Vol. 275, Issue 3, H917-H929, September 1998
Functional and metabolic effects of extracellular magnesium in
normoxic and ischemic myocardium
John P.
Headrick,
James C.
McKirdy, and
Roger J.
Willis
Rotary Centre for Cardiovascular Research, School of Health
Science, Griffith University Gold Coast Campus, Southport,
Queensland 4217, Australia
 |
ABSTRACT |
Metabolic and
functional responses to extracellular
Mg2+ concentration
([Mg2+]o)
were studied in perfused rat heart. Elevations of
[Mg2+]o
from 1.2 to 2.4, 5.0, and 8.0 mM dose dependently reduced contractile function and myocardial oxygen consumption
(M
O2) up to 80%. Intracellular Mg2+ concentration
([Mg2+]i)
remained stable (0.45-0.50 mM) during perfusion with 1.2-5.0 mM
[Mg2+]o
but increased to 0.81 ± 0.14 mM with 8.0 mM
[Mg2+]o.
Myocardial ATP was unaffected by
[Mg2+]o,
phosphocreatine (PCr) increased up to 25%, and
Pi declined by up to 50%. Free
energy of ATP hydrolysis
(
GATP)
increased from 
60 to 64 kJ/mol. Adenosine efflux declined
in parallel with changes in
MO2 and
[AMP]. At comparable workload and
MO2, the effects of
[Mg2+]o
on cytosolic free energy were mimicked by reduced extracellular Ca2+ concentration
([Ca2+]o)
or Ca2+ antagonism with verapamil.
Moreover, functional and energetic effects of
[Mg2+]o
were reversed by elevated
[Ca2+]o.
Despite similar reductions in preischemic function and
MO2, metabolic and
functional recovery from 30 min of global ischemia was enhanced
in hearts treated with 8.0 mM
[Mg2+]o
vs. 2.0 µM verapamil. It is concluded that
1) 1.2-8.0 mM
[Mg2+]o
improves myocardial cytosolic free energy indirectly by reducing metabolic rate and Ca2+ entry;
2)
[Mg2+]i
does not respond rapidly to elevations in
[Mg2+]o
from 1.2 to 5.0 mM and is uninvolved in acute functional and metabolic
responses to
[Mg2+]o;
3) adenosine formation in rat heart
is indirectly reduced during elevated
[Mg2+]o;
and 4) 8.0 mM
[Mg2+]o
provides superior protection during ischemia-reperfusion
compared with functionally equipotent
Ca2+ channel blockade.
adenosine; calcium; cardioplegia; free energy of adenosine
5'-triphosphate hydrolysis; phosphorus-31 nuclear magnetic
resonance spectroscopy; rat hearts
| |
INTRODUCTION |
ELEVATED EXTRACELLULAR
Mg2+
([Mg2+]o)
is employed in cardioplegic solutions, where it improves myocardial
recovery (29). Mg2+ treatment is
used in a variety of cardiovascular disorders including myocardial
infarction and arrhythmias (3, 44, 47). Moreover, a number of
cardiovascular disorders have been associated with low extracellular or
cellular [Mg2+]
including hypertension, myocardial infarction, arrhythmias, and
congestive heart failure (3, 44). Given the variety of clinical uses of
elevated Mg2+, and the implication
of Mg2+ imbalances in
cardiovascular pathophysiologies, it is relevant to develop our
understanding of the mechanisms of action of
[Mg2+]o.
There remains some controversy regarding the functional effects of
[Mg2+]o
in myocardial tissue, and few studies have examined the metabolic or
energetic effects of
[Mg2+]o
in mammalian heart.
In most studies
[Mg2+]o
is shown to exert negative inotropic and chronotropic effects in
cardiac tissue (8, 12, 17, 22, 31, 48, 49), although there is
paradoxical evidence for positive inotropic effects of moderate
elevations in
[Mg2+]o
(6, 32). Additionally, there are conflicting data regarding underlying
mechanisms of action of
[Mg2+]o.
Some studies support extracellular mechanisms of action (e.g., inhibition of sarcolemmal Ca2+
fluxes) (1, 32, 49), whereas others support modification of
intracellular processes [e.g., sarcoplasmic reticulum (SR) Ca2+ handling] by associated
elevations in intracellular Mg2+
([Mg2+]i)
(6, 51). Similarly, although it has been reported that a modest
elevation in
[Mg2+]o
improves myocardial energy state (6), the underlying mechanisms are
undefined. A Mg2+-dependent
improvement of cytosolic free energy state could be particularly
beneficial in settings of heart failure, ischemia, and
cardioplegia.
Generally speaking, elevations in
[Mg2+]o
may alter myocardial contractile function in two ways:
1) modulation of sarcolemmal Ca2+ fluxes by external
Mg2+, and/or
2) modification of myofibrillar and
SR function by associated elevations in
[Mg2+]i.
Similarly, elevations in
[Mg2+]o
may alter cytosolic free energy state by inhibition of sarcolemmal Ca2+ fluxes (reducing myocardial
work and energy demand) and/or improvement of substrate and
energy metabolism via elevations in
[Mg2+]i.
An additional indirect mechanism involves stimulation of adenosine formation (38) with resultant activation of cardiovascular adenosine receptors.
In an attempt to resolve some of the many remaining issues regarding
functional and metabolic effects of extracellular
Mg2+ in mammalian myocardium we
have 1) characterized functional and metabolic effects of graded elevations in
[Mg2+]o
in perfused rat heart; 2) compared
response to elevated
[Mg2+]o
with those to reduced extracellular
Ca2+ concentration
([Ca2+]o)
or to Ca2+ channel blockade;
3) examined the role of associated
changes in
[Mg2+]i
in metabolic and functional responses;
4) characterized effects of
[Mg2+]o
on myocardial adenosine formation; and
5) examined the relative metabolic
and functional benefit of elevated
[Mg2+]o
vs. Ca2+ antagonism during
ischemia-reperfusion.
| |
METHODS |
Isolated perfused rat hearts.
Hearts were isolated from mature male Wistar rats as described
previously (24-26, 28). Rats were anesthetized with 50 mg/kg pentobarbital sodium, a thoracotomy was performed, and the hearts were
rapidly excised into ice-cold perfusion fluid. The aorta was
immediately cannulated and hearts were perfused in a retrograde manner
at a constant hydrostatic pressure of 100 mmHg with modified Krebs
bicarbonate buffer containing (in mM) 120 NaCl, 25 NaHCO3, 4.7 KCl, 1.25 CaCl2, 1.2 MgCl2, 15 glucose, and 0.05 EDTA. The perfusate was equilibrated with 95%
O2-5%
CO2 at 36.5°C, giving a pH of
7.4. The pulmonary artery was cannulated for collection of venous
effluent for determination of effluent
PO2 and analysis of venous purine
metabolite levels using HPLC (27, 40).
MO2 was calculated as the
difference between perfusate and venous
PO2 (µl
O2/ml) multiplied by coronary flow (ml · min
1 · g1)
(27, 40). The left ventricle was vented with a polyethylene apical
drain and a fluid-filled latex ventricular balloon inserted via the
mitral valve. The balloon was inflated to yield a left ventricular
end-diastolic pressure of 2-4 mmHg. The balloon was attached to a
P23XL pressure transducer (Viggo-Spectramed, Oxnard, CA) by
fluid-filled polyethylene tubing, and ventricular function was
continuously monitored on a two-channel MacLab data acquisition unit
(ADInstruments, Castle Hill, Australia). The rate-pressure product
(heart rate × left ventricular developed pressure) was calculated
as an index of ventricular contractile function. After instrumentation
hearts were stabilized for a period of 40 min.
31P nuclear magnetic resonance
spectroscopic techniques and determination of phosphate metabolite
levels.
To assess changes in metabolic state and
[Mg2+]i,
we perfused rat hearts in the bore of a nuclear magnetic resonance
(NMR) magnet in a manner similar to that described previously (28).
Consecutive 31P spectra were
acquired at 121.47 MHz using a 90° radiofrequency pulse with an interpulse delay of 1.7 s. Spectral width was 4 kHz, and
a total of 4K data points were obtained. Individual
spectra consisted of 256 signal-averaged free-induction decays (FID)
acquired over consecutive 8-min periods or 160 FID acquired over
consecutive 5-min periods in the case of ischemia-reperfusion
experiments. FID were multiplied by a 25-Hz line-broadening factor to
improve spectral signal to noise.
