Effects of Ischemia and Hypercarbic Acidosis on Myocyte Calcium

1525
Effects of Ischemia and Hypercarbic Acidosis
on Myocyte Calcium Transients, Contraction,
and pHi in Perfused Rabbit Hearts
Rajendra Mohabir, Hon-Chi Lee, Robert W. Kurz, and William T. Clusin
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
The time courses of changes in pHI and cytosolic calcium were compared in isolated perfused
rabbit hearts with the use of the calcium-sensitive fluorescent indicator indo-1 and the pH
indicator 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF). Cell-permeant forms of
these indicators were loaded into myocytes by arterial infusion or by direct infusion into the
extravascular space. Indo-1 fluorescence was recorded from the epicardial surface of the left
ventricle at an excitation wavelength of 360 nm and emission wavelengths of 400 and 550 nm.
BCECF fluorescence was recorded at an excitation wavelength of 490 nm and an emission
wavelength of 530 nm. Calibration procedures were developed for each indicator that allowed
[Ca2"], and pHi to be quantified during ischemia. Global ischemia decreased contractility and
caused a rapid increase in both the systolic and end-diastolic levels of the calcium transients.
Ninety seconds of ischemia increased peak systolic [Ca2"]i from 609±29 to 1,341±+159 nM,
while end-diastolic [Ca2"]I increased from 315+±25 to 553+52 nM. The observed increase in
diastolic [Ca21]1 was shown not to arise from indo-1-loaded endothelial cells. The initial
increase in [Ca21]i was followed by a gradual decline and then a secondary rise occurring
between 5 and 15 minutes of ischemia. In contrast, ischemia caused a monotonic decrease in
pH; from a baseline of 7.03+0.06 to 6.83±0.02 after 2 minutes, 6.32±0.1 after 10 minutes, and
6.11±0.04 after 15 minutes. Perfusion of hearts with acidified (hypercarbic) saline increased
the systolic and diastolic levels of the calcium transients, but only when pH; fell below a
threshold value, which was more acidic than values achieved during the first 2 minutes of
ischemia (6.83+±0.03). Lesser degrees of acidification caused a decrease in contractility but did
not affect the calcium transients. Effects of pHi on the calcium transients were not due to
altered calcium sensitivity of indo-1. These results suggest that cytosolic acidification may
contribute to the increase in [Ca2`]; during the first 15 minutes of global ischemia, but the
[Ca2`J, increase during the first 2 minutes is mediated by other factors. (Circulation Research
1991;69:1525-1537)
Recent studies have revealed that an increase
in [Ca2"]i occurs during the reversible phase
of myocardial ischemia, when contraction
strength is declining.1'2 Although the decline in contraction strength was initially taken as evidence
From the Division of Cardiovascular Medicine, Falk Cardiovascular Research Center, Stanford University School of Medicine,
Stanford, Calif.
Supported by a grant from the National Institutes of Health
(HL-32093) and by a Grant-in-Aid from the American Heart
Association, California Affiliate. R.M. and H.-C.L. were recipients
of individual postdoctoral fellowships from the Canadian Heart
Foundation and the California Affiliate of the American Heart
Association. R.W.K. was supported by the Max Kade Foundation,
and W.T.C. was an Established Investigator of the American
Heart Association.
Address for correspondence: William T. Clusin, MD, Division
of Cardiovascular Medicine, Falk Cardiovascular Research Center, Stanford University School of Medicine, Stanford, CA 94305.
Received February 13, 1989; accepted June 28, 1991.
against an early [Ca2"]i increase, several factors are
now believed to decrease the sensitivity of the myofilaments to Ca2' during ischemia3 and prevent the
increase in [Ca2"]i from being manifested as increased force. One of these factors is intracellular
acidosis.4 An early decrease in pHi has been observed
during ischemia by a variety of techniques (see
Reference 5 for review), and there is evidence that
the degree of acidification that occurs in the first few
minutes of ischemia would be sufficient to reduce
contraction strength. However, continuous measurement of pHi has only recently become feasible in the
intact heart.
Intracellular acidification may also be a causative
factor in mediating the increase in [Ca2"]i during
ischemia. Recordings of calcium transients in aequorin-loaded papillary muscles show that acidification with lactate or CO2 causes a rapid increase in the
1526
Circulation Research Vol 69, No 6 December 1991
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
peak systolic level of the calcium transients.6,7 A
resting [Ca2"]i increase has also been recorded in
acidified cardiac fibers impaled with calcium-selective
intracellular electrodes.8 These observations have
fueled speculation that acidification of the cytoplasm
during ischemia may contribute to, or perhaps wholly
explain, the concomitant increase in [Ca2+1i.9-11
In the present study we used cell-permeant fluorescent indicators to obtain measurements of pHi and
[Ca2+]i from the left ventricular surface of rabbit
hearts during global ischemia and during perfusion
with C02-enriched saline. pHi was measured with
2 ',7 ' -bis(2-carboxyethyl)- 5(6)-carboxyfluorescein
(BCECF),12-14 whereas [Ca2+i] was measured with
indo-1. The indo-1 method was refined to permit
conversion of fluorescence to [Ca2+]i and to allow
examination of whether the increase in diastolic
[Ca 2]i during ischemia reflects changes occurring in
the cardiac myocytes. These measurements confirm
that cytoplasmic acidification can contribute to the
decrease in contraction strength and the increase in
[Ca 2]i during the first 15 minutes of ischemia. However, the early elevation of myocyte calcium transients during the first 2 minutes of ischemia cannot
be accounted for by cytoplasmic acidification and
must therefore be due to other causes.
Materials and Methods
Isolated Heart Preparation
Male albino New Zealand rabbits (1.8-2.2 kg) were
killed by cervical dislocation. The heart was rapidly
removed and perfused with a saline solution via the
aorta at a constant flow rate of 20-30 ml/min. The
perfusate contained (mM) NaCl 115, KCl 4.7, CaCl2
2.0, MgCl2 0.7, NaHCO3 28, NaH2P04 0.5, glucose 20,
and probenecid 0.3 as well as insulin (10 units/l) and
bovine serum albumin (40 mg/l). The pH was adjusted
to 7.4. Probenecid has been shown to prevent loss of
tetracarboxylate fluorescent indicators from cells by
blocking an organic anion transport system.15 The
buffer was equilibrated with 95% 02-5% CO2 and
heated to maintain the heart at 30+ 1°C. Left ventricular pressure was recorded with an isovolumic intracavitary latex balloon containing a fiberoptic pressure
transducer (Camino Laboratories). The cannulated
latex balloon was inserted through the mitral orifice of
the left ventricle. The cannula was secured by sutures
placed at the lower left atrium. A syringe filled with
deaerated water was attached to the cannula and was
used to fill the balloon. The heart was paced at 180
beats/min using an epicardial plunge electrode placed
in the right ventricle.
Optical Recording Apparatus
Calcium transients were recorded from the epicardial surface of the left ventricle as previously reported.1 Illumination from a 100 W mercury vapor lamp
was filtered at 360±5 nm and directed via a silica
fiberoptic cable onto the surface of the heart. The
fiberoptic cable was attached to the heart by a plastic
sleeve and rubber girdle to minimize relative motion.
Ultraviolet illumination was confined to a circular
region on the left ventricular epicardial surface 1 cm
in diameter. Fluorescence emissions were collected
by a ring of eight coaxial fiberoptics and were directed through a beam splitter into two photomultipliers (Hamamatsu) fitted with optical band-pass
filters (400±5 and 550±5 nm). The output of the
photomultipliers was passed into an electronic ratio
circuit to obtain the fluorescence ratio (F400/F550).
The output of the ratio signal was filtered by a
low-pass filter at settings of 30-70 Hz and recorded
on a strip-chart recorder.
