I
50”
5. Action potentials and mechanical parameters during ischemia. This septum was perfused at 37°C and 72 beats/min. Microelectrode remained in place throughout 7 min of ischemia. Action potentials are shown above dP/dt and tension recordings at a paper speed of 125 mm/s. Time-to-peak tension was measured from onset of positive dP/dt deflection until its return to base line immediately before its negative deflection. Some variation in base line of microFIG.
AND
RICCHIUTI
minl at 28°C. The Qlo for this process was 1.9. Recovery paralleled that of the other parameters. The values for +dP/& and -dP/& as a percentage of control levels were divided by the percentage of developed tension remaining during the periods of ischemia. In all cases the value of this ratio was greater than unity. Resting tension. The effects of 10 min of total ischemia upon resting tension were usually small. When perfusion stopped blood drained from the vasculature causing a decline in resting tension which averaged 3.1 t 3.6% of developed tension at 37OCafter perfusion with blood containing 1.5 mM Ca and 0.5 t 2.1% of developed tension at 28OCwith the same perfusate. Resting tension, measured 15 min after reperfusion, increased in comparison to its preischemic values by 10.3 t 2.8% of developed tension at 37OCand 3.6 t 2.3% at 28°C. These differences were not statistically significant. However, the maximal increase in resting tension during reperfusion ofte-n occurred after 8-10 min and produced an inflection in the recovery of other mechanical parameters, particularly in high-calcium or high-potassium perfusates, as will be shown. Since cessation of perfusion could alter developed tension and dP/& by decreasing resting muscle length, resting tension was reduced by 5 g in three muscles by releasing the force transducer suture. In all casesthe declines in positive and negative dPldt were symmetrical. Time-to-peak tension and action potential duration. The decline in developed tension immediately upon cessation of perfusion, while the maximal rate of tension development was well maintained, indicated that timeto-peak tension shortened immediately. This is shown in Figs. 5 and 6. In Fig. 5 a record is shown from a septum made ischemic at 37°C. The decline in tension, shortening of TPT, and maintenance of maximal positive dP/dt are shown with action potentials recorded from a microelectrode which remained in place throughout ischemia. There is some wandering of the action potential base line, but no shortening is demonstrable at either 50 or 90% repolarization during the first 90 s of ischemia, whereas the TPT has shortened to 55% of
r--LContro
DOUGLAS,
90”
/--l..__..------md-----p 120”
360”
electrode recording was seen. Time-to-peak tension shortened within 30 s of onset of &hernia and continued to decline to 55% of control value by 90 s, whereas action potential was unchanged. Subsequently, action potential shortened and its amplitude declined. Positive dP/dt was initially maintained, though negative dP/dt declined at same time as developed tension, and time-to-peak tension fell.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1229
SEPTA 050%
APD
Minutes
6. Relationships of action potentials to mechanical parameters in ischemia. Durations of action potentials at 50 and 90% repolarization are shown as a percentage of these durations during control preischemic periods for 12 muscles in which continuous recordings were available. Developed tensions, times-to-peak tension, and maximal rates of tension development for these septa are plotted below. Vertical bars indicate + SE. Note that times-to-peak tension declined rapidly with onset of ischemia but action potential durations were well maintained for 1st min of ischemia, after which these durations did decline. Maximal rates of tension development began to decline rapidly at time that action potential duration began to shorten. FIG.
control values. Thereafter abbreviation of action potential duration and diastolic depolarization of the Imembrane occurred. In Fig. 6 these relationships are shown for 12 muscles in which a microelectrode remained in place for varying periods from 3 to 10 min after initiation of ischemia. The 12 muscles represent those among 40 septa impaled in which action potential amplitude exceeded 60 mV, significant overshoot occurred and a continuous recording from a single cell was obtained. In 10 of the 12 septa action potential amplitude exceeded 80 mV. The declines in developed tension, positive and negative dP/& are shown on a linear scale for this group of muscles. The shortening of TPT for the group is shown as well as the durations of the action potentials at 50 and 90% repolarization. Although action potential duration shortened, the TPT declined far more rapidly. Indeed action potential duration decreased by less than 3% of control after 1 min, although TPT had decreased by 25% at that time. Time-to-peak tension upon reperfusion did lengthen by lo-12% over preischemia values as previously described (17). Potassium, 10 mM. Eight septa were perfused with blood containing 10 mM K at 37°C and 72 beats/min for 45 min prior to 10 min of total ischemia. The elevation of perfusate potassium did not affect the rate of tension or negative dP/& decline (Fig. 7). The lag in decline of the maximal rate of tension development was unchanged
from that obtained in 4.0 mM K, although the rate was slightly more rapid. Upon reperfusion the rate of recovery significantly (P < 0.01) exceeded that observed in 4.0 mM K, and the absolute level of tension development significantly (P < 0.001) exceeded that observed in 4.0 mM K. Time-to-peak tension returned to preischemia duration within l-2 min after reperfusion. The improvement in developed tension was the result of a more rapid and complete recovery of the maximal rate of tension development which approached 90% of preischemic levels after 5 min of reperfusion. Effects of calcium. Five muscles were perfused with blood containing 5.0 mM Ca prior to ischemia and a second group of five muscles was perfused with blood containing 0.5 mM for 60 min prior to ischemia at 37°C and 72 beats/min. The rate of decline of all parameters was similar at high and low calcium concentrations, except for positive dP/d.L This parameter fell with h = 0.26 min-l in 5.0 mM Ca (Fig. 7) compared to X = 0.39 min -l in 1.5 mM Ca, a significant (P < 0.01) difference. The delay in decline of the maximal rate of tension development was similar at 0.5 and 5.0 mM Ca. Mechanical recovery was initially very complete at 5.0 mM Ca, though a transient similar to that obtained in 10 mM KC1 occurred. Resting tension decreased an average of 16 t 87% during ischemia in 0.5 mM Ca and increased by 6.7 t 4.4% upon reperfusion. Resting tension increased in 5.0 mM Ca by 0.7 t 1.8% during ischemia and 1.3 t 0.7% upon reperfusion. However, these values are expressed as a percentage of developed tension which was substantially greater in 5.0 mM Ca. There was no significant difference in absolute values for the change in resting tension during ischemia or reperfusion in the 0.5, 1.5, or 5.0 mM Ca muscles. DISCUSSION
The decline of mechanical function in cardiac muscle during ischemia is often more rapid than that observed during hypoxemia (3). Significant differences can be expected as a result of the substrate deprivation and failure to remove accumulated metabolic products during ischemia (4, 11) in contrast to continued perfusion without oxygen. The responses to ischemia and hypoxemia may also be profoundly influenced by the condition of the myocardium prior to ischemic or hypoxemic insults. The isol ated blood-perfused interventricular septurn provided a reproducible, we1 l-controlled model for studies of global ischemia. Perfusion prior to ischemia was conducted at physiological perfusion rates, at heart rates of 72 beatslmin, and with excellent oxygenation. Ischemia was essentially instan taneous except for oxygen extracted from blood within the muscle at the point perfusion ceased. Ischemia was total, with no possible collateral flow and, particular] .y for action potential measurements, with an environment containing so little oxygen that subendocardial cells could receive no appreciable oxygen supply by diffusion. The preparation’s temperature and surface humidity was well controlled and numerous simultaneous measurements were possible.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
SHINE,
DOUGLAS,
AND
RICCHIUTI
1.5mMCa /OmMK ---0 +df/df -....-A-df/dt
y=g4e-0.4~*~~‘ *.:
‘0
5
I 10
I 20
15
MI
I
I
5
10
I
15
20
NUTES
7. Ischemia during perfusion with 10 mM KC1 and 5.0 mM Ca. On right, declines and recoveries of mechanical function are shown on semilogarithmic plots for 8 septa made ischemic under same conditions as previously described except for elevation of potassium concentration to 10 mM. Rates of declines were similar to those at 37°C in 4.0 mM K. Recoveries were significantly (P < 0.01) more rapid and magnitudes of recovery of developed tension and maximal rates of tension development were increased in comparison to 4.0
mM K. On left, developed tension, + dP/dt, and -dP/dt are plotted on semilogarithmic scales for 5 muscles made ischemic for 10 min after perfusion for 60 min in 5.0 mM Ca at 37°C. Though recoveries of function were not significantly different from those observed in 1.5 mM Ca, decline of +dP/dt was significantly slower in 5.0 mM Ca. Developed tension and -dP/dt also began to plateau at ‘25-30% of control values, giving a less satisfactory fit to computer-derived curves.