31P spectral intensities were
determined by integration and corrected for partial relaxation using
spin-lattice relaxation times (T1) of 0.94, 2.17, and 1.75 s for 
-ATP, phosphocreatine (PCr), and
Pi (28), respectively. To convert
spectral intensities to metabolite concentrations, we determined
cytosolic ATP and total creatine (Cr) levels in a group of
freeze-clamped control hearts (n = 6).
Powdered frozen tissue was extracted with 0.6 M perchloric acid, and
neutralized samples were analyzed for ATP, PCr, and Cr as outlined
previously (28). Concentrations were calculated on the basis of a
measured intracellular volume of 0.474 ml/g wet wt (see below). The
intracellular ATP concentration determined in this way was assigned to
the saturation-corrected -ATP intensity from baseline spectra, and
all saturation-corrected intensities were normalized against this ATP
concentration-to-intensity ratio (28).
Extracellular and cytosolic volumes were determined for control
Langendorff hearts perfused at 100 mmHg pressure with 1.2 mM
[Mg2+]o
and
[Ca2+]o
(n = 7). After 40 min of stabilization
hearts were supplied with perfusate containing 0.1 µCi/ml of
[14C]mannitol (ICN
Biochemicals, Costa Mesa, CA) for a period of 5 min. Perfusate from
each heart was sampled for determination of
14C activity. After the 5-min
period of perfusion with
[14C]mannitol-enriched
perfusate, the hearts were removed and gently blotted, and ~300 mg of
left ventricular myocardium were frozen in liquid
N2. Approximately 300 mg were also
weighed and then oven-dried at 80°C to a stable weight to measure
total water content. The frozen ventricular tissue samples were
subsequently pulverized, and powdered tissue was extracted with 0.6 M
perchloric acid as described previously (28). Neutralized extracts were
analyzed for
[14C]mannitol in a
scintillation counter. The extracellular space was calculated from the
ratio of the myocardial
[14C]mannitol activity
per gram wet wt divided by the
[14C]mannitol activity
per milliliter of perfusate. Intracellular volume was then calculated
by subtraction of the extracellular [14C]mannitol space
from total water content. In this way, total tissue water was
calculated to be 0.835 ± 0.004 ml/g, extracellular water content
was 0.361 ± 0.010 ml/g, and intracellular water was calculated to
be 0.474 ± 0.011 ml/g in control hearts.
Calculation of myocardial intracellular pH and
[Mg2+]i.
Intracellular pH (pHi) was
calculated from the chemical shift of
Pi relative to PCr
(
Pi)
Myocardial
[Mg2+]i
was calculated from the shift of the 
-P and -P resonances of ATP
in 31P spectra (21, 26, 28, 40)
where
[ATP]free
is the total unchelated ATP species,
[ATP]total is the sum
of the total chelated and unchelated cytosolic ATP species, 
is the chemical shift
difference [in parts per million (ppm)] between the -P
and -P resonances of ATP in the heart, ATP is the chemical shift
difference between the -P and -P resonances in a solution of
unchelated ATP, and Mg-ATP is
the chemical shift difference in a solution of ATP saturated with
Mg2+. We recently estimated the
dissociation constant
(KMg-ATPD) to be
0.058 mM at an ionic strength = 0.25, pH = 7.20, and 37°C (56).
This KMg-ATPD value was
adjusted to observed pHi in
perfused hearts (35)
Calculation of free cytosolic [ADP], [AMP],
[ATP]/[ADP][Pi],
and free energy of ATP hydrolysis.
Free cytosolic [ADP], [AMP],
[ATP]/[ADP][Pi],
and free energy of ATP hydrolysis
(GATP)
cannot be directly measured and are estimated from creatine kinase,
adenylate kinase, and ATPase equilibria, as described by us previously
(28, 40). Cytosolic [ADP] was calculated as
where
[Cr] was determined by subtraction of PCr from the total
chemically measured creatine pool (
PCr + Cr), and
K'ck is the observed
equilibrium constant for creatine kinase, adjusted for measured
pHi and
[Mg2+]i
as outlined in detail by Lawson and Veech (36) but using more recent
dissociation constants listed in Teague and Dobson (50). The
phosphorylation ratio
([ATP]/[ADP][Pi])
was then calculated from the free cytosolic [ADP] and the
31P NMR-determined ratio,
[ATP]/[Pi].
Free cytosolic [AMP] was calculated from the adenylate
kinase equilibrium
where
the observed equilibrium constant for adenylate kinase
(K'ak) was also
adjusted for measured pHi and [Mg2+]i.
The free energy of ATP hydrolysis
(GATP) was
calculated as
where
R is the ideal gas constant (8.31 J · K1 · mol1),
T is temperature in kelvin (310 K),
and the standard free energy of ATP hydrolysis
(GoATP) is calculated as
RTln K'ATP,
where K'ATP is the
equilibrium constant for ATPase adjusted for
pHi and
[Mg2+]i
(28).
Myocardial ion contents.
Hearts were analyzed for myocardial
Na+,
K+, and
Mg2+ contents at the end of 12-min
perfusion with either 1.2, 2.4, 5.0, or 8.0 mM
[Mg2+]o
after 40-min stabilization with normal perfusate (1.2 mM
[Mg2+]o).
At the end of the experimental period hearts were immersed in ice-cold
20 mM Tris · HCl buffer (pH 7.4) containing 300 mM sucrose and were perfused for 3 min with 20 ml of this solution to wash
out extracellular ions. All hearts were removed and blotted, and ~300
mg of left ventricular myocardium were weighed before drying at
80°C to a constant dry weight. Dried ventricular samples were
weighed and extracted in 12 M HNO3
for 24 h at 80°C. Extracted samples were diluted with HPLC-grade
water and analyzed for Na+,
K+, and
Mg2+ contents by atomic absorbance
spectroscopy. Perfusion with ion-free solution washes >90% of ions
from the extracellular space. To assess the degree of vascular washout
with this protocol and thereby correct for residual extracellular ion
contamination, hearts perfused with 1.2 mM
[Mg2+]o
were supplied with perfusate containing 0.1 µCi/ml of
[14C]mannitol (ICN
Biochemicals) for a period of 5 min before perfusion with the
cation-free solution. An effluent sample was acquired for measurement
of [14C]mannitol
activity, and the hearts were then flushed with ion-free solution as
described above.
[14C]mannitol content
was measured in these hearts (as described above for determination of
total extracellular volume) and was used to calculate perfusate ion
contamination by dividing myocardial [14C]mannitol/g by
perfusate
[14C]mannitol content.
Residual perfusate content determined in this way was 0.010 ± 0.004 ml/g wet wt (or 0.061 ± 0.024 ml/g dry wt). Thus the washing
procedure flushed ~97% of
[14C]mannitol from the
extracellular compartment. Assuming a comparable washout of
extracellular ions, we calculated intracellular
Na+,
K+, and
Mg2+ contents as
|
|
with
ion contents in moles per milliliter.
The free energies of transsarcolemmal ion gradients for
Na+
(G[Na+]o/i)
and Mg2+
(G[Mg2+]o/i)
were estimated. The Nernst potentials for distribution of
K+,
Na+, and
Mg2+ across the sarcolemma
(E[ion]o/i)
were first calculated as
where
T is 310 K,
n is the sign and valence of the ion,
F is the Faraday constant (96.45 J · V1 · mol1),
and [ion]o and
[ion]i are
extracellular and intracellular ion concentrations, respectively.
Perfusate Na+,
K+, and
Mg2+ concentrations (extracellular
concentrations) were measured using atomic absorbance spectroscopy.
Total tissue contents for Na+,
K+, and
Mg2+ (mol/g dry wt) were converted
to molar concentrations based on an intracellular volume of 0.474 ml/g
wet wt and measured total myocardial water contents.