Vascular and Extravascular Loading Techniques
Indo-1 AM was solubilized in anhydrous dimethyl
sulfoxide containing pluronic F-127 (25% wt/vol) and
infused at a final concentration of 2.5 ,uM with 5%
fetal calf serum. Initial experiments were performed
in hearts that had been perfused with indo-1 AM for
30 minutes followed by a 30-minute washout to
eliminate extracellular indo-1 AM. High-quality calcium transients were obtained by this technique, but
several factors suggest that much of the total fluorescence arose from nonmyocytes. 1) The total fluorescence at 400 and 550 nm is 14- to 20-fold greater than
the net amplitude of the calcium transients.12 A
baseline fluorescence of this magnitude is not expected from single myocyte experiments or from the
in vitro spectra of indo-1. 2) The maximum calciumdependent fluorescence level, determined with high
calcium and ionophore concentrations, is also too
large in relation to the calcium transients.' 3) Infusion of bradykinin (10-5 M) into the heart produced
a parallel shift in the systolic and diastolic levels of
the calcium transients, even though bradykinin (10`5
M) had no effect on [Ca 2]i in isolated rabbit ventricular myocytes (M. Lauer, unpublished observations).
Bradykinin is known to increase [Ca 21i in vascular
endothelial cells, and therefore, uptake of indo-1 AM
by endothelium could explain the effects of bradykinin on surface fluorescence in the intact heart.16
To minimize the uptake of indo-1 AM by endothelial cells, indo-1 AM was introduced directly into the
extravascular space by intramyocardial infusion. A
25-gauge hypodermic needle was inserted 1-2 mm
beneath the epicardium of the left ventricle, at the
site where the fiberoptic cable was attached. Solution
containing indo-l AM was then infused by a syringe
pump at 0.5 ml/min for 30 minutes. An additional
period of 30 minutes was allowed for removal of
indo-1 AM after the syringe pump was turned off.
This method produced a bright region of localized
fluorescence 6-10 mm in diameter from which calcium transients could be recorded.
The extent of nonmyocyte fluorescence in hearts
loaded by the extravascular infusion was tested with
bradykinin (10-5 M). The response of the calcium
transient to bradykinin was reduced by extravascular
loading and could be eliminated completely by pretreatment of the heart with a low concentration of
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
ionomycin. Infusion of 100 nM ionomycin caused a
parallel upward shift in the peak and baseline levels of
the calcium transients, similar to that produced by
bradykinin. This response was not due to fluorescence
changes in myocytes, because diastolic ventricular pressure did not increase. Addition of i0-` M bradykinin in
the presence of ionomycin had no effect on the calcium
transients. The bradykinin response was also blocked
by ionomycin in cultured rat aortic endothelial cells
containing indo-1 (authors' unpublished observations).
We infer that elevation of endothelial [Ca2+]i is the
mechanism by which ionomycin blocks the bradykinin
response. The effects of ionomycin on the bradykinin
response are temporary if the infusion is discontinued
after 5-10 minutes, but prolonged exposure (45 minutes) causes permanent loss of the response, even
though the myocardial calcium transients remain unchanged. Loss of the bradykinin response is accompanied by a reduction in the baseline fluorescence at 400
and 550 nm, with no reduction (and therefore a relative
increase) in the phasic component of the calcium
transient. These observations suggest that there is
selective loss of indo-1 from nonmyocytes after prolonged ionomycin treatment. The basis for such an
effect is not known, but experiments in fura-2-loaded
mast cells show loss of the indicator from the cytoplasm
during calcium-mediated exocytosis.17 Loss of indo-1
has been observed in isolated muscle cells exposed to
oxygen radicals,18 and it is possible that the nitric oxide
radical, which is thought to be produced in response to
ionomycin, has similar effects.
Calibration Procedure
Calibration of myocyte [Ca2"]i is feasible if there is
no significant calcium-dependent fluorescence from
nonmyocytes. Calibration was undertaken whenever
the response to bradykinin was shown to be absent.
The calibration procedure was similar to that reported previously.19 This method involves determination of maximum (Fm,a) and minimum (Fmin) fluorescence at the 400±+5 nm emission wavelength using
ionomycin. The heart was first infused with 100 mM
CaCl2 plus 1.5 ,uM ionomycin to obtain Fm. This
solution also contained 5 mM HEPES (pH 7.4) along
with 6% fetal calf serum to prevent binding of
ionomycin to the tubing. In contrast to low concentrations of ionomycin, this solution produced a sustained contracture that was stronger than the normal
systole. The heart was then perfused with a solution
containing 1.5 ,uM ionomycin and 100 mM MnCl2.
Entry of Mn21 into myocytes was also facilitated by
ionomycin, resulting in quenching of indo-1 fluorescence. Fmjn, the indo-1 fluorescence in the absence of
Ca +,was then given by the equation
Fmin==FMn2+ + O.18(Fmax-FMn2+)
where FMn2+ is the fluorescence measured after infusion of MnCI2 and 0.18 is a factor derived from
measurements performed in cuvettes.20 The concentration of [Ca2+]; was then obtained as
1527
[Ca2+1] i=Kd(F-Fmin)/(Fmax-F)
where F is the fluorescence measured and Kd is 250 nM.
In some experiments the fluorescence ratio F400/
F430 was used for calibration. F430 is fluorescence at
the in vivo isosbestic wavelength of indo-1.21 The
F40J/F430 ratio cancels motion artifact and autofluorescence changes during ischemia.2 For most measurements including Fmax, the F400 and F400/F430 signals
can be used interchangeably. However, F400/F430 cannot be used to measure Fmin because of the quenching
effect of Mn21 on the F430 signal. Recordings at 400
nm were therefore obtained in all experiments for
which a complete calibration was performed.
pHi Recordings
Recordings of pHi were obtained by perfusing
hearts with 2.8 ,M BCECF AM. The fiberoptic
apparatus used to measure the fluorescence of
BCECF was similar to that used to record indo-1
fluorescence transients. Fluorescence excitation was
provided by a 50 W halogen lamp filtered at 490±5
nm, with fluorescence emissions recorded at 530±5
nm. In some experiments, the excitation filter was
manually switched between 490±5 and 440±5 nm,
the latter being a frequency at which the 530 nm
fluorescence emission does not change as a function
of pH. Fluorescence excited at 440 nm did not
change appreciably during prolonged (1-2-hour) periods of recording and remained constant during
infusion of C02-rich saline, suggesting dye loss was
minimal during periods of recording in the presence
of probenecid (see Reference 13). BCECF AM was
dissolved in a dimethyl sulfoxide-pluronic solution
(25% wt/vol) and infused into the heart for 30
minutes followed by a 30-minute washout. The final
concentration of BCECF AM in the loading solution
was 2.8 4uM. This loading procedure caused fluorescence emissions to increase sevenfold to 15-fold
(mean, 9.8±1.1-fold; n=5), as compared with autofluorescence (excitation frequency, 490 nm). At the
end of each experiment, fluorescence emission was
calibrated by exposing the heart to a high [K']
solution containing 10-5 M nigericin, a H+-K' antiporter, which clamps the pHi with buffer pH (pHO)
when external and intracellular K' are the same. The
calibration solution contained (mM) KCl 150,
K2HPO4 2.5, CaCl2 1.0, MgCl2 1.0, glucose 10, and
HEPES 20. The calibration solution was adjusted to
four different pH values by addition of KOH (1N or
1ON) and infused into the heart until cellular fluorescence reached a new steady state (about 250 ml
each). The mean pH values of the four calibration
solutions were 7.26+0.07, 6.84±0.08, 6.35 ±0.05, and
5.79±0.09. A fluorescence versus pHi curve was
obtained at the end of each experiment, and pHi
values were derived by linear interpolation of fluorescence calibrations.
1528
Circulation Research Vol 69, No 6 December 1991
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
Simultaneous Calcium Transient and pHi Recordings
In some experiments, indo-1 AM and BCECF AM
were loaded simultaneously for 30 minutes followed
by a 30-minute washout. Two sets of similar fiberoptic cables were used for recordings of calcium transients and pHi. These sets of cables were positioned
so that they recorded from adjacent 1-cm regions of
the left ventricular surface. Optical interference between the cable systems was absent because neither
signal was affected by switching on or off the lamp
providing excitation to the opposite set of cables.