The monoexponential decline in developed tension observed upon production of ischemia was unexpected. At 37°C it was paralleled by the rate of decline of the maximal rate of relaxation and, after a delay of 1.5-3 min, by the decline of the maxi ma1 rate of tension development. There was significant temperature dependence of the decline of mechanical function, which was most significant for the maximal rate of relaxation Q10 = 1.9. Although the decline in developed tension and maximal rate of tension development showed a Qlo of only 1.4, these temperature effects may be related to many different processes which behave in opposing ways. A &m of 1.4 does suggest that a physical process such as competition at a binding site or diffusion of a crucial substance could be the rate limiting step. The rate of recovery of function was not at all temperature dependent, and &as slower than the rate of decline. This was also consistent with a nonmetabolic process. The principal effect of temperature upon recovery was upon its magnitude, with extrapolated as well as measured recoveries of function lo-17% lower at 37°C than at 28°C. The rate of decline of the maximal rate of relaxation, which showed a Q10 of 1.9, may reflect the h igh metabolic requirements of the relaxation process. The early decline in developed tension resulted from a rapid decrease in TPT in the first l-3 min of ischemia, during which time the maximal rate of tension development (+dP/&) declined slowly. Thereafter, the decline of the latter parameter paralleled the declines of developed tension and maximal rate of relaxation. The rate of
recovery of positive dP/& was identical to that of the two other parameters. Although TPT could decrease as a consequence of a shortening action potential, these phenomena were clearly dissociated during the action potential measurements. Although action potential durations shortened, they did so much more slowly than time to peak tension. This dissociation was present during the first l-2 min of ischemia in each of 50 or more impalements in a number of locations throughout the septa. Shortening of action potential duration has been reported durin -g hypoxemia in isolated, bath-perfused muscle fibers (16). Shorteni ng became apparent 4-5 min after initiation of hypoxemia and mechanical function declined more slowly than observed in ischemia. Although attempts have been made (6) to correlate changes in tension with action potential durations in the blood-perfused Langendorff dog heart preparation, the plateau duration was maintained for E-20 min after force declined to almost 50% of control values in that study. These data (6) may reflect the problem of inhomogeneity of flow previously noted. In our model action potential shortening is not immediate but does commence after 60-90 s. Changes in action potential duration could contribute significantly to decline in maximal rate of tension development (+dP/dt), but shortening of TPT did consistently precede the action potential shortening. The mechanical responses during ischemia in 10 mM KC1 were evaluated to determine if the lag in decline of
FIG.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1231
SEPTA
positive dP/& or the rate of decline would be altered by an intervention which shortens TPT and action potential duration and decreases the intracellular-to-extracellular potassium concentration gradient. No significant effect was observed in comparison to the declines in 4.0 mM K, indeed the rate constants were almost identical (Figs. 3, 4, and 7). This was not consistent with the hypothesis that potassium accumulation itself or alteration in the action potential durations explained the mechanical events. A concentration of 10 mM K did not impair the recovery upon reperfusion after 10 min of ischemia; indeed, the rate of recovery in comparison to control levels was enhanced. Reduction of preischemic extracellular calcium concentration for 60 min to 0.5 mM had no effect upon the decline of mechanical function. Recovery was slightly but not significantly more rapid than 1.5 mM Ca. Perfusion with 5.0 mM Ca did slow the rate of decline of positive dP/dt from X = 0.39 min-l in 1.5 mM to X = 0.26 min+ in 5.0 mM Ca. In addition, the lag phenomenon was less striking, although this change largely resulted from the slower overall rate of decline of positive dP/&. The detailed mechanism by which ischemia results in loss of mechanical function is unknown. Cellular energy stores remain adequate long after mechanical function has been markedly reduced (4), and contractile proteins are intact for at least 5-10 min after the onset of ischemia (4). Moreover, the degree of reversibility observed in our septal preparations emphasizes the degree of reversible inhibition of force development in the face of total ischemia. A decrease in action potential duration prior to any evidence of diastolic depolarization has been described in hypoxia (16). Since evidence has accumulated for contraction related calcium current during the action potential plateau, this shortening of action potential duration has been suggested (4) as a mechanism whereby calcium release may be diminished during ischemia. Although this phenomenon may contribute to the declines in mechanical function observed in the septa, there was no consistent parallelism of effects. Particularly during the initial 60-90 s of ischemia, the shortening of time-to-peak tension was much more rapid than the action potential changes. Subsequent action potential changes were considerably slower in both rate and magnitude than the mechanical events. This same dissociation was shown by the data of Miller and Gilmore (6). The shortening of contraction, here the result of a decrease in time to peak tension, was demonstrated by Tennant and Wiggers (15). This decline paralleled the fall in developed tension and in the maximal rate of relaxation. Nakamaru and Schwartz (7) reported an enhanced affinity of cardiac sarcoplasmic reticulum for Ca2+ at higher H+ concentrations. This increased binding might shorten the time during which calcium acted at the myofilaments. However, the decrease in maximal rate REFERENCES 1. BLESA, E. S., G. A. LANGER, A. J. BRADY, AND S. D. SERENA. Potassium exchange in rat ventricular myocardium: its relation to rate of stimulation. Am. J. Physiol. 219: 747-754, 1970.