E[ion]o/i
values were then used to calculate
G values for transsarcolemmal ion movement
(G[ion]o/i)
The
above method of estimating intracellular
Na+ and
K+ from total tissue levels minus
extracellular contamination has been employed in previous studies (see
Ref. 39 and references therein), and we are unaware of evidence
supporting significant binding or compartmentation of intracellular
Na+ and
K+. Indeed, microelectrode studies
(e.g., Ref. 9) yield estimates of
[Na+]i
comparable to those obtained by analysis of total tissue
Na+ here (see
RESULTS) and by others (39). Additionally, the Nernst potential for K+ calculated from
the total tissue data (~85 mV) agrees well with direct measures of
membrane potential in myocytes (~83 mV, see Ref. 39), which is
predicted assuming a near equilibrium between K+ distribution and the resting
membrane potential.
Measurement of myocardial purine efflux.
Venous effluent samples collected from
Mg2+-treated hearts were frozen at
80°C until analysis for adenosine and inosine by HPLC, as
described in detail previously (27, 40). Adenosine and inosine were
identified by retention times and were quantitated by comparison of
peak areas with those for known standards run daily with the effluent
samples. Data analysis was performed using the Waters Maxima software
package (Waters, Milford, MA).
Experimental protocol.
All hearts were stabilized for a 40-min period with control perfusate
containing 1.2 mM Mg2+. Basal
31P NMR spectra and functional
measurements were then acquired. Functional and metabolic properties of
hearts were studied during elevations in
[Mg2+]o,
reductions in
[Ca2+]o,
infusion of verapamil, and during ischemia-reperfusion in nonarrested and KCl-arrested hearts with or without elevated
[Mg2+]o
or verapamil. Additionally, a group of control hearts was perfused with
1.2 mM
KH2PO4
present in the perfusate to assess the impact of extracellular
phosphate on metabolic parameters and
[Mg2+]i.
To examine effects of elevated
[Mg2+]o,
we subjected stabilized hearts to stepped elevations in
[Mg2+]o
from 1.2 to 2.4, 5.0 and 8.0 mM for periods of 12 min at each concentration. The hearts were stabilized at each
[Mg2+]o
for 4 min before acquisition of
31P spectra and functional
measurements. Functional responses to [Mg2+]o
were rapid and stabilized within 1-2 min at all
[Mg2+]o.
Preliminary studies in which consecutive 4-min spectra were acquired
throughout the protocol revealed all metabolic changes were stable
after the initial 4-min period (data not shown).
To examine the concentration-dependent inotropic effects of
[Mg2+]o
without interference from chronotropic responses, a series of hearts
(n = 7) were stabilized for 30 min as
described above before being switched to constant-flow perfusion (at
16.5 ± 1.7 ml · min1 · g1
at a mean aortic pressure of 95 ± 7 mmHg) and were electrically paced at 4 Hz. After an additional 10-min stabilization period, baseline functional measurements (1.2 mM
[Mg2+]o)
were acquired and the hearts were then subjected to stepped elevations
in
[Mg2+]o
(2.4, 5.0 and 8.0 mM). The hearts were stabilized at each
[Mg2+]o
for 4 min before acquisition of stable functional measurements.
To examine the ability of alterations in
[Ca2+]o
to mimic effects of elevated
[Mg2+]o,
we stabilized a group of hearts
(n = 8) under control
conditions (1.2 mM
[Mg2+]o,
1.2 mM
[Ca2+]o)
before acquisition of baseline measurements. This was followed by
graded reductions in
[Ca2+]o
to 0.7 and 0.2 mM (0.75 and 0.25 mM
[Ca2+]o
in the presence of 0.05 mM EDTA) and an elevation to 2.5 mM (2.55 mM
[Ca2+]o
with 0.05 mM EDTA). Hearts were stabilized at each concentration for 4 min before acquisition of 31P
spectral and functional data.
Effects of Ca2+ antagonism were
examined in a third group of hearts (n = 8). After acquisition of baseline data, hearts were subjected to
stepped elevations in perfusate verapamil from 0 to 0.5 and 2.0 µM.
Hearts were stabilized for 4 min at each dose before acquisition of
31P spectral and functional data.
[Ca2+]o
reversal of the functional and metabolic effects of elevated
[Mg2+]o
was studied in a fourth series of experiments. Hearts
(n = 8) were stabilized under control
conditions (1.2 mM
[Mg2+]o,
1.2 mM
[Ca2+]o)
before acquisition of baseline measurements. The hearts were then
switched to perfusion with 5.0 mM
[Mg2+]o
and 2.5 mM
[Ca2+]o
and were paced at the basal rate (210 ± 10 beats/min). Pacing was
necessary as elevated
[Ca2+]o
reversed the inotropic effects of
[Mg2+]o
but not the chronotropic response (data not shown). Hearts were
stabilized with the
high-Mg2+/Ca2+
buffer for 4 min before acquisition of
31P spectral and functional data.
In the final series of experiments effects of elevated
[Mg2+]o
or Ca2+ channel blockade with
verapamil were studied in ischemic-reperfused hearts. After baseline
measurements were acquired, hearts were either untreated
(n = 7) or subjected to cardioplegic
arrest by perfusion with modified buffer containing 25 mM KCl alone
(n = 7), 25 mM KCl + 8.0 mM
[Mg2+]o
(n = 8), or 25 mM KCl + 2.0 µM
verapamil (n = 8). The NaCl concentration was reduced by 25 mM in all cardioplegic solutions. Cardioplegic perfusion was maintained for 6 min, after which all hearts
were subjected to 30 min of total global normothermic ischemia followed by 20 min of reperfusion with normal perfusion fluid (1.2 mM
[Mg2+]o,
4.7 mM KCl, 0 µM verapamil).
Data analysis.
Data reported are means ± SE. The data were analyzed using an ANOVA
followed by the Newman-Keuls post hoc test for individual comparisons.
In all tests significance was accepted at the 95% confidence
level (P < 0.05).
| |
RESULTS |
Functional and metabolic effects of elevated
[Mg2+]o.
Baseline functional parameters for Langendorff-perfused rat hearts
supplied with control perfusate (i.e., 1.2 mM
[Mg2+]o,
1.2 mM
[Ca2+]o,
0 µM verapamil) are shown in Table 1.
Elevations of
[Mg2+]o
dose dependently reduced peak systolic pressure and heart rate and
elevated end-diastolic pressure. The dose dependence for the negative
inotropic and chronotropic effects of
Mg2+ was comparable (Fig.
1). The concentration-dependent reduction in inotropic state in the absence of heart rate changes is shown in
Fig. 2. Inotropic sensitivity to
[Mg2+]o
in these hearts is similar to that observed in nonpaced hearts (Fig.
1). MO2 decreased in parallel
with contractile function (see Figs. 1 and 4). A linear correlation was
obtained between MO2 and the
rate-pressure product during elevations in
[Mg2+]o
[MO2 = 30.8 + 0.0075 · (rate-pressure product)] (Fig. 4). Coronary flow was not substantially altered by
[Mg2+]o,
decreasing slightly (by up to 10%) as
[Mg2+]o
was elevated (data not shown). This modest decline is most likely a
result of elevated end-diastolic pressure (Fig. 1) coupled with reduced
metabolic rate and thereby metabolic vasodilatation (e.g., see data
below for adenosine release).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1.
Effects of elevated extracellular
Mg2+ concentration
([Mg2+]o)
on end-diastolic (filled symbols) and peak systolic pressure (open
symbols) (A), heart rate
(B), and rate-pressure product
(filled symbols) and myocardial oxygen consumption
(MO2; open symbols)
(C) in Langendorff-perfused rat
hearts. Data are also shown for functional effects of 5.0 mM
[Mg2+]o
in the presence of 2.5 mM extracellular
Ca2+ concentration
([Ca2+]o)
(triangles). All values are means ± SE;
n = 8. * P < 0.05 vs. baseline
values.  P < 0.05 vs. values in the presence of normal
(1.2 mM) extracellular Ca2+.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 2.
Effects of elevated
[Mg2+]o
on first derivative of left ventricular pressure
(+dP/dt) in Langendorff-perfused rat
hearts paced at 4 Hz. Values are means ± SE;
n = 7. * P < 0.05 vs. baseline
values.
|
|
Metabolically,
[Mg2+]o
significantly elevated [PCr] and reduced
[Pi] without modifying
[ATP] (Table 2).