Indo-1 fluorescence was recorded only at 400 nm for
these experiments since the 550 nm emissions could
be contaminated by fluorescence energy transfer in
which light emitted by indo-1 (490 nm) produced
secondary excitation of BCECF (excitation, 490 nm;
emission, 530 nm). This effect would not be present
in the 400 nm emissions since the emission spectrum
of BCECF falls off sharply at wavelengths less than
500 nm.14 Recordings at the isosbestic wavelength of
indo-1 (430 nm) confirmed that the calcium transients recorded at single wavelengths were essentially
free of motion artifact (not shown).
Experimental Intervention
Global ischemia was produced by complete cessation of coronary flow. The temperature of the heart
during ischemia was maintained at 30+ 1°C by the use
of a humidified warm-air convection system. Acidosis
was produced by perfusion of the heart with a
hypercarbic saline solution as described previously.2
The hypercarbic saline solution was equilibrated with
oxygen and varying amounts of CO2 to obtain pH
values at graded intervals between 6.3 and 7.4. The
Pco2, Po2, and pH of the hypercarbic solutions were
measured with a blood gas analyzer (Corning Instruments) before infusion into the heart.
All recordings were displayed on a Gould-Brush
strip-chart recorder.
Chemicals
Probenecid, nigericin, and bovine serum albumin
were obtained from Sigma Chemical Co., St. Louis,
Mo. Probenecid was initially dissolved as 100 mg/ml
1N NaOH. This solution was further diluted 1:1,000
with saline solution and was used throughout the
experiment. Nigericin was initially dissolved in ethanol
to yield a stock solution of 10-3 M, which was further
diluted in 1 1 high [K'] solution to obtain a final
concentration of 10-` M. Indo-1 AM and BCECF AM
were obtained from Molecular Probes, Eugene, Ore.
Statistics
All data are presented as mean+SEM. Statistical
significance was determined using Student's t test.
Results
Isolated perfused rabbit hearts loaded with indo-1
produce calcium transients that are very similar to
those observed in isolated myocytes. Figure 1 shows
100F
IT'
E
E
0L
g 1.5
0
U-
'-6
0
,c
u-
i.o L
100_
0
0
ll1
L'
0
sec 1
FIGURE 1. Calcium-dependent fluorescence transients recorded from the epicardial surface of a rabbit heart after
infusion of indo-1 AM into the extravascular space. Binding of
calcium to indo-1 causes an increase in fluorescence at 400
nm (F400, middle tracing) and a decrease in fluorescence at
550 nm (F550, bottom tracing). The fluorescence ratio, F400/
F550 (top tracing), shows calcium transients with a rapid
upstroke and smooth decay. Movement artifact present in the
individual tracings is canceled in the ratio. Recordings differ
from those obtained by arterial loading insofar as the transients represent a larger percentage of the baseline fluorescence: 17% of end-diastolic fluorescence at 400 nm and 16%
of end-diastolic fluorescence at 550 nm.
beat-to-beat changes in indo-1 fluorescence recorded
from the epicardial surface of the left ventricle. For
this recording indo-1 AM was infused into the extravascular space within the field of the optical probe.
Resulting calcium transients had rapid upstrokes and
a slight distortion of the falling phase, which is due to
motion artifact and is canceled in the ratio. Compared with those in arterially loaded hearts, calcium
transients obtained by extravascular infusion of
indo-1 AM are twofold to threefold larger in proportion of total fluorescence. Thus, the transients shown
in Figure 1 represent 17% and 16%, respectively, of
the end-diastolic fluorescence at 400 and 550 nm.
Approximately one third of the diastolic fluorescence
in Figure 1 is not calcium dependent (autofluorescence plus nondeesterified indo-1 AM), as shown by
infusion of Mn'+ and ionomycin at the end of the
experiment. The phasic change in calcium-dependent
fluorescence of each beat is therefore about 25% of the
total indo-1 fluorescence signal. A change of this magnitude would be expected if 1) indo-1-free acid is
primarily confined to the cytoplasm of the cardiac
myocytes; 2) values of [Ca'+]i in systole and diastole are
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
CONTROL
m
BRADYKININ
80 [
cm
I
E
E
C
UC)
IL
0
11' 0.75
U0
'I'
0.50 -
.LU.ri.J.J
LfLJ&.L
F400
r~~~~~~r
r
........................................
1
0
sec
0
1
sec
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
FIGURE 2. Effect of bradykinin (10-5 M) on calcium transients and contraction in a heart loaded with indo-1AMby the
extravascular infusion technique. Bradykinin causes a parallel
elevation of the systolic and end-diastolic level of the calcium
transients, which is maximal about 30 seconds after the onset
of the effect. There is no change in the amplitude or diastolic
level of the contraction (top tracing). There is also no change
in the amplitude or shape of calcium transients recorded at
fluorescence 400 nm (F400, bottom tracing) or 550 nm (F550,
not shown). The middle tracing shows calcium transients
displayed as the F400/F550 ratio. Dotted lines in the right panel
show original end-diastolic signals.
consistent with those measured in isolated myocytes;
and 3) resulting fluorescence changes are consistent
with in vitro spectroscopy. For example, published
spectra of indo-120 show a 26% increase in fluorescence
at 400 nm as [Ca2+]i increases from 300 to 600 nM.
Methods to Eliminate the Nonmyocyte Components of
Indo-1 Fluorescence
Endothelial cells are a major source of nonmyocyte
fluorescence in hearts loaded with indo-1 AM by
aortic perfusion.16 This conclusion is based on the
ability of bradykinin to produce a large upward shift
in the baseline level of the calcium transients, even
though it has no effect on myocyte [Ca2+]1.16 To
determine whether extravascular infusion of indo-1
AM prevents nonmyocyte uptake, we administered
bradykinin (10-` M) after completion of the loading
procedure. In one of 12 hearts, there was no response
to bradykinin. In the other 11 hearts, a response was
present, which was smaller than that in hearts loaded
by aortic perfusion. As shown in Figure 2, the peak
response to bradykinin (right panel) is typically one
half the amplitude of the calcium transient. Bradykinin does not change the shape or net amplitude of
calcium transients at 400 or 550 nm (bottom tracing).
This suggests that effects of bradykinin are confined
to nonmyocytes.
Several strategies were used to eliminate the remaining bradykinin response, the most successful of
which was infusion of low concentrations of ionomy-
0
0
LL
CONTROL
75 [
1529
BRADYKININ
OL
075
0.55
1
_-
1
0
1
sec
sec
FIGURE 3. Absence of a bradykinin response in a heart that
has been pretreated with 100 nM ionomycin for 5 minutes.
Ionomycin elevates calcium transients in the fluorescence ratio
(F40,/F550) but has no effect on ventricular pressure or on the
net amplitude of calcium transients at 400 nm or 550 nm
(signals not shown). Addition of 10-5 M bradykinin for 30
seconds in the continued presence of ionomycin has no effect
on either the calcium transients or contraction.
0
cin (100 nM). Low concentrations of calcium ionophores cause endothelium-dependent relaxation in
vascular smooth muscle resulting from elevation of
endothelial cell [Ca2"]j, which leads to release of
endothelium-derived relaxing factor.22 Figure 3
shows calcium transients and ventricular contraction
in a heart that had been infused with 100 nM
ionomycin for 5 minutes. Infusion of ionomycin
caused a sustained upward displacement of the F4W/
F550 recording. Systolic and end-diastolic pressures
were unaffected, suggesting that the changes in fluorescence were confined to nonmyocardial cells.
These observations indicate that bradykinin and ionomycin act on the same population of cells and that
ionomycin can prevent the bradykinin response by
raising [Ca2+]i to a ceiling value.