of relaxation occurred at the same rate as the decrease in developed tension at 37°C during the first 5 min of ischemia (Fig. 6). This result implies that a change in the rate of calcium sequestration by the sarcoplasmic reticulum did not necessarily occur. Moreover, if calcium release from the sarcoplasmic reticulum was important in the development of tension, an enhanced affinity for calcium by these sites would be expected to decrease the rate of tension development (+dP/&) immediately as TPT declines. Our current hypothesis emphasizes changes at the sarcolemmal binding sites during ischemia to expl .ain our observations, an interpretation consistent with the results of Nayler et al. (9). According to this hypothesis a proportion of such binding sites was rendered ineffective for rebinding or release of calcium per unit time of ischemia, although the rate at which calcium was released with each depolarization was un impaired. As a consequence of an intact rate of calcium release + dP/dt was well maintained. If it is assumed that the sequestration capabilities of the sarcoplasmic reticulum are maintained during this period, sequestration might reduce the available calcium at the myofilaments earlier in the contraction, thereby shortening TPT. Subsequently, as the amount of sarcolemmal calcium available further declined, +dP/dt declined at the same rate as the calcium releasing sites diminished. According to this hypothesis, the principal effect of ischemia during the initial 60-90 s was to decrease the magnitude of calcium release rather than upon the rate of release. When equilibrated with 5.0 mM Ca enough calcium was available to saturate the binding sites prior to ischemia and to antagonize the process limiting calcium release or rebinding. A depletion process of the kind described is particularly attractive because of the monoexponential decline of function and the low temperature dependence of the process. The onset of the effect immediately upon cessation of flow and the persistence of the effect as a monoexponential function while the action potential effects are discontinuous argues against an electrical explanation for this phenomenon. Further exploration of this hypothesis including a demonstration of the substance or substances responsible for these effects awaits further investigation. We believe that the isolated septum is an excellent preparation for the study of these phenomena. The search will be particular1 .y directed toward processes which will correlate with the monoexponenti .a1 behavior of the mechanical parameters. The authors express their technical assistance and to many useful discussions. This study was supported 05 from the Public Health Association Grant LACHA Borun Foundation, and the Send reprint requests to Medical Center, Los Angeles, Received
for publication
appreciation Glenn Langer
to James Buchanan and Alan J. Brady
for for
by Grants HE-05909-01 and HE-11074Service, by Los Angeles County Heart 481-Cl, by the Anna Borun and Harry Castera Foundation. K. I. Shine, Room A3-381 BRI, UCLA Calif. 90024.
29 October
1975.
A., E. ROSENGREN, A. BERTHER, AND J. NILSSON. 2. CARLSSON, Effect of reserpine in the metabolism of catecholamines. In: Psychotropic Drugs, edited by S. Garattini and V. Ghetti. Milan:
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1232 Elsevier, 1957, p. 363-372. 3. FISHER, V. J., R. A. MARTINO, R. S. HARRIS, AND F. KAVALER. Coronary flow as an independent determinant of myocardial contractile force. Am. J. PhysioZ. 217: 1127-1133, 1969. 4. KATZ, A. M. Effects of ischemia on the contractile processes of heart muscle. Am. J. Cardiol. 32: 456-460, 1973. 5. MAROKO, P. R., P. LIBBY, W. R. GINKS, C. M. BLOOR, W. %. SHELL, B. E. SOBEL, AND T. Ross, JR. Coronary artery reperfusion: early effects on local myocardial function and the extent of myocardial necrosis. J. CZin. Invest. 51: 2710-2716, 1972. 6. MILLER, D. T., AND J. P. GILMORE. Excitation contraction correlates in true ischemia. J. EZectrocardioZ. 5: 257-264, 1972. 7. NAKAMARU, Y., AND A. SCHWARTZ. Possible control of intracellular calcium metabolism by [H+]: sarcoplasmic reticulum of skeletal and cardiac muscle. Biochem. Biophys. Res. Commun. 41: 830-836, 1970. 8. NAKHJAVAN, F. K., R. PARAMESWARAN, C. Y. Lu, N. V. SRINIVASAN, AND H. GOLDBERG. Effects of hypoxia, reoxygenation, and temperature on cat papillary muscle. Am. J. Physiol. 220: 1289-1293, 1971. 9. NAYLER, W. G., J. STONE, V. CARSON, AND D. CHIPPERFIELD. Effect of ischemia on cardiac contractility and calcium exchangeability. J. MOL. CeZZ. CardioZ. 2: 125-143, 1971. 10. POOLE-WILSON, P. A., AND G. A. LANGER. The effect of pH on ionic exchange and function in rat and rabbit myocardium. Am. J. Physiol. 229: 570-581, 1975.
SHINE,
DOUGLAS,
AND
RICCHIUTI
11. ROVETTO, M. J., J. T. WHITMER, AND J. R. NEELY. Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circulation Res. 32: 699-711, 1973. 12. SHINE, K. I., AND A. DOUGLAS. Magnesium effects on ionic exchange and mechanical function in rat ventricle. Am. J. Physiol. 227: 317-324, 1974. 13. SHINE, K. I., S. D. SERENA, AND G. A. LANGER. Kinetic localization of contractile calcium in rabbit myocardium. Am. J. PhysioZ. 221: 1408-1417, 1971. 14. TATOOLES, C. J., AND W. C. RANDALL. Local ventricular bulging after acute coronary occlusion. Am. J. Physiol. 201: 451-456, 1961. 15. TENNANT, R., AND C. J. WIGGERS. The effect of coronary occlusion on myocardial contraction. Am. J. Physiol. 112: 351-361, 1935. 16. TRAUTWEIN, W., AND J. DUDEL. Aktionspotential and Kontraktion des Herzmuskels in Sauerstoffmangel. PfZuegers Arch. 263: 23-32, 1956. 17. TYBERG, J. V., L. A. YEATMAN, W. W. PARMLEY, C. W. URCHEL, AND E. H. SONNENBLICK. Effects of hypoxia on mechanics of cardiac contraction. Am. J. Physiol. 218: 1780-1788, 1970. 18. WOODBURY, J. W., AND A. J. BRADY. Intracellular recording from moving tissues with flexibly mounted ultramicroelectrode. Science 123: 100-101, 1956.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
in U.S.A.