[Mg2+]i
was unaltered during perfusion with <8.0 mM
[Mg2+]o
(0.45-0.50 mM) but increased by 80% with 8.0 mM
[Mg2+]o
(Table 1). pHi was also unaltered
during perfusion with <8.0 mM
[Mg2+]o,
and increased slightly but significantly with 8.0 mM
[Mg2+]o
(by 0.04 pH units). As a result of these metabolic changes the
cytosolic phosphorylation ratio
([ATP]/[ADP][Pi])
and GATP were
significantly and dose dependently elevated, and free cytosolic [AMP] significantly reduced, by elevations in
[Mg2+]o
(Table 3). Metabolic parameters
([ATP], [PCr],
pHi,
[Mg2+]i,
and [AMP]) were unaltered by addition of 1.2 mM phosphate to perfusion fluid in control hearts (Tables 2 and 3). Myocardial [Pi],
[ATP]/[ADP][Pi],
or GATP were
not calculated in these hearts owing to the presence of the
extracellular phosphate resonance in
31P spectra and resultant
interference with integration of the intracellular Pi resonance.
View this table:
[in this window]
[in a new window]
|
Table 2.
Metabolic parameters in functioning hearts perfused with varying levels
of [Mg2+]o,
[Ca2+]o, or verapamil
|
|
View this table:
[in this window]
[in a new window]
|
Table 3.
Metabolic parameters in functioning hearts perfused with varying levels
of [Mg2+]o,
[Ca2+]o, or verapamil
|
|
Functional and metabolic effects of altered
[Ca2+]o
and Ca2+
antagonism with verapamil.
Reductions in
[Ca2+]o
increased end-diastolic pressure and reduced peak systolic pressure and
the rate-pressure product (Fig. 3). An
elevation in
[Ca2+]o
to 2.5 mM resulted in a slight elevation in contractile function. Interestingly, heart rate was resistant to changes in
[Ca2+]o,
not varying by more than 10% (data not shown). The magnitude of the
changes in rate-pressure product during reductions in
[Ca2+]o
was similar to those observed during elevations in
[Mg2+]o.
Infusion of the L-type Ca2+
channel antagonist verapamil (0.5 and 2.0 µM) significantly increased end-diastolic pressure and decreased contractile function (Fig. 3). In
contrast to reduced
[Ca2+]o,
infusion of verapamil substantially reduced heart rate by ~55% (data
not shown). Coronary flow was not significantly altered by reduced
[Ca2+]o
but increased slightly at the highest dose of verapamil.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3.
Effects of
[Ca2+]o
(n = 8)
(A) and verapamil
(n = 8)
(B) on end-diastolic ( ) and peak
systolic ( ) pressure and rate-pressure product ( ) in
Langendorff-perfused rat hearts. Values are means ± SE.
* P < 0.05 vs. baseline
values.
|
|
Metabolically, reductions in
[Ca2+]o
and verapamil treatment produced changes similar to those observed with
elevated
[Mg2+]o
(i.e., increased [PCr] and reduced
[Pi] without changes
in [ATP]) (Table 2).
[ATP]/[ADP][Pi]
and GATP were
increased by reduced
[Ca2+]o
and by verapamil treatment (Table 3). As shown in Table 3, changes in
extracellular Mg2+,
Ca2+, and verapamil concentrations
produced comparable improvements in cytosolic free energy state at
similar workload and MO2.
An elevation in
[Ca2+]o
from 1.2 to 2.5 mM abolished the negative inotropic effect of 5.0 mM
[Mg2+]o
(Fig. 1A) without markedly
altering end-diastolic pressure (Fig.
1B). It was found that elevated
[Ca2+]o
did not reverse the negative chronotropic effect of
[Mg2+]o.
Electrical pacing at the control rate in the presence of elevated [Ca2+]o
fully normalized the rate-pressure product between treated and
untreated hearts (Fig. 1C). With
this normalized rate-pressure product
MO2 was almost
iden- tical in control and 5.0 mM
[Mg2+]o-treated
hearts (Fig. 1C). In addition,
intracellular metabolic state remained stable in these hearts (Tables 2
and 3). Thus reversal of functional effects of 5.0 mM
[Mg2+]o
by elevated
[Ca2+]o
(and pacing) also effectively counteracted the metabolic effects of
elevated
[Mg2+]o.
Relationships between contractile function, metabolic rate
(MO2), and
GATP in
hearts perfused with varying
[Mg2+]o,
[Ca2+]o,
or verapamil.
As shown in Figs. 1 and 3, the functional effects of elevated
[Mg2+]o,
reduced
[Ca2+]o,
and verapamil treatment are similar. Moreover, the effects of these
treatments on
GATP are also
comparable (Table 3). Figure 4 depicts the
relationship between contractile function (rate-pressure product) and
MO2 in all groups. A
consistent linear correlation was obtained under all conditions studied
[MO2 = 34.7 + 0.0069 · (rate-pressure product)]. Thus the
oxygen cost of isovolumic work appears to be comparable in
Mg2+-,
Ca2+-, and verapamil-treated
hearts. Collectively, the data indicate a basal metabolic rate of ~35
µl
O2 · min1 · g1
(i.e., at zero isovolumic work). It was also found that
GATP was
comparably and linearly correlated with contractile function and
MO2 in all experimental
groups (Fig. 5). Collectively, these data
indicate that the rate of mitochondrial respiration is consistently related to workload and that cytosolic free energy state is
consistently related to workload and
MO2 in hearts treated with
either Mg2+,
Ca2+, or verapamil (Fig. 5).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Correlation between contractile function (rate-pressure product) and
MO2 in Langendorff-perfused
rat hearts subjected to varying
[Mg2+]o
(n = 8),
[Ca2+]o
(n = 8), or verapamil concentration
(n = 8). Values are means ± SE.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 5.
Relationships between free energy of ATP hydrolysis
(GATP) and
contractile function (A) and between
GATP and
MO2
(B) in Langendorff-perfused rat
hearts subjected to varying
[Mg2+]o
(n = 8),
[Ca2+]o
(n = 8), or verapamil concentration
(n = 8). Values are means ± SE.
|
|
Effects of elevated
[Mg2+]o
on myocardial adenosine formation.
The myocardial efflux of adenosine and inosine was significantly
reduced in a dose-dependent manner by elevations in
[Mg2+]o
(Fig. 6). The
[Mg2+]o-dependent
reduction in myocardial adenosine formation paralleled reductions in
MO2,
GATP, and free
cytosolic [AMP] (Fig. 7). Similar reductions in purine efflux were also observed in hearts subjected to reduced
[Ca2+]o
(data not shown).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6.
Effects of elevated
[Mg2+]o
on myocardial adenosine and sum () of adenosine + inosine efflux in
Langendorff-perfused rat hearts. Values are means ± SE;
n = 8.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 7.
Relationships between myocardial adenosine and adenosine + inosine efflux and MO2
(A),
GATP
(B), and free cytosolic
[AMP] (C) in
Langendorff-perfused rat hearts subjected to varying
[Mg2+]o.
Values are means ± SE; n = 8.
|
|
Ionic changes associated with elevations in
[Mg2+]o.
Perfusion of hearts with elevated
[Mg2+]o
did not alter intracellular K+ but
produced a small reduction in intracellular
Na+ and an elevation in total
tissue [Mg2+] (Table
4). As a result of these changes
G[Na+]o/i increased slightly but significantly with 8.0 mM
[Mg2+]o.
As
[Mg2+]o
was increased,
G[Mg2+]o/i increased from 19 to 23 kJ/mol. Inclusion of 1.2 mM phosphate in the perfusion fluid did not alter myocardial contents
of Na+,
K+, or
Mg2+ or the
G values for
Na+ and
Mg2+ (Table 4). As shown
in Fig. 8, it was calculated that
G[Mg2+]o/i increased gradually with
GATP in
the Mg2+-treated hearts.
However, in Ca2+- and
verapamil-treated hearts this relationship did not hold (Fig. 8). In
addition it was calculated that
G[Mg2+]o/i and
G[Na+]o/i
both increased as [Mg2+]o
was elevated, although
G[Mg2+]o/i remained consistently higher than
G[Na+]o/i at all times (Fig. 8).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 8.