Block of the bradykinin response is reversible if
ionomycin is removed from the perfusate within 5-10
minutes. However, prolonged infusion of ionomycin
(45 minutes) causes permanent unresponsiveness to
bradykinin that does not revert in ionomycin-free
saline. The basis for this phenomenon is uncertain,
but several factors point to loss of indo-1 from
nonmyocytes. In one heart (used subsequently for
Figure 4), ionomycin pretreatment caused reductions
of 29% and 38% in baseline fluorescence at 400 and
550 nm. This improved the signal-to-noise ratio and
caused a corresponding increase in the F4J/F550 phasic transient, but there was no increase in contraction
strength. This relative increase in the transient component of fluorescence does not occur if ionomycin is
omitted from the perfusate.
Calibration of [Ca2`]i Transients in
Bradykinin- Unresponsive Hearts
Fluorescence transients that are not affected by
bradykinin are considered myocyte specific and can
be calibrated. Figure 4 shows changes in 400 nm
fluorescence in a bradykinin-unresponsive heart infused with high calcium saline (100 mM CaCl2, zero
1530
Circulation Research Vol 69, No 6 December 1991
lonomycin, 100 mM Ca2+
Fmax
2 1000
+
-. 3001--=
m
100
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
20
0
60
40
sec
FIGURE 4. Calibration of [Ca2"]i transients in a heart rendered unresponsive to bradykinin. Superficial myocytes were loaded with
indo-1 AM by extravascular infusion, which was followed with 100 nM ionomycin for 45 minutes. This procedure caused a relative
increase in the transient component offluorescence (see text). Bradykinin was infused 15 minutes after washout of ionomycin, with
no effect on fluorescence. Calibration was performed 30 minutes later as follows: First, the heart was infused with a 100 mM CaC12
solution containing 1.5 ,M ionomycin. This solution caused a sustained contracture and increased the fluorescence signal at 400
nm to a maximum level, F,.. Then the zero calcium level (Fmin) was obtained by infusion of 100 mM MnCl2 plus ionomycin (see
text). The position of Fma,.. relative to the calcium transients and the resulting values of [Ca2+]i are similar to results obtained in single
cardiac myocytes.
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
sodium) containing ionomycin (1.5 ,uM). The highcalcium saline produces sustained contracture and
elevation of the fluorescence signal to Fm,,. As in
isolated myocytes,19 the displacement of Fm,, above
the diastolic baseline is approximately three times
the amplitude of the calcium transients. This result
contrasts with findings in intact heart loaded by
arterial infusion of indo-1 AM, for which the displacement of Fma above the baseline is 10 times
larger than the calcium transients (Figure 3 of Reference 1).
Calibration of indo-1 fluorescence in Figure 4 gives
a peak [Ca2"]i value of 642 nM and a diastolic [Ca2+]i
value of 333 nM. Similar results have been obtained
in eight hearts for which the mean systolic [Ca2+]i is
626+34 nM and mean diastolic [Ca2+]i is 313+ 19 nM.
Values are similar in one heart that did not require
infusion of 100 nM ionomycin to become unresponsive to bradykinin.
on [Ca2"]i Transients
The effects of ischemia were studied in hearts
loaded with indo-1 by the extravascular route that did
not respond to bradykinin. Figure 5 shows the effect
of 100 seconds of ischemia in such a heart where
ionomycin (100 nM) is present before and during the
recording. Ischemia produces an increase in the
systolic and end-diastolic levels of the [Ca2"]i transients together with broadening of the transients and
an increase in net amplitude (Figure 6). The increased amplitude of the calcium transients is discernible in single wavelength recordings (Figure 6,
bottom tracing), which confirms that this change
arises from the myocytes. Elevation of end-diastolic
[Ca2"]i is also myocyte specific since this response
cannot be mimicked by pharmacological stimulation
of nonmyocytes (i.e., by bradykinin).
Ninety seconds of ischemia increased peak systolic
[Ca2"]i from 609±29 to 1,341±159 nM (p=0.005)
Effects of Ischemia
ISCHEMIA
i REPERFUSE
75
E
o
.....
0.85
_
o
U-
U-
MEmb
. _
0.60L
---------------
--------
.~~~
0
0.5
1.5
1.0
2.0
2.5
min
FIGURE 5. Effect of ischemia on calcium transients in a bradykinin-unresponsive heart. Ionomycin (100 nM) is present in the
perfusate at the onset of ischemia and for 15 minutes before ischemia. Ionomycin prevented elevation of the calcium transients by
bradykinin, as shown in Figure 3. Ischemia is produced by cessation of coronary perfusion for 90 seconds while the heart is beingpaced
at 180 beats/min by an epicardial plunge electrode. Ischemia produces a prompt decline in systolic pressure along with progressive
elevation of the calcium transients (shown by the fluorescence ratio, F4JF550). The net amplitude of the calcium transients is also
increased. Reperfusion causes return of both signals to their original level but with a temporary increase in end-diastolic pressure that
may be related to the persisting elevation of [Ca2"]j. The recording is interrupted to obtain fast time base records.
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
CONTROL
CD
ISCHEMIA
E
0 1.2 r
tEw
75
LW
L
I
E
1531
z
0
W
O_
c)
W
a:
0 1.1 k
° 0.85
D
LL
U11
0
W
l,'
o
o
N
0.60
-J
0
z 1.0 L
F400
.._* \------
0
10
5
15
min
r
0
1
sec
0
1
sec
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
FIGURE 6. Same experiment as Figure 5 but at a faster time
base. Left panel: Recordings before ischemia. Right panel:
Recordings at 80 seconds of ischemia. Besides elevation of the
systolic and diastolic levels of the transients, the transients are
broadened during ischemia and their net amplitude increases.
These changes are seen in the fluorescence recording at 400
nm (F400, bottom tracing) as well as in the F40o/F550 ratio
(middle tracing). Changes in the net amplitude and shape of
the calcium transients must arise from the cardiac myocytes.
Elevation of the end-diastolic level of the transients is also
ascribed to the myocytes because changes in nonmyocyte
[Ca2"]i are not demonstrable in the presence of ionomycin
(same experiment as Figure 3).
while increasing end-diastolic [Ca2"]i from 315±25 to
553±52 nM (p=0.006) (n=5). The rise in diastolic
[Ca2+]i is unlikely to be an artifact of nonmyocyte
stores of indo-1. If such stores existed, they would
spuriously elevate Fmax (see Figure 4), so that systolic
[Ca21]i before ischemia would be implausibly low.
Another source of potential artifact besides ischemia
is a change in autofluorescence. Autofluorescence
changes have been measured in a previous study2 and
found to be significantly less than changes in indo-1
fluorescence during ischemia. After 90 seconds of
ischemia, the change in autofluorescence at 400 nm,
measured before loading with indo-1, was 37% of the
fluorescence change during a comparable trial after
loading. Taking this value as typical, it is then possible to determine the net change in indo-1 fluorescence at 400 nm for a group of ischemic hearts. From
this net fluorescence change, a corrected mean value
for diastolic and systolic [Ca2+]i can be obtained.
Resulting values for the five ischemic hearts mentioned above were 444 nM for end-diastolic [Ca2+]i
and 965 nM for peak systolic [Ca2+]j. These values
represent an increase of 41% and 58%, respectively,
over the preischemic values.
The effects of longer periods of ischemia were also
examined. Figure 7 shows the peak systolic value of
the calcium transient during 15 minutes of global
ischemia in five hearts. The initial increase in the
transient was maximal at about 2 minutes of ischemia
FIGURE 7. Mean systolic fluorescence ratio in five hearts
during 15 minutes ofglobal ischemia. The systolic fluorescence
ratio from each heart is normalized to the corresponding ratio
value at the onset of ischemia. Ischemia produces a prompt
increase in peak systolic [Ca2"]i by about 2 minutes of
ischemia. This is followed by a gradual decline in [Ca2+]i
between 2 and 5 minutes and then a progressive secondary rise
between 5 and 15 minutes.
and was followed by a slow decline to an intermediate level (2-5 minutes). Then there was a secondary
increase that reached or exceeded the 2-minute level
by 15 minutes of ischemia. These results indicate that
15 minutes of global ischemia produces two phases of
[Ca2+]i increase.