Ischemia in isolated interventricular septa: mechanical events KENNETH I. SHINE, ANNIE M. DOUGLAS, AND NICHOLAS RICCHIUTI Departments of Physiology and Medicine and Los Angeles County Cardiovascular Laboratory, University of California at Los Angeles, Center for the Health Sciences, Los Angeles, California 90024
I., ANNIE M. DOUGLAS, AND NICHOLAS Ischemia in isolated interventricular septa: mechanical events. Am. J. Physiol. 231(4): 12251232. 1976. Isolated blood-perfused rabbit interventricular septa were adapted for studies of global ischemia by enclosure in a constant-humidity nitrogen atmosphere. During ischemia, developed tension (DT) and maximal rate of relaxation (-dP/dt> declined monoexponentially, h = 0.39 min-’ at 37°C and 72 beats/min with a Q10 of 1.4 for DT and a Q10 of 1.9 for -dP/dt. After a 60- to 90-s delay the maximal rate of tension development (+dP/dt) declined at the same rate as DT. Time-to-peak tension (TPT) shortened immediately with ischemia but action potential duration shortened after 60-90 s. Calcium at a concentration of 5 mM slowed the rate of decline of +dPldt to h = 0.26 min. Upon reperfusion after 10 min of ischemia the rates of recovery of DT, +dP/dt, and --dP/dt were similar, h = 0.21-0.23 min-‘, and were not temperature dependent. The magnitude of recovery was lo-17% less at 37°C than 28°C. Potassium at a concentration of 10 mM did not alter the rate of decline of mechanical function, but significantly (P < 0.01) increased the magnitude of mechanical recovery. The results suggest depletion and/or repletion of single compartments as the rate-limiting steps in ischemia and reperfusion. SHINE, RICCHIUTI.
reperfusion;
KENNETH
calcium;
potassium;
rabbit
septa
PRECISE MECHANISMS for loss of force development in ischemic myocardium are unknown (4). Although several satisfactory preparations are available for studies of hypoxia in isolated myocardial muscle (17), the controlled examination of ischemia has been less easily achieved. Ligation of a coronary artery in the in situ heart presents problems of blood flow via collateral coronary and bronchopulmonary vessels, thereby producing variable degrees of ischemia and hypoperfusion in the affected myocardium. Other models have included graded decreases in coronary flow without total ischemia (11). Langendorff- ‘type preparations generally require coronary perfusion rates during the control period which are severalfold greater than physiological levels, with considerable interstitial edema and potentially large heterogeneity of flow during low flow states and upon reperfusion (11). In all these preparations the venous effluent collected may originate in areas of markedly different degrees of perfusion and oxygenation. We have adapted the isolated blood-perfused interTHE
Research
ventricular septal preparation to studies of ischemia so that abrupt total ischemia can be produced with reproducible consequences under a wide variety of conditions. Ischemia and reperfusion at physiological perfusion rates can be accomplished with rapid analysis of mechanical, electrophysiological, metabolic, and ionic events. The same preparation can be used for studies of hypoxemia. Mechanical and electrophysiological results of ischemia in isolated blood-perfused septa are reported here. These data provide the basis for additional observations of metabolic changes and ionic movement in this preparation. METHODS
The experimental preparation was the isolated bloodperfused interventricular septum of 1.5 to 2.0-kg white male New Zealand rabbits according to the technique described by Shine et al. (12, 13). The animals were anticoagulated with 10 mg sodium heparin intravenously, anesthetized with an intravenous injection of pentobarbital, 40 mg/kg, and the heart was removed. The septal artery, a branch of the left coronary artery, was cannulated with a small polyethylene cannula and perfused at a constant flow rate by a Harvard pump. A triangular portion of the septum with the perfusion cannula at its base was dissected and suspended. The two lower corners were held in clamps while the apex of the triangle was attached by a suture to a Statham UC4 transducer (Fig. 1). The transducer recorded only that vector of tension developed along its axis, but the proportion of total force represented by this vector remained constant throughout each experiment. A rabbit septum was accepted for study if 10.0 g or more of developed tension were recorded with a resting tension up to 10 g. Rabbit septa were perfused at 1.0 ml/min (1.5-2.0 ml/g per min). Rabbit septa perfused under these conditions maintained stable tension for 6-8 h at rates of contraction up to 120/min. At the perfusion rates described steady-state tissue water and tissue potassium and calcium contents were achieved within 45 min. The temperature of the perfusate was maintained by passing current through a 500-a power resistor placed around the metal perfusion cannula which attached to the polyethylene cannula. The muscle was placed under a rectangular plastic
1225
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1226
SHINE, TO
RECORDER
/
FIG. 1. Apparatus for studies text for description of technique. tors against artifacts in tension gular septum is shown in position. removed in diagram to facilitate
of ischemia and hypoxemia. See Tennis balls were efficient insulaand dP/dt measurements. TrianAnterior section of hood has been visualization.
hood through which nitrogen was passed at 6 liters/min. The nitrogen was bubbled through a heated water column (temperature humidification well) so as to maintain constant temperature and h .umidity around the suspended muscle. A thermistor located close to the muscle surface activated the humidification-well power controller (RFL Industries, no. 7872) which passed current through high-resistance wire wrapped around the water column. The well water temperature was continuously adjusted to maintain muscle surface temperature constant. Simultaneously a thermistor probe inserted in the .muscle monitored muscle core temperature. Sealable ports located on both sides of the hood permitted manipulation of the muscle within the hood. Within the apparatus muscle temperat ure was stable to rtO.25”C throughout each experiment including the period of no perfusion. Ischemia was produced by switching off the Harvard pump driving the perfusion syringes. With nitrogen flow of 6 literslmin the oxygen content of the environment was reduced to less than 1 ~01% (8 mmHg) by Van Slyke determination. Cessation of perfusion produced almost immediate deep cyanosis of the muscle surface which persisted throughout the period of no perfusion as continual evidence for ischemia of the surface of the muscle. Reperfusion was accomplished by starting the perfusion pump. With the seital tissue mounted it was possible to record a number of parameters continuously and simultaneously: a) frequency of contraction with the rate controlled by external stimulation; b) isometric tension, maximal rate of tension development (positive dP/&), maximal rate of relaxation (negative dP/&), time-topeak tension (TPT); c) the temperature of the muscle (*0.25”C) by means of the inserted thermistor needle; and d) environmental temperature under the dome by an inserted thermometer. The preparation is also suita-
DOUGLAS,
AND
RICCHIUTI
ble for measurements of isotopic activity in the muscle by a closely apposed G-M probe and for collection of effluent to determine radioisotopic or chemical content. Experiments in the present study were conducted at 28 or 37 t 0.25”C with heart rates of 72 beats/min, except as described subsequently. Action potentials were recorded by a flexibly mounted microelectrode as described by Woodbury and Brady (18) with the tip less than 0.5 ,um filled with 3 M KCl. A microelectrode preamplifier was placed through the dome on a balland-socket hinge which allowed manipulation before, during, and after ischemia without perturbing the dome environment (Fig. 1). Action potentials were recorded on a type 564 Tektronix storage oscilloscope and on a Brush 440 recorder at paper speeds of 125 mm/s. Results reported are only for muscles in which the microelectrode remained in the same cell for the duration of measurement. Standard solutions for perfusion contained (in mM): NaCl, 142; KCl, 4-10; MgClz, 1.0; NaH,PO,, 0.435; dextrose, 5; sodium pyruvate, 2; CaCl,, 0.5-5.0. After equilibration of the aqueous solution at 98% O,-2% CO, at 24°C red blood cells were added to a final hematocrit of 20%. As previously described (13), the buffer capacity of the red blood cells was adequate to maintain a stable pH throughout perfusion without addition of bicarbonate at 28°C. For experiments at 37°C 12 mM NaHCO, was added and the NaCl was reduced to 130 mM in order to maintain a final pH at 7.35-7.40 at both temperatures. Perfusate Pco2 was 24-26 mmHg by Van Slyke determinations. The blood perfusate contained 0.08 ml 0, per milliliter with effluent venous content of 0.04 ml 0, per milliliter at a heart rate of 160 beats/min. Homogeneity of perfusion has been previously demonstrated in these preparations and changes in the quantity of muscle perfused observed by following the washing out of 42K (1). Potassium contents calculated from 42K tissue uptake and 42K effluent analysis gave identical results and agreed with values obtained from flame photometry. A change in the amount of muscle perfused after labelling with 42K produced an abrupt rise or fall in 42K counts appearing in the efluent (1). To obtain catechol-depleted septa rabbits were injected intravenously with reserpine 3 mg/kg, 36 and 18 h prior to sacrifice. A dosage of 1 .O mglkg has been shown to produce almost complete depletion of tissue catecholamines in rabbit atria within 16-22 h (2). All values have been expressed as means t standard errors of the means. Statistical significance of a difference between two means was determined by the Student test and/or by the paired-t test. Experimental mechanical data during ischemia were fitted to give the least-squares fit on a 360/91 computer in APL language (conversational nonlinear curvefitting program by C. Clausen). During reperfusion, the experimental mechanical data were similarly fitted except in experiments with perfusates containing 5.0 mM Ca or 10 mM K (Fig. 8). Because of the shape of these curves, only the initial 6 min of reperfusion were analyzed. In all experiments the +dP/& data at 1 min were not fitted because of the delay in decline of that parameter (Fig. 3 and 4).