Relationship of free energy of transsarcolemmal
Mg2+ concentration gradient
(G[Mg2+]o/i)
vs. GATP
(A) and vs. free energy of
transsarcolemmal Na+ concentration
gradient
(G[Na+]o/i)
(B) in Langendorff-perfused rat
hearts subjected to varying
[Mg2+]o
(n = 7),
[Ca2+]o
(n = 7), or verapamil concentration
(n = 6). Values are means ± SE.
|
|
Functional and metabolic effects of
[Mg2+]o
vs. verapamil in ischemic hearts.
Cardioplegic arrest with 25 mM KCl abolished contractile function and
increased GATP
before ischemia (data not shown). Addition of 8.0 mM
Mg2+ or verapamil to cardioplegic
solutions produced no further changes in energy state, function, or
MO2. Functional responses to
ischemia and reperfusion differed significantly between
treatment groups. The extent and time to ischemic contracture varied
significantly. End-diastolic pressure increased to 58 ± 12 mmHg at
the end of the ischemic period in control hearts and to 55 ± 8, 35 ± 7, and 20 ± 4 mmHg in hearts subjected to
high-K+ cardioplegic arrest alone
or with verapamil or Mg2+,
respectively. Although the K+
cardioplegia alone did not significantly alter contracture, verapamil and Mg2+ treatment significantly
reduced contracture compared with both control and
K+ cardioplegia hearts.
Contracture in the Mg2+-treated
group was significantly lower than that in the verapamil-treated group
(P < 0.05). Time to contracture was
also altered, significantly increasing from 8 ± 1 min in control
hearts to 12 ± 1, 15 ± 2, and 20 ± 2 min in hearts
subjected to high K+ cardioplegic
arrest alone or with verapamil or
Mg2+, respectively.
During reperfusion, hearts that were not arrested before
global normothermic ischemia recovered ~40% of preischemic
contractile function (Table 5).
Cardioplegic arrest with K+
alone significantly improved contractile recovery, and addition of
verapamil or Mg2+ further
improved contractile recovery during reperfusion (Table 5).
Metabolically, cardioplegia reduced the decline in ATP and elevation in
Pi during the ischemic period
(Fig. 9). Final recovery of ATP and PCr was
modestly improved in the high-K+
plus Mg2+ treatment group but not
significantly improved in any of the other groups (Fig. 9).
Pi recovered to preischemic levels
in all groups. No differences were observed in responses of
pHi or
[Mg2+]i
in the different treatment groups (data not shown).
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 9.
Metabolic responses to 30 min of global normothermic ischemia
and 20 min of reperfusion. Hearts were either untreated (control,
n = 7) or subjected to
high-K+ cardioplegia
(K+,
n = 7),
high-K+ cardioplegia + 8.0 mM
[Mg2+]o
(K+ + Mg2+,
n = 8), or
high-K+ cardioplegia + 2.0 µM
verapamil (K+ + verapamil,
n = 8). Values are means ± SE.
* P < 0.05 vs. preischemic
values; P < 0.05 vs.
control hearts;  P < 0.05 vs. K+ cardioplegia alone.
PCr, phosphocreatine.
|
|
| |
DISCUSSION |
We characterized metabolic and functional effects of varying levels of
extracellular Mg2+ in the rat
myocardium. To date the dose dependence of the metabolic effects of
Mg2+ has not been documented in
mammalian myocardium. We also compared effects of elevated
[Mg2+]o
with those of reduced
[Ca2+]o
and Ca2+ antagonism, and the
functional and metabolic impact of
Mg2+ vs. verapamil treatment
during ischemia. Our results indicate that
[Mg2+]o
within the 1.2- to 8.0-mM range primarily modifies cytosolic energy
state via reducing energy demand and/or
Ca2+ activation of the heart.
These effects are mediated by extracellular and not intracellular
Mg2+. Additionally, we find no
evidence for direct
Mg2+-dependent modulation of
adenosine formation in rat heart. Finally, it was found that beneficial
effects of 8.0 mM
[Mg2+]o
exceed those for a functionally equipotent
Ca2+ antagonist during
ischemia-reperfusion.
Functional effects of 1.2-8.0 mM
[Mg2+]o.
It is generally considered that
[Mg2+]o
exerts negative inotropic effects in hearts. However, conflicting data
exist (e.g., Refs. 6 and 32), and the role of intracellular
Mg2+ in mediating these responses
is undefined. In the present study heart rate and contractile function
were dose dependently reduced by
[Mg2+]o,
in agreement with documented electrophysiological and functional effects of Mg2+ in individual
cells, isolated muscle, and intact hearts (1, 2, 8, 12, 22, 31, 48, 49,
55). Responses in rat tissues are more pronounced than those in other
species (8, 22, 30). Inhibitory effects of
[Mg2+]o
may be mediated by suppression of sarcolemmal
Ca2+ fluxes (1, 2, 11, 22, 49, 55)
and/or inhibitory effects of associated changes in
[Mg2+]i.
We eliminate the latter on the basis of
1) rapidity of functional alterations (maximal in 1-2 min) and
2) absence of detectable changes in
[Mg2+]i
with <8.0 mM
[Mg2+]o.
In addition, inotropic effects of elevated
[Mg2+]o
were effectively reversed by elevating
[Ca2+]o
(Fig. 1). These data, and comparable relationships observed between
function, MO2, and
GATP in
[Mg2+]o-,
[Ca2+]o-,
and verapamil-treated hearts (Figs. 4 and 5), are consistent with a
Ca2+-dependent mechanism of action
of elevated
[Mg2+]o
(1, 2, 12, 30, 31, 41, 48, 51, 55).
There is controversy with respect to contractile effects of moderate
elevations in
[Mg2+]o
in heart. Most studies demonstrate negative inotropic effects of
Mg2+ in isolated myocytes (22,
49), papillary or septal muscle (22, 48), perfused hearts (8), and the
in situ heart (17). However, Barbour et al. (6) report
positive inotropic effects of 4.8 mM
[Mg2+]o
in perfused hearts, and James et al. (32) report positive inotropic
effects in the baboon. Explanations for these responses included
displacement of bound Ca2+ by
intracellular Mg2+, activation of
adenylate cyclase, increased cross-bridge number, and intracellular
alkalinization (6). The present data demonstrate that 1.2-8.0 mM
[Mg2+]o
consistently reduces inotropic state and heart rate in intact myocardium, irrespective of
[Mg2+]i.
Thus we eliminate potential effects of
[Mg2+]i
on Ca2+ mobilization, adenylate
cyclase activity, and cross-bridge number. With respect to inotropic
effects of alkalinization (6), we also observe increased pH with
elevated
[Mg2+]o
(Table 2), and a comparable alkalinization occurred with verapamil (Table 2), suggesting alkalinization is related to
Ca2+ channel inhibition, reduced
metabolic rate, and/or elevated
GATP, rather
than a specific effect of Mg2+.
Importantly, alkalinization in both experiments was not associated with
positive inotropism. The present findings, in conjunction with previous
studies (22, 48, 49), show that moderate elevations in
[Mg2+]o
exert negative inotropic effects in myocardial tissue, and these
effects are mediated via extracellular mechanisms, probably involving
Ca2+ channel blockade. The
positive inotropism observed by Barbour et al. (6) in working hearts
may have resulted from the inverse dependence of ventricular force and
Ca2+ flux on heart rate (e.g., see
Ref. 14), coupled with increased ventricular filling and stroke volume
during Mg2+-dependent bradycardia.
Metabolic effects of 1.2-8.0 mM
[Mg2+]o.
In addition to a positive inotropic response, Barbour et al. (6)
documented improved cellular energy state, at a given workload, in
response to a fourfold elevation in
[Mg2+]o.
The authors suggested this improvement involved elevated
[Mg2+]i.
According to a study by Saks et al. (46), elevated
[Mg2+]i
directly enhances ATP formation. Moreover, because enzymes in
carbohydrate metabolism are Mg2+
dependent (19, 46), elevated
[Mg2+]i
may improve substrate metabolism.
[Mg2+]i
also inhibits mitochondrial Ca2+
transport (13, 53), which could reduce energetic competition between
Ca2+ transport and ATP production
(13). It is therefore feasible that elevated
[Mg2+]i
could contribute to improved cytosolic free energy level during elevations in
[Mg2+]o.