The sequence of changes in Figure 7 was similar in
hearts loaded with indo-1 AM by intravascular and
extravascular infusion. However, the changes shown
for the systolic fluorescence ratio are not representative of diastolic values or of time-averaged
[Ca21]j. The reason for this is that block of conduction or stimulus capture reduces the frequency of
calcium transients by one half after 3-5 minutes of
ischemia. Reduced frequency of calcium transients
causes [Ca2+]i to fall to a lower level at end diastole
and also increases the duration of diastole as a
fraction of the cardiac cycle. Thus, while time-averaged [Ca21]i has not been calculated from our data,
the fall in this value between 2 and 5 minutes of
ischemia must be more pronounced than the fall in
peak [Ca2]1i. It is therefore possible that the timeaveraged [Ca2+]i measured at 5 minutes of ischemia
would not be elevated compared with preischemic
measurements.9,10
Effects of Ischemia on pH,
Cytosolic acidification during ischemia may be a
causal factor responsible for the associated increase
in cytosolic calcium. To correlate the time course of
changes in intracellular calcium transients with
changes in pHi during ischemia, hearts were loaded
with the fluorescent pHi indicator BCECF. Loading
with BCECF had no significant effects on left ventricular developed pressure. Phasic fluorescence
changes in the 530 nm emission signal were generally
Circulation Research Vol 69, No 6 December 1991
1532
0C02
IC) 100[
v[
722r
C02 OFF
.
I 6.8
00
c
0
I 6.0[
7.0
-.
0
2
1
!8 6.4
3
min
FIGURE 8. Effects ofinfusion of an acidic, hypercarbic saline
solution (pH 6.72) on left ventricular pressure (top tracing) and
pHi in an isolated rabbit heart loaded with 2', 7'-bis(carboxyethyl)-5(6)-carboxy fluorescein. Perfusion with hypercarbic saline produced a rapid decrease in left ventricular pressure and
pHI,. The pHi remained stable after 1 minute of perfusion with
C02-rtich saline and retumed toward the normal value when
perfusion with normal saline was resumed.
6..L
0
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
10 minutes, and 6.11±0.04 after 15 minutes. Similar
values for the pHi decrease during ischemia have been
observed using other methods.24-26
Simultaneous Recordings of Calcium Transients
and pHi
Figure 1lA shows the effects of intracellular acidification with CO2 on the calcium transient (middle
tracing), left ventricular pressure (top tracing), and
pHi (bottom tracing) in a heart loaded with both
indo-1 and BCECF. Acidosis (pH. 6.88) caused a
rapid decrease in contraction strength along with a
decrease in pHi to 6.85. However, this level of
intracellular acidification did not produce any elevation of the calcium transient. The pHi of 6.85
achieved in Figure 1lA is similar to values obtained
after 2 minutes of ischemia (6.83 ±0.02 in Figure 10).
This observation suggests that the decrease in pH,
cannot explain the elevation of the calcium transients
that occurs during the first 2 minutes of ischemia. In
contrast, a reduction of pHi of the magnitude shown
would be sufficient to reduce contraction strength
during the initial minutes of ischemia. The experiment in Figure 1lA shows an excellent temporal
correlation between pHi and the reversible decrease
in contraction strength. Furthermore, the absence of
any change in the calcium transients suggests that the
ISCHEMIA
I1
6 I[
7
.....................
--_
L
0
1
2
3
4
min
15
FIGURE 10. Effects of prolonged ischemia on cytosolic pH
(mean+±SEM) in four isolated rabbit hearts. Cytosolic pH
values for each heart were determined by calibration with a
high [K'] solution and nigericin at the end of the experiment.
Ischemia caused a progressive decline in cytosolic pH which
reached a value of 6.11+±.04 after 15 minutes.
less than 3% of the total signal, which is substantially
smaller than the fluctuations related to pH. The
ability of BCECF to monitor pHi in the intact heart
could be confirmed by perfusion of the heart with
C02-rich saline. As shown in Figure 8, perfusion with
hypercarbic saline (pH 6.16) produced a rapid decrease in contractility, which occurred synchronously
with the increase in BCECF fluorescence. Calibration of the fluorescence recording shows that pH,
reached a steady-state value of 6.0 after about 1
minute of perfusion with hypercarbic saline and
returned to baseline value when perfusion with normal saline solution was resumed. Of particular importance is the fact that very small changes in pH,
(less than 0.1 unit) can substantially diminish contraction strength.
Figure 9 shows the effects of 15 minutes of global
ischemia on left ventricular pressure (top tracing) and
pHi (bottom tracing) in an isolated rabbit heart. The
pHi of the heart under normal conditions was
7.08±0.04 (n=10), which is in good agreement with
the pHi recorded in cardiac muscle by other methods.5
When the heart was made ischemic, both pHi and
contraction strength promptly declined with a similar
time course, suggesting that they are causally related.
Figure 10 shows the mean pHi values in four hearts
subjected to 15 minutes of global ischemia. The mean
control pHi (t=0) is 7.03+0.06. During global ischemia, there was a monotonic decrease in pHi, reaching
values of 6.83+0.02 after 2 minutes, 6.32±0.01 after
1
I
10
DURATION OF ISCHEMIA (min)
5
14
FIGURE 9. Effect of a 15-minute episode of
ischemia on left ventricular pressure (top tracing)
and pHI (bottom tracing). Ischemia produced a
prompt decrease in left ventricular pressure and a
progressive decline in pHI,. pHI, decreased to 5.9
after 15 minutes of ischemia. pHi values were
derived from a calibration curve obtained at the end
of the experiment. The calibration procedure in15 volved the use of a high [K'] solution containing
nigericin, which was adjusted to various pH values.
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
ranged from 6.55 to 6.76 (mean, 6.65 ±0.05). Thus the
pHi threshold for calcium transient elevation in these
hearts is bounded by the mean values of 6.65±0.05
and 6.83±0.03.
The existence of a pHi threshold for calcium
transient elevation is further shown by the delay in
calcium transient elevation that occurs when pHi
barely exceeds the threshold value. The steady-state
pHi achieved in Figure liB is 6.76, which is only
slightly more acidic than the pHi value of 6.83, which
did not elevate the calcium transients in Figure 1lA.
The greater acidification in Figures 11B and 12
produces elevation of both systolic and diastolic
[Ca2"]i along with broadening of the peak (Figure
12). However, there is a delay in the effects of pHi on
the calcium transients so that no elevation of the
transients occurs until 45 seconds after pHi has begun
to decline. Furthermore, changes in the calcium
transient occur slowly so that an additional 2.5 minutes is required for the full effect to become apparent
(Figure 12). These observations confirm that the
degree of acidification observed during the first 2
minutes of ischemia (Figure 10) does not account for
the concomitant elevation of the calcium transients.
Also, the disparity between the time course of
changes in pHi and indo-1 fluorescence confirms that
the latter does not reflect a direct effect of protons on
the fluorescence properties of indo-1.
TABLE 1. Threshold pHi Change Associated With Elevation of
Calcium Transients
No [Ca2"]i
elevation
[Ca2"]i elevation
Heart
No.
pHi
pHi
pH.
pH.,
1
6.40
6.55
6.78
6.79
2
6.60
6.70
6.80
6.89
3
6.31
6.60
6.51
6.77
4
6.76
6.75
6.90
6.88
6.51
6.65
6.75
6.83
0.10
0.05
0.08
0.03
For each heart, the right two columns show the lowest pH. and
pH, values that failed to produce elevation of the calcium transients. The left two columns show the highest pH,, and pH, values
that do produce elevation of the calcium transient.
Mean
SEM
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
decrease in contraction strength may be due to a
direct effect of pHi on the myofilaments.
The result in Figure 11A suggests that for any
given heart, there is a threshold change in pHi at
which elevation of the calcium transients first occurs.
This possibility has been explored systematically in
four hearts loaded with both indicators and then
infused with a series of four to six increasingly
hypercarbic solutions whose pH values differed by
0.15-0.40 pH units. For the four hearts, the lowest
pHi that failed to elevate the calcium transients
ranged from 6.77 to 6.89 (mean, 6.83+0.03; Table 1).