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1227
SEPTA
RESULTS
Sixteen septa were subjected to ischemia, eight at 37°C and eight at 28°C. The mechanical responses are demonstrated in Fig. 2A. In the upper panel a septum perfused at 37°C and 72 beatslmin is shown. Cessation of perfusion produced an abrupt and immediate decline in developed tension which fell to zero tension. Resting tension showed a decline of about 0.2 g concomitant with the loss of perfusion, then rose by 0.4 g after 10 min of ischemia. The decline in dP/& was not symmetrical. The maximal rate of relaxation, negative dP/&, declined as rapidly as did developed tension. The maximal rate of tension development, positive dP/&, showed a lmin delay before its decline began. Thereafter the rate of decline was rapid. During the first 4 min of reperfusion, tension recovered rapidly; then, after a transient decline, it gradually recovered to 87% of its preischemia values. The recovery of dP/& was symmetrical. The time course of the response to ischemia is further documented at faster paper speed, under the same conditions in the lower panels on the left (Fig. 2B). Here the asymmetrical decline of dP/&, with a delay in fall of the maximal rate of tension development, is shown during the first 5 min of ischemia. The contribution of catecholamine release to the mechanical events was evaluated in three septa from reserpinized rabbits and in three septa perfused with propranolol, lo-” M, prior to ischemia. The declines in dP/ dt and tension were similar in all six septa to those previously described. In one of the reserpinized septa, shown in Fig. 2C, there was a small transient decline in maximal rate of tension development for 20 s after cessation of perfusion. Despite this transient variation in tension and dP/dt, the decline in maximal rate of tension development lagged 1.2 min behind the declines
in maximal rate of relaxation and developed tension. In Fig. 3 the mean values for developed tension, maximal rates of relaxation, and maximal rates of tension development are shown as a percentage of preischemic control values in eight muscles made ischemic at 37°C. During 10 min of total ischemia developed tension declined to 2.9 t 0.6% of control tension. The decline was linear on a semilogarithmic plot with a rate constant (A) of 0.39 min. Upon reperfusion recovery was slower, h = 0.21 min-‘, with recovery to 73.3 t 3% of preischemic values during 15 min of reperfusion. The data were fit to the curves shown by the iterative computer technique. The best-fit equations are shown. The maximal rate of relaxation (-dP/&) also declined linearly on a semilogarithmic plot at 37°C until this rate had fallen to 10% of control values. With the exception of this terminal slowing, the maximal rate of relaxation declined with a rate constant not significantly different from that for the decline of developed tension. The rate constant for recovery was identical to that for developed tension. In comparison with developed tension and maximal rate of relaxation, the maximal rate of tension development (+dP/dt) declined more slowly during the first-2 min of ischemia. Thereafter the decline was at the same rate as that observed for the other two parameters. Upon reperfusion no lag in recovery was observed. The rate constant for decline was identical to those observed for developed tension and maximal negative dP/dt, when the 1-min value was excluded from the computer fit. The rate constant for recovery was identical to those during recovery of the other two parameters. The responses of eight muscles made ischemic for 10 min at 28°C are shown in Fig. 4. Developed tension again declined along a single exponential to 4.9 t 1.2% of control values with X = 0.29 min. This decline was y=8?
1OOK
I
( I -e-‘.**+)
/y=93t
I-e-0.2’+)
I
10 mln
6 “-5U-J E 0 t-&
c
25
i
0 I
1 min
I
I
10 min
I
FIG. 2. Mechanical responses to ischemia. A: onset of ischemia noted with 1st arrow and reperfusion with 2nd arrow. Delay in ,decline of maximal rate of tension development (+dP/dt) in contrast to immediate decline of maximal rate of relaxation and developed tension is shown. Recovery of dP/dt was symmetrical. No significant change in resting tension occurred. B: onset of ischemia is shown at faster paper speed. C: same phenomenon is shown in reserpinized septum. One-gram decline in resting tension occurred upon cessation of perfusion with transient fluctuation of maximal rate of tension development. Delay in decline of this parameter is shown. All experiments shown were at 37°C and 72 beatslmin.