Indeed, elevations in
[Mg2+]o
from 1.2 to 2.4, 5.0, and 8.0 mM significantly elevate
GATP (Table
3). However, we obtain no evidence of a specific/direct effect of
[Mg2+]o
on energy metabolism, and no evidence for mediation of the effect by
[Mg2+]i,
because substantial changes were observed in the absence of altered
[Mg2+]i.
Changes in
[Mg2+]o,
[Ca2+]o,
or verapamil all produce almost identical changes in
GATP at
similar MO2 and rate-pressure
product (Figs. 4 and 5, Table 3). Myocardial
GATP displayed
a consistent linear correlation with contractile function and
MO2 in all groups (Fig. 5).
Moreover, effects of
[Mg2+]o
on GATP were
reversed by elevated
[Ca2+]o
and pacing to normalize function. Therefore, although we agree that
modestly elevated
[Mg2+]o
improves cytosolic free energy state, we find no evidence that this is
specific for Mg2+ or that it
involves changes in
[Mg2+]i.
These effects of 1.2-8.0 mM
[Mg2+]o
appear secondary to altered cardiac work and are consistent with
Ca2+-dependent functional effects
of
[Mg2+]o
at extracellular sites (49).
Impact of
[Mg2+]o
on
[Mg2+]i.
As already noted,
[Mg2+]i
was stable during perfusion with <8.0 mM
[Mg2+]o.
Using 31P NMR spectroscopic
techniques, we calculated baseline
[Mg2+]i
to be ~0.45 mM in Langendorff-perfused rat heart (Table 2). This is
consistent with our previous studies in in situ rat and rabbit
myocardium (28, 40) and with 31P
and 19F NMR measurements in rat
and guinea pig hearts (6, 24-26, 34, 38, 42). Estimation of
[Mg2+]i
in guinea pig hearts from the
Mg2+-dependent
glyceraldehyde-3-phosphate dehydrogenase/phosphoglycerate kinase couple
yields values of ~0.6 mM (37, 38), which also agree with the present
estimate. Ion-selective microelectrode studies yield similar myocyte
[Mg2+]i
values (9, 10, 22). Alternatively, estimates of rat heart
[Mg2+]i
from the aconitase equilibrium are variable, ranging from 0.4 to 1.2 mM
(33, 39). This latter technique may be inappropriate in rapidly
respiring myocardium owing to displacement of the reaction from
equilibrium (37).
The effects of elevated
[Mg2+]o
on
[Mg2+]i
are controversial. Stability of
[Mg2+]i
with up to 5.0 mM
[Mg2+]o
was predicted on the basis of previous observations in isolated myocardial tissue (9, 10, 22, 23, 35, 49). Recent ion-selective
microelectrode studies reveal very modest (if any) elevations in
myocardial
[Mg2+]i
with elevations in
[Mg2+]o
from 0.5 to 10-20 mM (9, 10). Studies with fluorescent Mg2+ probes show that
[Mg2+]i
is stable during exposure of myocytes to 5-15 mM
[Mg2+]o
(23, 35, 49). Studies in vascular myocytes also demonstrate stability
of
[Mg2+]i
with up to 10 mM
[Mg2+]o
(20). These previous observations and the present data demonstrate that
quite large changes in
[Mg2+]o
are necessary to modify
[Mg2+]i.
The substantial sarcolemmal electrochemical gradients for Mg2+ require active
Mg2+ extrusion mechanisms: if
permitted to reach electrochemical equilibrium at a membrane potential
of 85 mV (Table 4),
[Mg2+]i
would increase to ~150 mM. There is some evidence for
Na+-dependent
Mg2+ efflux in mammalian myocytes
(18, 23, 45). Nevertheless, the existence of myocardial
Na+/Mg2+
exchange is controversial, and there is good evidence against such a
mechanism (9, 10). Handy et al. (23) recently obtained data favoring a
myocardial
Na+-Mg2+
antiport. However, their results were also consistent with the operation of other Na+-
and/or Ca2+-dependent
pathways (23).
In examining a potential energy dependence of the transsarcolemmal
Mg2+ gradient, we calculated that
G[Mg2+]o/i increases gradually as
GATP rises
during Mg2+ infusion (Fig. 8).
Although this tends to support energy dependence of sarcolemmal
Mg2+ efflux, or coupling of
Mg2+ efflux to other
energy-dependent ion movements (e.g.,
Na+ transport), this relationship
was not observed during verapamil- or
[Ca2+]o-dependent
changes in
GATP (Fig. 8).
Thus this response appears to be due to
Mg2+-dependent elevations in
GATP
coincident with direct elevations in
G[Mg2+]o/i
during perfusion with high
[Mg2+]o.
We find no evidence for energy-dependent
Mg2+ extrusion. Interestingly, we
observe that
G[Mg2+]o/i increases as the transsarcolemmal
Na+ gradient rises under all
conditions, and
G[Mg2+]o/i is consistently higher than
G[Na+]o/i (Fig. 8). These data are compatible with electroneutral 2:1
Na+/Mg2+
exchange (15, 16, 18). Movement of two
Na+ down an electrochemical
gradient, equivalent to a G of 15 kJ/mol, more than matches the G for
countermovement of a single Mg2+
against its electrochemical gradient (19-23 kJ/mol). A final potential mechanism for Mg2+
extrusion is the action of a
Mg2+-ATPase (15, 16), similar to
those located in bacteria. However, such an enzyme has not been
localized in heart tissue, and, as shown here, we observe no consistent
relationship between
G[Mg2+]o/i and GATP,
which would be predicted if the
Mg2+ gradient were dependent on a
sarcolemmal ATPase in a Gibbs-Donnan near-equilibrium system
(39).
Effects of elevated
[Mg2+]o
in ischemic myocardium.
Because functional and metabolic effects of
Mg2+ and verapamil were similar,
we compared ischemic cardioprotection afforded by 8.0 mM
[Mg2+]o
vs. verapamil. Cardioplegic arrest improved functional and metabolic
outcome from ischemia, and addition of
[Mg2+]o
or verapamil to cardioplegic solutions further enhanced recovery (Table
5, Fig. 9). Importantly, 8.0 mM
[Mg2+]o
provided superior functional and metabolic cardioprotection compared
with the Ca2+ antagonist. Because
all cardioplegic solutions examined reduced contractile function to
comparable levels before ischemia, we eliminate a role for
preischemic metabolic rate in the differences in recovery (although
reduced preischemic workload generally plays a role in cardioplegic
protection). Curiously, in examining metabolic effects of cardioplegia
with and without Mg2+ as an
additive, we observed discrepancies between NMR-detected changes in
[Pi] in control vs.
cardioplegia-treated hearts (Fig. 9). For example,
[Pi] was ~20 mM
lower in Mg2+-treated hearts vs.
control hearts despite only a 3 mM difference in [ATP]
(which would explain a 9 mM difference in liberated
Pi, leaving a discrepancy of ~10
mM Pi). We have no explanation
for reduced [Pi]
during ischemia in cardioplegic versus control hearts but
recognize that complex relationships exist between total and freely
mobile (NMR detected) Pi during
ischemia as a result of saturable binding of intracellular
Pi (4). Although speculative, it
may be that cardioplegia enhances this binding to reduce free [Pi] at any given
total tissue concentration.
Previous studies have documented beneficial functional effects of
Mg2+ in
high-K+ cardioplegia (5, 53, 54),
although the mechanism remains unclear. Effects of elevated
[Mg2+]o
will involve inhibition of ischemic
Ca2+ accumulation (5, 54).
However, because Mg2+ and
verapamil both inhibit Ca2+ entry
via sarcolemmal Ca2+ channels (1,
2, 22, 30, 31, 55), and similar degrees of
Ca2+ channel antagonism are
indicated by comparable effects of
[Mg2+]o
and verapamil on preischemic function, additional mechanisms must be
involved in enhanced protection with
Mg2+. Interestingly,
Ca2+ channel blockade has been
shown to only partially attenuate myocardial Ca2+ entry, with remaining entry
likely mediated by
Na+/Ca2+
exchange (7). Elevated
[Mg2+]o
could therefore inhibit Ca2+ entry
via
Na+/Ca2+
exchange, as suggested by Koss and Grubbs (35). However,
Mg2+ exerts only minor effects on
sarcolemmal
Na+/Ca2+
exchange (52). Additional mechanisms include effects of elevated [Mg2+]i
on ATP synthesis and creatine kinase kinetics (45), inhibition of
mitochondrial Ca2+ accumulation
(13, 53), and modulation of SR
Ca2+ uptake and release (12, 41,
51).