In the same hearts, the highest pHi value at which
elevation of the calcium transients was discernible
Discussion
The principal objective of these experiments was to
compare calibrated measurements of [Ca2+]i and pHi
FIGURE 11. Panel A: Effects of hypercarbic saline (pH 6.76, Pco2 130,
Po2 504) on left ventricular pressure
(top tracing), cytosolic calcium transients (F400, middle tracing), and intracellular pH (bottom tracing) in a heart
loaded simultaneously with indo-1 and
2', 7'-bis(carboxyethyl)-5(6)-carboxyfluorescein. Perfusion with this CO2
mixture decreased left ventricular pressure and pHI (to 6.85). However, the
change in pH, was insufficient to cause
an elevation of the calcium transients.
Panel B: Effects of C02-enriched saline
(pH 6.75) on the left ventricular pressure (top tracing), calcium transient
(F4e, middle tracing), and pH, (bottom
tracing). Perfusion with C02-enriched
solution decreased left ventricular pressure and pHi (to 6.76) and caused
delayed elevation of the calcium transient. Elevation of the calcium tran-
A
C02
[0
m100
_
C02 OFF
_F
E
F400
--
6.0
f7
--
--
w-
11
---rl.
1
-,
7.0-
30
0
60
90
sec
120
B
COO.j:
I
100
E
C02 OFF
[_
0
E
uu
Ae
-_.
sients
associated with a 34% inin the left ventricular developed
pressure. Resumption ofperfusion with
normal saline caused all recordings to
return to normal within 8 minutes.
crease
0
1
2
min
4
5
1533
6
8
was
Circulation Research Vol 69, No 6 December 1991
1534
the effect of bradykinin-induced fluorescence
changes. In rare cases extravascular loading is sufficient to eliminate the bradykinin response.
C02
CONTROL
cD 100[<
F400
6.0
7.0
0
0.5
sec
1.0
0
0.5
1.0
sec
FIGURE 12. Same experiment as Figure I1B but with faster
speed. With pHI 6.76 the systolic and diastolic level of
the calcium transient (F400) is elevated. The transient is also
broader and its decay ends more abruptly.
sweep
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
during global ischemia in the rabbit heart. A threepart strategy has been used. First, methods were
developed for calibrating the indicators indo-1 and
BCECF in the intact heart. Second, BCECF was used
to determine the time course of cytosolic acidification
during global ischemia. Third, [Ca2"]i and pHi were
monitored simultaneously in hearts acidified with hypercarbic saline to determine whether the degree of
acidification seen with ischemia could explain quantitatively the rise in [Ca'+1i. The usefulness of this
approach is evident from the temporal disparity in the
two measurements and from the existence of a threshold pHi that must be surpassed for [Ca'+]i to rise.
Calibration of Indo-1 in the Intact Heart
A major difficulty in interpreting the indo-1 fluorescence signals in the intact heart has been the lack
of a successful calibration technique. Without plausible measurements of systolic and diastolic [Ca'+]i it
is difficult to prove that calcium-dependent fluorescence changes are confined to the cytoplasm of the
cardiac myocytes. The two most obvious sources of
nonmyoplasmic indo-1 would be nonmyocardial cells,
such as endothelial cells,16 and organelles, such as
nuclei or mitochondria.27 Conversely, if calibration of
systolic and diastolic [Ca'+Ii gives values comparable
to those in single myocytes, then most if not all the
indo-1 in the heart must be localized within the
cytoplasm of cardiac myocytes.
Our present work suggests that calibration of
myocyte calcium transients is not possible in hearts
loaded with indo-1 AM by arterial perfusion. The
reason for this is significant uptake of indo-1 AM by
cells in the vessel wall that are sensitive to bradykinin. These cells include endothelial cells and possibly
other cells, such as smooth muscle cells. Nonmyocyte
fluorescence can be minimized by using an extravascular loading technique. Extravascular loading increases the magnitude of calcium transients relative
to total fluorescence and to F.., and also minimizes
Potential Sources of Artifact With Indo-1
Two other sources of artifact must be considered
in interpreting indo-1 results: a change in myoglobin
screening and mitochondrial uptake of indo-1 AM.
In the case of myoglobin screening, deoxygenation of
myoglobin would increase the perceived fluorescence
at 400 nm and decrease the fluorescence at 550 nm.
The importance of this effect would depend on light
penetration through the tissue. Several observations
indicate that changes in screening are not responsible
for effects of ischemia on indo-1 fluorescence. First,
ischemic trials performed in the absence of indo-1 do
not change the F40/ F550 ratio (Figure 4B of Reference 2). If myoglobin screening were important, the
effects of ischemia on the indo-1 fluorescence and
autofluorescence ratios should be similar. This point
can be made even more clearly if the heart is
perfused with cell-impermeant indo-1-free acid, so
that indo-1 fluorescence greatly exceeds autofluorescence. Ischemia cannot change the calcium saturation of indo-1 under these conditions, and we observe
no significant change in the F400 fluorescence level or
in the F400/F550 ratio. This confirms that ischemia
does not alter the screening of indo-1 fluorescence in
our experiments. Finally, the effect of ischemia on
520 nm emissions has been studied in hearts loaded
with indo-1 AM; 520 nm is an isosbestic wavelength
for the myoglobin screening effect (R. Balaban, personal communication, 1989). Ischemia produces a
decrease in fluorescence at 520 nm, similar to that
observed at 550 nm. This observation further shows
that the effect of ischemia on indo-1 fluorescence is
not an artifact of tissue screening.
A second source of potential artifact is trapping
of indo-1 AM in mitochondria. Although previous
studies indicate that very little of the Mn2'-quenchable fluorescence resides in mitochondria (Reference 2), it is still possible that an increase in
mitochondrial free calcium could contribute to elevation of the F400/ F550 ratio. Such an event would be
of physiological importance insofar as it would
represent a net transfer of calcium from other
organelles or from the extracellular space. Nonetheless, it is clear from the shape of the transients
that at least part of the elevation in baseline during
ischemia is due to a rise in cytosolic calcium.
Transients recorded during ischemia are falling
faster at the onset of the next stimulus than are the
preischemic transients (see Figure 6). It follows
that free calcium in the cytosolic compartment at
end diastole is farther away from the resting level
during ischemia than during control transients.
Incomplete reuptake of calcium eventually leads to
[Ca'+]i alternans, in which systolic and diastolic
calcium fluctuate from beat to beat.2 These observations confirm that elevation of the baseline of the
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
calcium transients during ischemia is at least partly
due to elevation of [Ca2+]i.
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
Time Course of the [Ca2+]i Increase During Ischemia
Our recordings indicate that intracellular calcium
increases very early during cardiac ischemia.1 2'9-11'28
We previously reported that brief periods of ischemia
produce elevation of calcium transients that reaches
a plateau after 90 seconds.1'2 In these studies, ischemia caused an increase in both systolic and diastolic
levels of the transients. The rise in [Ca2+]i occurred at
a time when contraction strength was falling. A
similar observation has been made in ischemic ferret
hearts loaded with aequorin by subepicardial injections.28 In that model, an increase in systolic and
diastolic [Ca2+]i begins within 2 minutes and coincides with decay of contraction strength. An increase
in time-averaged [Ca2+]i during the first 10 minutes of
ischemia has also been recorded using the calciumsensitive nuclear magnetic resonance probe 5F
BAPTA.9,10 An increase in [Ca2+]i within the first 5
minutes was not detected with this technique but
might have been missed owing to the length of time
required for signal acquisition and to the transient
fall in [Ca2+]i that is seen in Figure 7.
In the present study, we have extended our experiments with indo-1 to longer periods of ischemia. We
found that with prolonged ischemia, there is partial
decline of the calcium transients from the level
observed at 90 seconds, followed by a slow increase
between 5 and 15 minutes of ischemia. The observation of a progressive increase in [Ca2+]i between 5
and 15 minutes of ischemia is consistent with the SF
BAPTA measurements.