! 11 0
37” c 72 6fM 1.5mM Co 4.0mM K Developed - dP/dt
/ \ ‘..I hl i
y=
l()()p37+
I 5
. . . . . . . + d,P/dj
'\\
\
fenslon
\
I 10
I -1’ 15
20
25
MINUTES FIG. 3. Mechanical function during and after ischemia at 37°C. Developed tension, maximal rate of relaxation (-dP/dt), and maximal rate of tension development (+dP/dt) as percentages of preischemia values are plotted against time on a semilogarithmic scale for 8 muscles maintained at 37°C. Vertical bars indicate * SE. Computer-fitted curves and their equations are shown. Declines and recoveries fit monoexponential functions closely.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1228
SHINE,
significantly (P < 0.001) slower than that at 37OC and would indicate a Qlo of 1.4 for this process. Upon reperfusion recovery occurred at the same rate (h = 0.23 min-l) as that observed at 37°C but the magnitude of the recovery during 15 min of reperfusion was greater, to 79 t 4% 4% of of control control values. values. The maximal rate of tension development again showed an initial delay in decline and then fell parallel to the decline in developed tension, h = 0.29 min+, and recovery of this parameter also para paralleled that lleled th .at of developed tension during the reperfusion period. The maximal rate of relaxation was the most sensitive parameter to the decrease in temperature. Although it declined along a single exponential, the rate constant, h, decreased from 0.39 min-’ at 37OCto 0.23
28°C 7251% l.5mM Co 4.0mM K - --A Developed -0 - dP/df -....-m+ dP/df
tension
Tension gme ,,[o .i..A I
50”
5. Action potentials and mechanical parameters during ischemia. This septum was perfused at 37°C and 72 beats/min. Microelectrode remained in place throughout 7 min of ischemia. Action potentials are shown above dP/dt and tension recordings at a paper speed of 125 mm/s. Time-to-peak tension was measured from onset of positive dP/dt deflection until its return to base line immediately before its negative deflection. Some variation in base line of microFIG.
AND
RICCHIUTI
minl at 28°C. The Qlo for this process was 1.9. Recovery paralleled that of the other parameters. The values for +dP/& and -dP/& as a percentage of control levels were divided by the percentage of developed tension remaining during the periods of ischemia. In all cases the value of this ratio was greater than unity. Resting tension. The effects of 10 min of total ischemia upon resting tension were usually small. When perfusion stopped blood drained from the vasculature causing a decline in resting tension which averaged 3.1 t 3.6% of developed tension at 37OCafter perfusion with blood containing 1.5 mM Ca and 0.5 t 2.1% of developed tension at 28OCwith the same perfusate. Resting tension, measured 15 min after reperfusion, increased in comparison to its preischemic values by 10.3 t 2.8% of developed tension at 37OCand 3.6 t 2.3% at 28°C. These differences were not statistically significant. However, the maximal increase in resting tension during reperfusion ofte-n occurred after 8-10 min and produced an inflection in the recovery of other mechanical parameters, particularly in high-calcium or high-potassium perfusates, as will be shown. Since cessation of perfusion could alter developed tension and dP/& by decreasing resting muscle length, resting tension was reduced by 5 g in three muscles by releasing the force transducer suture. In all casesthe declines in positive and negative dPldt were symmetrical. Time-to-peak tension and action potential duration. The decline in developed tension immediately upon cessation of perfusion, while the maximal rate of tension development was well maintained, indicated that timeto-peak tension shortened immediately. This is shown in Figs. 5 and 6. In Fig. 5 a record is shown from a septum made ischemic at 37°C. The decline in tension, shortening of TPT, and maintenance of maximal positive dP/dt are shown with action potentials recorded from a microelectrode which remained in place throughout ischemia. There is some wandering of the action potential base line, but no shortening is demonstrable at either 50 or 90% repolarization during the first 90 s of ischemia, whereas the TPT has shortened to 55% of
r--LContro
DOUGLAS,
90”
/--l..__..------md-----p 120”
360”
electrode recording was seen. Time-to-peak tension shortened within 30 s of onset of &hernia and continued to decline to 55% of control value by 90 s, whereas action potential was unchanged. Subsequently, action potential shortened and its amplitude declined. Positive dP/dt was initially maintained, though negative dP/dt declined at same time as developed tension, and time-to-peak tension fell.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1229
SEPTA 050%
APD
Minutes
6. Relationships of action potentials to mechanical parameters in ischemia. Durations of action potentials at 50 and 90% repolarization are shown as a percentage of these durations during control preischemic periods for 12 muscles in which continuous recordings were available. Developed tensions, times-to-peak tension, and maximal rates of tension development for these septa are plotted below. Vertical bars indicate + SE. Note that times-to-peak tension declined rapidly with onset of ischemia but action potential durations were well maintained for 1st min of ischemia, after which these durations did decline. Maximal rates of tension development began to decline rapidly at time that action potential duration began to shorten. FIG.
control values. Thereafter abbreviation of action potential duration and diastolic depolarization of the Imembrane occurred. In Fig. 6 these relationships are shown for 12 muscles in which a microelectrode remained in place for varying periods from 3 to 10 min after initiation of ischemia. The 12 muscles represent those among 40 septa impaled in which action potential amplitude exceeded 60 mV, significant overshoot occurred and a continuous recording from a single cell was obtained. In 10 of the 12 septa action potential amplitude exceeded 80 mV. The declines in developed tension, positive and negative dP/& are shown on a linear scale for this group of muscles. The shortening of TPT for the group is shown as well as the durations of the action potentials at 50 and 90% repolarization. Although action potential duration shortened, the TPT declined far more rapidly. Indeed action potential duration decreased by less than 3% of control after 1 min, although TPT had decreased by 25% at that time. Time-to-peak tension upon reperfusion did lengthen by lo-12% over preischemia values as previously described (17). Potassium, 10 mM. Eight septa were perfused with blood containing 10 mM K at 37°C and 72 beats/min for 45 min prior to 10 min of total ischemia. The elevation of perfusate potassium did not affect the rate of tension or negative dP/& decline (Fig. 7). The lag in decline of the maximal rate of tension development was unchanged
from that obtained in 4.0 mM K, although the rate was slightly more rapid. Upon reperfusion the rate of recovery significantly (P < 0.01) exceeded that observed in 4.0 mM K, and the absolute level of tension development significantly (P < 0.001) exceeded that observed in 4.0 mM K. Time-to-peak tension returned to preischemia duration within l-2 min after reperfusion. The improvement in developed tension was the result of a more rapid and complete recovery of the maximal rate of tension development which approached 90% of preischemic levels after 5 min of reperfusion. Effects of calcium. Five muscles were perfused with blood containing 5.0 mM Ca prior to ischemia and a second group of five muscles was perfused with blood containing 0.5 mM for 60 min prior to ischemia at 37°C and 72 beats/min. The rate of decline of all parameters was similar at high and low calcium concentrations, except for positive dP/d.L This parameter fell with h = 0.26 min-l in 5.0 mM Ca (Fig. 7) compared to X = 0.39 min -l in 1.5 mM Ca, a significant (P < 0.01) difference. The delay in decline of the maximal rate of tension development was similar at 0.5 and 5.0 mM Ca. Mechanical recovery was initially very complete at 5.0 mM Ca, though a transient similar to that obtained in 10 mM KC1 occurred. Resting tension decreased an average of 16 t 87% during ischemia in 0.5 mM Ca and increased by 6.7 t 4.4% upon reperfusion. Resting tension increased in 5.0 mM Ca by 0.7 t 1.8% during ischemia and 1.3 t 0.7% upon reperfusion. However, these values are expressed as a percentage of developed tension which was substantially greater in 5.0 mM Ca. There was no significant difference in absolute values for the change in resting tension during ischemia or reperfusion in the 0.5, 1.5, or 5.0 mM Ca muscles. DISCUSSION