Effects of
[Mg2+]o
on myocardial adenosine formation.
Release of adenosine is important in control of cardiovascular function
and in protecting the heart from ischemic injury. Adenosine formation
is considered to be partially Mg2+
dependent owing to the Mg2+
sensitivity of 5'-nucleotidase (24, 38). It is therefore possible
that beneficial effects of Mg2+
could stem in part from adenosine receptor activation subsequent to
Mg2+-dependent elevations in
adenosine formation (38). However, we observed consistent reductions
rather than elevations in myocardial adenosine and inosine efflux with
increased
[Mg2+]o
(Fig. 6). Reductions in adenosine efflux parallel changes in MO2,
GATP, and free
cytosolic [AMP] (Fig. 7). These correlations support the
hypothesis that myocardial adenosine formation proceeds via
dephosphorylation of intracellular AMP (27, 40), which increases
substantially during reductions in energy state (24, 28, 40). In this
way formation of vasoactive adenosine is coupled to myocardial energy
status. The findings are contrary to the observations of Mallet et al.
(38), who recently documented Mg2+-dependent activation of
ecto-5'-nucleotidase and adenosine formation despite reduced
intracellular [AMP] in guinea pig hearts.
Reasons for the different observations in rat vs. guinea pig are
unclear but may involve species differences in 5'-nucleotidases. Whereas cytosolic 5'-nucleotidases from dog, rabbit, and rat
hearts display absolute requirements for
Mg2+, with 50% maximal activation
at 1-3 mM (11, 43, 57), membrane-bound ectoenzyme in rat heart has
no absolute requirement for Mg2+
and is only modestly activated (by 70-80%) by elevations in
Mg2+ from 0 to 10 mM (43). On the
other hand, whereas the canine membrane-bound enzyme also has no
absolute requirement for Mg2+, it
is much more sensitive, being activated by 300-400% with elevations in Mg2+ from 0 to only
6 mM (11). Thus Mg2+ sensitivity
of ecto-5'-nucleotidase appears to be very modest in rat heart
and greater in other species, possibly contributing to the enhanced
adenosine release observed in
Mg2+-treated guinea pig hearts
(38) but not in Mg2+-treated rat
hearts (present data).
In conclusion, our data indicate that graded elevations
in
[Mg2+]o
from 1.2 to 8.0 mM improve myocardial energy metabolism through reductions in workload,
MO2, and/or
Ca2+ activation. Effects of
elevated
[Mg2+]o
on function and energy metabolism are independent of intracellular [Mg2+], supporting an
extracellular locus of action. Functional and metabolic effects of
[Mg2+]o
are mimicked by direct reductions in
[Ca2+]o
or treatment with a Ca2+
antagonist and can be countered by elevations in
[Ca2+]o,
supporting Ca2+ dependence of the
effects of
[Mg2+]o.
Mg2+ therefore does not exert
unique effects on function and energy state in intact heart.
Nevertheless, the data provide experimental support for direct
beneficial effects of high
[Mg2+]o
on metabolic and functional variables in hearts subjected to cardioplegic arrest and ischemia. Finally, myocardial adenosine formation in rat heart is not enhanced by
[Mg2+]o
but declines in parallel with
Mg2+-dependent reductions in
workload and improvement in energy state. Therefore, the functional and
metabolic effects of elevated
[Mg2+]o
in rat myocardium do not involve adenosine receptor activation, and
optimal protection of ischemic heart with
Mg2+ does not involve enhanced
adenosine-mediated cardioprotection.
| |
ACKNOWLEDGEMENTS |
This research was supported in part by grants from the National
Heart Foundation of Australia and the National Health and Medical
Research Council. J. McKirdy was supported by a Ph.D. scholarship from
Griffith University.
| |
FOOTNOTES |
Address reprint requests to J. P. Headrick.
Received 3 November 1997; accepted in final form 8 May 1998.
| |
REFERENCES |
1.
Agus, Z. A.,
E. Kellepouris,
I. Dukes,
and
M. Morad.
Cytosolic magnesium modulates calcium channel activity in mammalian ventricular cells.
Am. J. Physiol.
256 (Cell Physiol. 25):
C452-C455,
1989[Abstract/Free Full Text].
2.
Almers, W.,
and
E. W. McCleskey.
Non-selective conductance in calcium channels of frog muscle: calcium selectivity in single-file pore.
J. Physiol. (Lond.)
353:
585-608,
1984[Abstract/Free Full Text].
3.
Altura, B. M.,
and
B. T. Altura.
New perspectives on the role of magnesium in the pathophysiology of the cardiovascular system. II. Experimental aspects.
Magnesium
4:
226-244,
1985[Medline].
4.
Armiger, L. C.,
J. P. Headrick,
L. R. Jordan,
and
R. J. Willis.
Bound inorganic phosphate and early contractile failure in global ischaemia.
Basic Res. Cardiol.
90:
482-488,
1996.
5.
Ataka, K.,
D. Chen,
J. McCully,
S. Levitsky,
and
H. Feinberg.
Magnesium cardioplegia prevents accumulation of cytosolic calcium in the ischemic myocardium.
J. Mol. Cell. Cardiol.
25:
1387-1390,
1993[Medline].
6.
Barbour, R. L.,
B. M. Altura,
S. D. Reiner,
T. L. Dowd,
R. K. Gupta,
F. Wu,
and
B. T. Altura.
Influence of Mg2+ on cardiac performance, intracellular free Mg2+, and pH in perfused hearts as assessed with 31P nuclear magnetic resonance spectroscopy.
Magnesium Trace Elem.
10:
99-116,
1992.
7.
Barry, W. H.,
and
T. W. Smith.
Mechanisms of transmembrane calcium movement in cultured chick embryo ventricular cells.
J. Physiol. (Lond.)
325:
243-260,
1982[Abstract/Free Full Text].
8.
Bersohn, M. M.,
K. I. Shine,
and
W. D. Sterman.
Effect of increased magnesium on recovery from ischemia in rat and rabbit hearts.
Am. J. Physiol.
242 (Heart Circ. Physiol. 11):
H89-H93,
1982.
9.
Buri, A.,
S. Chen,
C. H. Fry,
H. Illner,
E. Kickenweize,
J. A. S. McGuigan,
D. Noble,
T. Powell,
and
V. W. Twist.
The regulation of intracellular Mg2+ in guinea-pig heart, studied with Mg2+-selective microelectrodes and fluorochromes.
Exp. Physiol.
78:
221-233,
1993[Abstract].
10.
Buri, A.,
and
J. A. S. McGuigan.
Intracellular free magnesium and its regulation, studied in isolated ferret ventricular muscle with ion-selective microelectrodes.
Exp. Physiol.
75:
751-761,
1990[Abstract].
11.
Darvish, A.,
R. W. Pomerantz,
P. G. Zografides,
and
P. J. Metting.
Contribution of cytosolic and membrane-bound 5'-nucleotidases to cardiac adenosine production.
Am. J. Physiol.
271 (Heart Circ. Physiol. 40):
H2162-H2167,
1996[Abstract/Free Full Text].
12.
Fabiato, A.,
and
F. Fabiato.
Effects of pH on the myofilaments and the sarcoplasmic reticulum of skinned cells from cardiac and skeletal muscle.
J. Physiol. (Lond.)
249:
497-517,
1975[Abstract/Free Full Text].
13.
Ferrari, R.,
P. Pedersini,
M. Bongrazio,
G. Gaia,
P. Bernocchi,
F. Di Lisa,
and
O. Visioli.
Mitochondrial energy production and cation control in myocardial ischaemia and reperfusion.
Basic Res. Cardiol.
88:
495-512,
1993[Medline].
14.
Field, M. L.,
A. Azzawi,
J. F. Unitt,
A.-M. L. Seymour,
C. Henderson,
and
G. K. Radda.
Intracellular [Ca2+] staircase in the isovolumic pressure-frequency relationship of Langendorff-perfused rat heart.