Cytosolic Acidification as a Potential Cause of the
[Ca2+J1 Increase During Ischemia
There is evidence that the amplitude of calcium
transients in cardiac muscle fibers can be increased
by cytosolic acidification achieved with either C02 or
lactic acid.2'67 Initial results with aequorin failed to
show an increase in diastolic [Ca2+]i during acidification,67 but results with ion-selective electrodes8 as
well as indo-12'29 show that acidification can increase
the diastolic or resting [Ca2+]i with approximately the
same time course as the increase in systolic [Ca2+]i.
The likelihood that sufficient acidification would
occur to elevate [Ca2 ]i during ischemia is well established from previous studies. Couper et a130 have
measured changes in extracellular pH in ischemic
rabbit ventricular myocardium and found that extracellular pH can change by 0.5 units during 90-second
ischemic trials at a pacing rate of 75 beats/min.
Changes in pH. are rate dependent and could occur
more rapidly in hearts beating at 180 beats/min. Case
et aP31 have measured accumulation of CO2 in ischemic canine myocardium and found that a doubling
of myocardial Pco2 can occur during the first 2
minutes of ischemia, with an eightfold increase during the first 8 minutes. CO2 accumulation is believed
to be more important than lactate accumulation in
1535
the ischemic heart because lactate accumulates too
slowly to account for the fall in pHi.32
The role of cytosolic acidity can be studied most
effectively by direct measurements of pHi. BCECF
fluorescence is a particularly useful method for making these measurements because recordings are obtained instantaneously and can be compared with
analogous recordings of [Ca21]i. Our measurements
show that ischemic myocardium becomes detectably
acidic within 15 seconds and that pHi declines monotonically by about 1.0 unit after 10-15 minutes of
ischemia. The time course and magnitude of the
decrease in pHi are comparable to what has been
recorded by 31P nuclear magnetic resonance spectroscopy in ischemic hearts containing no indicator.24-26
Correlation of Cystolic Acidification With Elevation of
Calcium Transients
The present experiments suggest that elevation of
the calcium transients does not occur unless pHi falls
below a threshold value of approximately 6.75. Furthermore, there is a delay between the fall in pHi and
the change in [Ca21]i. In cases for which the degree of
acidification barely exceeds the threshold value,
[Ca2"]i continues to rise for several minutes after pHi
has reached a steady value (Figure 12). These observations are in agreement with in vitro measurements
in which pH was found to have no direct effect on the
fluorescence of indo-1 over the range of pH values
studied in our experiments.2
There are three ways in which cytosolic acidification could produce an increase in [Ca21]i. First, it has
been suggested that acidosis enhances the release of
calcium from the sarcoplasmic reticulum.7 Alternately, acidosis may cause net uptake of calcium from
the extracellular space. According to this scheme,
intracellular acidosis would cause an increase in
intracellular sodium via the Na+-H+ exchange.33 The
resultant increase in [Nat]i would increase intracellular calcium via the Na+-Ca2' exchange.34 Either of
the above mechanisms would be consistent with the
time delay between cytosolic acidification and elevation of the calcium transients. A third mechanism
would be a simple exchange of calcium for protons at
cytosolic binding sites.35 This mechanism seems less
likely, since a simple exchange of ions at binding sites
ought to occur rapidly and should not be associated
with a time delay.
A major conclusion from the simultaneous indo-1
and BCECF recordings is that cytosolic acidification
during the first 2 minutes of ischemia is not sufficient
to explain the concomitant elevation of the calcium
transients. The mean value of pHi observed at 2
minutes of ischemia in Figure 10 is 6.85, which is
above the range of values that is associated with
elevation of calcium transients (see Table 1). Furthermore, the peak values of the calcium transients
during 15 minutes of ischemia (Figure 7) have a
triphasic time course, which differs from the monotonic decrease in pHi during ischemia (Figure 10).
This result would not be expected if the early eleva-
1536
Circulation Research Vol 69, No 6 December 1991
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
tion of the calcium transients were a simple consequence of cytosolic acidity.
Based on the above results, some other factor must
be responsible for producing the increase in [Ca2"]
during the first 2 minutes of ischemia. One substance
that might mediate this increase is endothelin,36
which is also released by hypoxia and is capable of
increasing [Ca2+]i in various cell types. Endothelin
can be detected in the serum of patients with recent
myocardial infarction37 and has been shown to increase the calcium current and [Ca2+]i in cardiac
myocytes.38
The temporary decrease in [Ca21]i after the first 2
minutes of ischemia (Figure 7) could be a consequence of conduction block or could represent "desensitization" of the stimulus producing the early
increase. The temporary decrease in [Ca2]]ireaches
its nadir at 4-5 minutes of ischemia, which is about
the time when pHi crosses the threshold for elevation
of calcium transients (Figure 6, Table 1). The secondary increase in [Ca2+]i could therefore be a consequence of progressive acidification.
In our study, the disparity noted by Jacobus et al is
less apparent. In Figure 11, for example, the acute
change in pHi associated with a 50% reduction in
developed pressure is 0.1 unit. The mean value obtained in four hearts exposed to hypercarbic saline is
0.14±0.01 unit. This reduction in pHi is similar to what
is seen in the first 2 minutes of ischemia. Thus, for a
given pHi value, ischemia and acidosis produce similar
impairment of contraction. In spite of this similarity,
we cannot conclude that acidosis is the only factor
producing contractile failure during ischemia. Analysis is complicated by the elevation of the calcium
transients during ischemia, which is greater, at least
initially, than that produced by acidosis alone.
Potential Role of pH, in Ischemic Contractile Failure
It is well recognized that ischemia produces a
decrease in contractility. The fact that there is a
decrease in contractility at a time when [Ca2+]i is
elevated suggests that the ability of the myofilaments
to produce tension is inhibited during ischemia (see
Reference 3 for review).
One of the factors that could inhibit the response
of the myofilaments to calcium is intracellular acidification. Katz and Hecht39 proposed that intracellular
acidification is the cause for decreased tension development during ischemia. Using skinned cardiac cells,
Fabiato and Fabiato4 showed that acidosis alters
myofilament sensitivity to calcium. Part of this inhibitory effect of acidosis results from competition between hydrogen ions and calcium for the calcium
binding sites on troponin C.40 Our experiments support this proposal by showing an excellent temporal
correlation between changes in contraction and
changes in pHi. In addition, very small changes in pHi
(0.1 unit or less) are found to have substantial effects
on contraction (see Figure lIA). Thus, the small
changes in pHi that occur during the first 2 minutes of
ischemia, while not sufficient to affect the calcium
transients, could still inhibit contraction.
The role of cytosolic acidity in ischemic contractile
failure has been investigated by Jacobus et a125 using
31P nuclear magnetic resonance. In their study, left
ventricular developed pressure was correlated with
pHi during acidification of rabbit hearts with CO2 to
determine the magnitude of the change in pHi associated with a 50% decline in developed pressure. The
mean change in pHi that produced this effect was
0.22 units. In contrast, when prolonged periods of
subtotal ischemia were produced, the steady-state
change in pHi associated with a reduction in developed pressure of 50% was 0.09 units.