The decline of mechanical function in cardiac muscle during ischemia is often more rapid than that observed during hypoxemia (3). Significant differences can be expected as a result of the substrate deprivation and failure to remove accumulated metabolic products during ischemia (4, 11) in contrast to continued perfusion without oxygen. The responses to ischemia and hypoxemia may also be profoundly influenced by the condition of the myocardium prior to ischemic or hypoxemic insults. The isol ated blood-perfused interventricular septurn provided a reproducible, we1 l-controlled model for studies of global ischemia. Perfusion prior to ischemia was conducted at physiological perfusion rates, at heart rates of 72 beatslmin, and with excellent oxygenation. Ischemia was essentially instan taneous except for oxygen extracted from blood within the muscle at the point perfusion ceased. Ischemia was total, with no possible collateral flow and, particular] .y for action potential measurements, with an environment containing so little oxygen that subendocardial cells could receive no appreciable oxygen supply by diffusion. The preparation’s temperature and surface humidity was well controlled and numerous simultaneous measurements were possible.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
SHINE,
DOUGLAS,
AND
RICCHIUTI
1.5mMCa /OmMK ---0 +df/df -....-A-df/dt
y=g4e-0.4~*~~‘ *.:
‘0
5
I 10
I 20
15
MI
I
I
5
10
I
15
20
NUTES
7. Ischemia during perfusion with 10 mM KC1 and 5.0 mM Ca. On right, declines and recoveries of mechanical function are shown on semilogarithmic plots for 8 septa made ischemic under same conditions as previously described except for elevation of potassium concentration to 10 mM. Rates of declines were similar to those at 37°C in 4.0 mM K. Recoveries were significantly (P < 0.01) more rapid and magnitudes of recovery of developed tension and maximal rates of tension development were increased in comparison to 4.0
mM K. On left, developed tension, + dP/dt, and -dP/dt are plotted on semilogarithmic scales for 5 muscles made ischemic for 10 min after perfusion for 60 min in 5.0 mM Ca at 37°C. Though recoveries of function were not significantly different from those observed in 1.5 mM Ca, decline of +dP/dt was significantly slower in 5.0 mM Ca. Developed tension and -dP/dt also began to plateau at ‘25-30% of control values, giving a less satisfactory fit to computer-derived curves.
The monoexponential decline in developed tension observed upon production of ischemia was unexpected. At 37°C it was paralleled by the rate of decline of the maximal rate of relaxation and, after a delay of 1.5-3 min, by the decline of the maxi ma1 rate of tension development. There was significant temperature dependence of the decline of mechanical function, which was most significant for the maximal rate of relaxation Q10 = 1.9. Although the decline in developed tension and maximal rate of tension development showed a Qlo of only 1.4, these temperature effects may be related to many different processes which behave in opposing ways. A &m of 1.4 does suggest that a physical process such as competition at a binding site or diffusion of a crucial substance could be the rate limiting step. The rate of recovery of function was not at all temperature dependent, and &as slower than the rate of decline. This was also consistent with a nonmetabolic process. The principal effect of temperature upon recovery was upon its magnitude, with extrapolated as well as measured recoveries of function lo-17% lower at 37°C than at 28°C. The rate of decline of the maximal rate of relaxation, which showed a Q10 of 1.9, may reflect the h igh metabolic requirements of the relaxation process. The early decline in developed tension resulted from a rapid decrease in TPT in the first l-3 min of ischemia, during which time the maximal rate of tension development (+dP/&) declined slowly. Thereafter, the decline of the latter parameter paralleled the declines of developed tension and maximal rate of relaxation. The rate of
recovery of positive dP/& was identical to that of the two other parameters. Although TPT could decrease as a consequence of a shortening action potential, these phenomena were clearly dissociated during the action potential measurements. Although action potential durations shortened, they did so much more slowly than time to peak tension. This dissociation was present during the first l-2 min of ischemia in each of 50 or more impalements in a number of locations throughout the septa. Shortening of action potential duration has been reported durin -g hypoxemia in isolated, bath-perfused muscle fibers (16). Shorteni ng became apparent 4-5 min after initiation of hypoxemia and mechanical function declined more slowly than observed in ischemia. Although attempts have been made (6) to correlate changes in tension with action potential durations in the blood-perfused Langendorff dog heart preparation, the plateau duration was maintained for E-20 min after force declined to almost 50% of control values in that study. These data (6) may reflect the problem of inhomogeneity of flow previously noted. In our model action potential shortening is not immediate but does commence after 60-90 s. Changes in action potential duration could contribute significantly to decline in maximal rate of tension development (+dP/dt), but shortening of TPT did consistently precede the action potential shortening. The mechanical responses during ischemia in 10 mM KC1 were evaluated to determine if the lag in decline of
FIG.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
ISCHEMIA
IN
INTERVENTRICULAR
1231
SEPTA
positive dP/& or the rate of decline would be altered by an intervention which shortens TPT and action potential duration and decreases the intracellular-to-extracellular potassium concentration gradient. No significant effect was observed in comparison to the declines in 4.0 mM K, indeed the rate constants were almost identical (Figs. 3, 4, and 7). This was not consistent with the hypothesis that potassium accumulation itself or alteration in the action potential durations explained the mechanical events. A concentration of 10 mM K did not impair the recovery upon reperfusion after 10 min of ischemia; indeed, the rate of recovery in comparison to control levels was enhanced. Reduction of preischemic extracellular calcium concentration for 60 min to 0.5 mM had no effect upon the decline of mechanical function. Recovery was slightly but not significantly more rapid than 1.5 mM Ca. Perfusion with 5.0 mM Ca did slow the rate of decline of positive dP/dt from X = 0.39 min-l in 1.5 mM to X = 0.26 min+ in 5.0 mM Ca. In addition, the lag phenomenon was less striking, although this change largely resulted from the slower overall rate of decline of positive dP/&. The detailed mechanism by which ischemia results in loss of mechanical function is unknown. Cellular energy stores remain adequate long after mechanical function has been markedly reduced (4), and contractile proteins are intact for at least 5-10 min after the onset of ischemia (4). Moreover, the degree of reversibility observed in our septal preparations emphasizes the degree of reversible inhibition of force development in the face of total ischemia. A decrease in action potential duration prior to any evidence of diastolic depolarization has been described in hypoxia (16). Since evidence has accumulated for contraction related calcium current during the action potential plateau, this shortening of action potential duration has been suggested (4) as a mechanism whereby calcium release may be diminished during ischemia. Although this phenomenon may contribute to the declines in mechanical function observed in the septa, there was no consistent parallelism of effects. Particularly during the initial 60-90 s of ischemia, the shortening of time-to-peak tension was much more rapid than the action potential changes. Subsequent action potential changes were considerably slower in both rate and magnitude than the mechanical events. This same dissociation was shown by the data of Miller and Gilmore (6). The shortening of contraction, here the result of a decrease in time to peak tension, was demonstrated by Tennant and Wiggers (15). This decline paralleled the fall in developed tension and in the maximal rate of relaxation. Nakamaru and Schwartz (7) reported an enhanced affinity of cardiac sarcoplasmic reticulum for Ca2+ at higher H+ concentrations. This increased binding might shorten the time during which calcium acted at the myofilaments. However, the decrease in maximal rate REFERENCES 1. BLESA, E. S., G. A. LANGER, A. J. BRADY, AND S. D. SERENA. Potassium exchange in rat ventricular myocardium: its relation to rate of stimulation. Am. J. Physiol. 219: 747-754, 1970.
of relaxation occurred at the same rate as the decrease in developed tension at 37°C during the first 5 min of ischemia (Fig. 6). This result implies that a change in the rate of calcium sequestration by the sarcoplasmic reticulum did not necessarily occur. Moreover, if calcium release from the sarcoplasmic reticulum was important in the development of tension, an enhanced affinity for calcium by these sites would be expected to decrease the rate of tension development (+dP/&) immediately as TPT declines. Our current hypothesis emphasizes changes at the sarcolemmal binding sites during ischemia to expl .ain our observations, an interpretation consistent with the results of Nayler et al. (9). According to this hypothesis a proportion of such binding sites was rendered ineffective for rebinding or release of calcium per unit time of ischemia, although the rate at which calcium was released with each depolarization was un impaired. As a consequence of an intact rate of calcium release + dP/dt was well maintained. If it is assumed that the sequestration capabilities of the sarcoplasmic reticulum are maintained during this period, sequestration might reduce the available calcium at the myofilaments earlier in the contraction, thereby shortening TPT. Subsequently, as the amount of sarcolemmal calcium available further declined, +dP/dt declined at the same rate as the calcium releasing sites diminished. According to this hypothesis, the principal effect of ischemia during the initial 60-90 s was to decrease the magnitude of calcium release rather than upon the rate of release. When equilibrated with 5.0 mM Ca enough calcium was available to saturate the binding sites prior to ischemia and to antagonize the process limiting calcium release or rebinding. A depletion process of the kind described is particularly attractive because of the monoexponential decline of function and the low temperature dependence of the process. The onset of the effect immediately upon cessation of flow and the persistence of the effect as a monoexponential function while the action potential effects are discontinuous argues against an electrical explanation for this phenomenon. Further exploration of this hypothesis including a demonstration of the substance or substances responsible for these effects awaits further investigation. We believe that the isolated septum is an excellent preparation for the study of these phenomena. The search will be particular1 .y directed toward processes which will correlate with the monoexponenti .a1 behavior of the mechanical parameters. The authors express their technical assistance and to many useful discussions. This study was supported 05 from the Public Health Association Grant LACHA Borun Foundation, and the Send reprint requests to Medical Center, Los Angeles, Received
for publication
appreciation Glenn Langer
to James Buchanan and Alan J. Brady
for for
by Grants HE-05909-01 and HE-11074Service, by Los Angeles County Heart 481-Cl, by the Anna Borun and Harry Castera Foundation. K. I. Shine, Room A3-381 BRI, UCLA Calif. 90024.
29 October
1975.
A., E. ROSENGREN, A. BERTHER, AND J. NILSSON. 2. CARLSSON, Effect of reserpine in the metabolism of catecholamines. In: Psychotropic Drugs, edited by S. Garattini and V. Ghetti. Milan:
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
1232 Elsevier, 1957, p. 363-372. 3. FISHER, V. J., R. A. MARTINO, R. S. HARRIS, AND F. KAVALER. Coronary flow as an independent determinant of myocardial contractile force. Am. J. PhysioZ. 217: 1127-1133, 1969. 4. KATZ, A. M. Effects of ischemia on the contractile processes of heart muscle. Am. J. Cardiol. 32: 456-460, 1973. 5. MAROKO, P. R., P. LIBBY, W. R. GINKS, C. M. BLOOR, W. %. SHELL, B. E. SOBEL, AND T. Ross, JR. Coronary artery reperfusion: early effects on local myocardial function and the extent of myocardial necrosis. J. CZin. Invest. 51: 2710-2716, 1972. 6. MILLER, D. T., AND J. P. GILMORE. Excitation contraction correlates in true ischemia. J. EZectrocardioZ. 5: 257-264, 1972. 7. NAKAMARU, Y., AND A. SCHWARTZ. Possible control of intracellular calcium metabolism by [H+]: sarcoplasmic reticulum of skeletal and cardiac muscle. Biochem. Biophys. Res. Commun. 41: 830-836, 1970. 8. NAKHJAVAN, F. K., R. PARAMESWARAN, C. Y. Lu, N. V. SRINIVASAN, AND H. GOLDBERG. Effects of hypoxia, reoxygenation, and temperature on cat papillary muscle. Am. J. Physiol. 220: 1289-1293, 1971. 9. NAYLER, W. G., J. STONE, V. CARSON, AND D. CHIPPERFIELD. Effect of ischemia on cardiac contractility and calcium exchangeability. J. MOL. CeZZ. CardioZ. 2: 125-143, 1971. 10. POOLE-WILSON, P. A., AND G. A. LANGER. The effect of pH on ionic exchange and function in rat and rabbit myocardium. Am. J. Physiol. 229: 570-581, 1975.
SHINE,
DOUGLAS,
AND
RICCHIUTI
11. ROVETTO, M. J., J. T. WHITMER, AND J. R. NEELY. Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circulation Res. 32: 699-711, 1973. 12. SHINE, K. I., AND A. DOUGLAS. Magnesium effects on ionic exchange and mechanical function in rat ventricle. Am. J. Physiol. 227: 317-324, 1974. 13. SHINE, K. I., S. D. SERENA, AND G. A. LANGER. Kinetic localization of contractile calcium in rabbit myocardium. Am. J. PhysioZ. 221: 1408-1417, 1971. 14. TATOOLES, C. J., AND W. C. RANDALL. Local ventricular bulging after acute coronary occlusion. Am. J. Physiol. 201: 451-456, 1961. 15. TENNANT, R., AND C. J. WIGGERS. The effect of coronary occlusion on myocardial contraction. Am. J. Physiol. 112: 351-361, 1935. 16. TRAUTWEIN, W., AND J. DUDEL. Aktionspotential and Kontraktion des Herzmuskels in Sauerstoffmangel. PfZuegers Arch. 263: 23-32, 1956. 17. TYBERG, J. V., L. A. YEATMAN, W. W. PARMLEY, C. W. URCHEL, AND E. H. SONNENBLICK. Effects of hypoxia on mechanics of cardiac contraction. Am. J. Physiol. 218: 1780-1788, 1970. 18. WOODBURY, J. W., AND A. J. BRADY. Intracellular recording from moving tissues with flexibly mounted ultramicroelectrode. Science 123: 100-101, 1956.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 17, 2019.