2. Lee HC, Mohabir R, Smith N, Franz MR, Clusin WT: Effect
of ischemia on calcium-dependent fluorescence transients in
rabbit hearts containing indo 1: Correlation with monophasic
action potentials and contradiction. Circulation 1988;78:
1047-1059
3. Allen DG, Orchard CH: Myocardial contractile function during ischemia and hypoxia. Circ Res 1987;60:153-163
4. Fabiato A, Fabiato F: Effects of pH on the myofilaments and
the sarcoplasmic reticulum of skinned cells from cardiac and
skeletal muscle. J Physiol (Lond) 1978;276:233-253
5. Poole-Wilson PA: Measurement of myocardial intracellular
pH in pathological states. JMol Cell Cardiol 1978;10:511-526
6. Orchard CH: The role of the sarcoplasmic reticulum in the
response of ferret and rat heart muscle to acidosis. J Physiol
(Lond) 1987;384:431-449
7. Orchard CH, Houser SR, Kort AA, Bahinski A, Capogrossi
MC, Lakatta EG: Acidosis facilitates spontaneous sarcoplasmic reticulum Ca2' release in rat myocardium. J Gen Physiol
1987;90:145-165
8. Bers DM, Ellis D: Intracellular calcium and sodium activity in
sheep heart Purkinje fibres: Effect of changes of external
sodium and intracellular pH. Pflugers Arch 1982;393:171-178
9. Marban E, Kitakaze M, Kusuoka H, Porterfield JK, Yue DT,
Chacko VR: Intracellular free calcium concentration measured with 19F NMR spectroscopy in intact ferret hearts. Proc
Natl Acad Sci U SA 1987;84:6005-6009
10. Steenbergen C, Murphy E, Levy L, London RE: Elevation in
cytosolic free calcium concentration early in myocardial isch-
Acknowledgments
We would like to thank Dagmar Truckses, Gordon
Grant, and David Profitt for assistance in conducting
the experiments.
References
1. Lee HC, Smith N, Mohabir R, Clusin WT: Cytosolic calcium
transients from the beating mammalian heart. Proc NatlAcad
Sci USA 1987;84:7793-7797
emia in perfused rat heart. Circ Res 1987;60:700-707
11. Allen DG, Lee JA, Smith GL: The consequences of simulated
ischaemia on intracellular Ca2' and tension in isolated ferret
ventricular muscle. J Physiol (Lond) 1989;410:297-323
12. Paradiso AM, Negulescu PA, Machen TE: Na'/HW and Cl-/
HCO3- exchange in gastric glands. Am J Physiol 1986;250:
G524-G534
13. Paradiso AM, Tsien RY, Demarest JR, Machen TE: Na+-H+
and Cl--HCO3- exchange in rabbit oxyntic cells using fluores-
cence microscopy. Am J Physiol 1987;253:C30-C36
14. Rink TJ, Tsien RY, Pozzan T: Cytoplasmic pH and free Mg2`
in lymphocytes. J Cell Biol 1982;95:189-196
15. Di Virgilio F, Steinberg JA, Silverstein SC: Inhibition of fura-2
sequestration and secretion with organic anion transport
blockers. Cell Calcium 1990;11:57-62
16. Lorell BH, Apstein CS, Cunningham MJ, Schoen FJ, Weinberg EO, Peeters GA, Barry WH: Contribution of endothelial
Mohabir et al Ischemia and Hypercarbic Acidosis in Rabbit Hearts
17.
18.
19.
20.
21.
22.
23.
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
24.
25.
26.
27.
28.
cells to calcium-dependent fluorescence transients in rabbit
hearts loaded with indo-1. Circ Res 1990;67:415-425
Almers W, Neher E: The Ca signal from fura-2 loaded mast
cells depends strongly on the method of dye loading. FEBS
Lett 1985;192:13-18
Josephson RA, Silverman HS, Lakatta EG, Stern MD, Zweier
JL: Study of the mechanisms of hydrogen peroxide and
hydroxyl free radical induced cellular injury and calcium
overload in cardiac myocytes. J Biol Chem (in press)
Lee HC, Clusin WT: Cytosolic calcium staircase in culturedmyocardial cells. Circ Res 1987;61:934-939
Hesketh TR, Smith GA, Moore JP, Metcalfe JC: Free cytoplasmic calcium concentration and mitogenic stimulation of
lymphocytes. J Biol Chem 1983;258:4878-4882
Peeters GA, Hlady V, Bridge JHB, Barry WH: Simultaneous
measurement of calcium transients and motion in cultured
heart cells. Am J Physiol 1987;253:H1400-H1408
Grynkiewicz G, Poenie M, Tsien RY: A new generation of
Ca2' indicators with greatly improved fluorescence properties.
JBiol Chem 1985;260:3440-3450
Singer HA, Peach MJ: Calcium- and endothelial-mediated
vascular smooth muscle relaxation in rabbit aorta. Hypertension 1982;4(suppl II):II-19-II-25
Bailey IA, Williams SR, Radda GK, Gadian DG: Activity of
phosphorylase in total global ischemia in the rat heart. Biochem J 1981;196:171-178
Jacobus WE, Pores IH, Lucas SK, Kallman CH, Weisfeldt
ML, Flaherty JT: The role of intracellular pH in the control of
normal and ischemic myocardial contractility: A 31P nuclear
magnetic resonance and mass spectrometry study, in Nuccitelli
R, Deamer DW (eds): Intracellular pH: Its Measurement,
Regulation and Utilization in Cellular Function. New York, Alan
R Liss, Inc, 1982, pp 537-565
Ochi S, Ogawa Y, Imai H, Nakazawa M, Imai S: Effects of
beta blockers on the fall in pH in the early phase of ischemia
in the isolated perfused rat heart: A nuclear magnetic resonance study. J Cardiovasc Pharmacol 1988;11:326-331
Spurgeon HA, Stern MD, Baartz G, Raffaeli S, Hansford RG,
Talo A, Lakatta EG, Capogrossi MC: Simultaneous measurement of Ca2 , contraction, and potential in cardiac myocytes.
Am J Physiol 1990;258:H574-H586
Kihara Y, Grossman W, Morgan JP: Direct measurement of
changes in intracellular calcium transients during hypoxia,
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
1537
ischemia, and reperfusion of the intact mammalian heart. Circ
Res 1989;65:1029-1044
Kohomoto 0, Barry WH: The effects of intracellular acidosis
on [Ca+]i, contraction, and 45Ca influx in cultured chick
embryo ventricular myocytes (abstract). Circulation 1988;
78(suppl II):II-152
Couper GS, Weiss J, Hiltbrand B, Shine KI: Extracellular pH
and tension during ischemia in the isolated rabbit ventricle.
Am J Physiol 1984;247:H916-H927
Case RB, Alexis F, Castellana FS: Rate of rise of myocardial
pCO2 during early myocardial ischemia in the dog. Circ Res
1979;45:324-330
Ichihara K, Haga N, Abiko Y: Is ischemia-induced pH
decrease of dog myocardium respiratory or metabolic acidosis? Am J Physiol 1984;246:H652-H657
Piwinca-Worms D, Jacob R, Horres CR, Lieberman M: Na/H
exchange in cultured chick heart cells pHi regulation. J Gen
Physiol 1985;85:43-64
Mullins LJ: The generation of electric currents in cardiac
fibers by Na/Ca exchange. Am J Physiol 1979;237:C103-C110
Vaughan-Jones RD, Lederer WJ, Eisner DA: Ca2' ions can
affect intracellular pH in mammalian cardiac muscle. Nature
1983;301:522-524
Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi
M, Mitsui Y, Yazaki Y, Goto K, Masaki T: A novel potent
vasoconstrictor peptide produced by vascular endothelial cells.
Nature 1988;332:411-415
Miyauchi T, Yanagisawa M, Tomizawa T, Sugishita Y, Suzuki
N, Fujino M, Ajisaka R, Goto K, Masaki T: Increased plasma
concentrations of endothelin-1 and big endothelin-1 in acute
myocardial infarction. Lancet 1989;2:53-54
Lauer MR, Gunn MD, Clusin W: Effect of endothelin on
cytosolic calcium and membrane current in single ventricular
myocytes (abstract). Circulation 1989;80(suppl II):II-194
Katz AM, Hecht HH: The early "pump" failure of the
ischemic heart. Am J Med 1969;47:497-502
Blanchard EM, Solaro RJ: Inhibition of the activation and
troponin calcium binding of dog cardiac myofibrils by acidic
pH. Circ Res 1984;55:382-391
KEY WORDS * ischemia * calcium transient * intracellular pH
indo-1 * BCECF
a
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
Effects of ischemia and hypercarbic acidosis on myocyte calcium transients, contraction,
and pHi in perfused rabbit hearts.
R Mohabir, H C Lee, R W Kurz and W T Clusin
Circ Res. 1991;69:1525-1537
doi: 10.1161/01.RES.69.6.1525
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/69/6/1525
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/