J. Mol. Cell. Cardiol.
28:
65-77,
1996[Medline].
15.
Flatman, P. W.
Magnesium transport across cell membranes.
J. Membr. Biol.
80:
1-14,
1983.
16.
Flatman, P. W.
Mechanisms of magnesium transport.
Annu. Rev. Physiol.
53:
259-271,
1991[Medline].
17.
Friedman, H. S.,
T. N. Nguyen,
A. M. Mokraoui,
R. L. Barbour,
T. Murakawa,
and
B. M. Altura.
Effects of magnesium chloride on cardiovascular hemodynamics in the neurally intact dog.
J. Pharmacol. Exp. Ther.
243:
126-130,
1987[Abstract/Free Full Text].
18.
Fry, C. H.
Measurement and control of intracellular magnesium ion concentration in guinea pig and ferret ventricular myocardium.
Magnesium
5:
306-331,
1986[Medline].
19.
Garfinkel, L.,
R. A. Altschuld,
and
D. Garfinkel.
Magnesium in cardiac energy metabolism.
J. Mol. Cell. Cardiol.
18:
1003-1013,
1986[Medline].
20.
Gilbert, E. K.,
H. A. Singer,
and
C. M. Rembold.
Magnesium relaxes arterial smooth muscle by decreasing intracellular Ca2+ without changing intracellular Mg2+.
J. Clin. Invest.
89:
1988-1994,
1994.
21.
Gupta, R. K.,
J. L. Benovic,
and
Z. B. Rose.
The determination of the free magnesium level in the human red blood cell by 31P NMR.
J. Biol. Chem.
253:
6172-6176,
1978[Abstract/Free Full Text].
22.
Hall, S. K.,
and
C. H. Fry.
Magnesium affects excitation, conduction, and contraction of isolated mammalian cardiac muscle.
Am. J. Physiol.
263 (Heart Circ. Physiol. 32):
H622-H633,
1992[Abstract/Free Full Text].
23.
Handy, R. D.,
I. F. Gow,
D. Ellis,
and
P. W. Flatman.
Na-dependent regulation of intracellular free magnesium concentration in isolated rat ventricular myocytes.
J. Mol. Cell. Cardiol.
28:
1641-1652,
1996[Medline].
24.
Headrick, J.,
and
R. J. Willis.
5'-Nucleotidase activity and adenosine formation in stimulated, hypoxic, and underperfused rat heart.
Biochem. J.
261:
541-550,
1989[Medline].
25.
Headrick, J. P.,
and
R. J. Willis.
Effect of inotropic stimulation on cytosolic Mg2+ in isolated rat heart: a 31P magnetic resonance study.
Magn. Reson. Med.
12:
328-338,
1989[Medline].
26.
Headrick, J. P.,
and
R. J. Willis.
Cytosolic free magnesium in stimulated, hypoxic, and underperfused rat heart.
J. Mol. Cell. Cardiol.
23:
991-999,
1991[Medline].
27.
Headrick, J. P.,
G. P. Matherne,
and
R. M. Berne.
Myocardial adenosine formation during hypoxia: effects of ecto-5'-nucleotidase inhibition.
J. Mol. Cell. Cardiol.
24:
295-304,
1992[Medline].
28.
Headrick, J. P.,
G. P. Dobson,
J. P. Williams,
J. C. McKirdy,
L. R. Jordan,
and
R. J. Willis.
Bioenergetics and control of oxygen consumption in the in situ rat heart.
Am. J. Physiol.
267 (Heart Circ. Physiol. 36):
H1074-H1084,
1994[Abstract/Free Full Text].
29.
Hearse, D. J.,
D. A. Stewart,
and
M. V. Baimbridge.
Myocardial protection during ischemic cardiac arrest. The importance of magnesium in cardioplegic infusates.
J. Thorac. Cardiovasc. Surg.
75:
877-885,
1978[Abstract].
30.
Hess, P.,
and
R. W. Tsien.
Mechanism of ion permeation through calcium channels.
Nature
309:
453-456,
1984[Medline].
31.
Iseri, L. T.,
and
J. H. French.
Magnesium: nature's physiologic calcium blocker.
Am. Heart J.
108:
188-193,
1984[Medline].
32.
James, M. F. M.,
R. C. Cork,
and
J. E. Dennett.
Cardiovascular effects of magnesium sulphate in the baboon.
Magnesium
6:
314-324,
1987[Medline].
33.
Kashiwaya, Y.,
K. Sato,
N. Tsuchiya,
S. Thomas,
D. A. Fell,
R. L. Veech,
and
J. V. Passonneau.
Control of glucose utilization in working perfused rat heart.
J. Biol. Chem.
269:
25502-25514,
1994[Abstract/Free Full Text].
34.
Kirkels, J. H.,
C. J. A. Van Echteld,
and
T. J. C. Ruigrok.
Intracellular magnesium during myocardial ischemia and reperfusion: possible consequences for postischemic recovery.
J. Mol. Cell. Cardiol.
21:
1209-1218,
1989[Medline].
35.
Koss, K. L.,
and
R. D. Grubbs.
Elevated extracellular Mg2+ increases Mg2+ buffering through a Ca-dependent mechanism in cardiomyocytes.
Am. J. Physiol.
267 (Cell Physiol. 36):
C633-C641,
1994[Abstract/Free Full Text].
36.
Lawson, J. W. R.,
and
R. L. Veech.
Effects of pH and Mg2+ on the Keq of the creatine kinase reaction and other phosphate hydrolyses and phosphate transfer reactions.
J. Biol. Chem.
254:
6528-6537,
1979[Abstract/Free Full Text].
37.
Mallet, R. T.,
Y.-H. Kang,
N. Mukohara,
and
R. Bunger.
Use of cytosolic metabolite patterns to estimate free magnesium in normoxic myocardium.
Biochim. Biophys. Acta
1139:
239-247,
1992[Medline].
38.
Mallet, R. T.,
J. Sun,
W.-L. Fan,
Y.-H. Kang,
and
R. Bunger.
Magnesium activated adenosine formation in intact perfused heart: predominance of ecto 5'-nucleotidase during hypermagnesemia.
Biochim. Biophys. Acta
1290:
165-176,
1996[Medline].
39.
Masuda, T.,
G. P. Dobson,
and
R. L. Veech.
The Gibbs-Donnan near-equilibrium system of heart.
J. Biol. Chem.
265:
20321-20334,
1990[Abstract/Free Full Text].
40.
Matherne, G. P.,
S. S. Berr,
and
J. P. Headrick.
Integration of vascular, contractile and metabolic responses to hypoxia: effects of maturation and the role of adenosine.
Am. J. Physiol.
270 (Regulatory Integrative Comp. Physiol. 39):
R895-R905,
1996[Abstract/Free Full Text].
41.
Meissner, G.,
and
J. S. Henderson.
Rapid calcium release from cardiac sarcoplasmic reticulum vesicles is dependent on Ca2+ and is modulated by Mg2+, adenine nucleotide, and calmodulin.
J. Biol. Chem.
262:
3065-3073,
1987[Abstract/Free Full Text].
42.
Murphy, E.,
C. Steenbergen,
L. A. Levy,
B. Raju,
and
R. E. London.
Cytosolic free magnesium levels in ischemic rat heart.
J. Biol. Chem.
26:
5622-5627,
1989.
43.
Naito, Y.,
and
J. M. Lowenstein.
5'-Nucleotidase from rat heart.
Biochemistry
20:
5188-5194,
1981[Medline].
44.
Rasmussen, H. S.,
P. Norregard,
O. Lindeneg,
P. McNair,
V. Backer,
and
S. Balslev.
Intravenous magnesium in acute myocardial infarction.
Lancet
1:
234-236,
1986[Medline].
45.
Romani, A.,
and
A. Scarpa.
Regulation of cell magnesium.
Arch. Biochem. Biophys.
298:
1-12,
1992[Medline].
46.
Saks, V. A.,
G. B. Chernousova,
D. E. Gukovsky,
V. N. Smirnov,
and
E. I. Chazov.
Studies of energy transport in heart cells. Mitochondrial isozyme of creatine phosphokinase: kinetic properties and regulatory actions of Mg2+ ions.
Eur. J. Biochem.
57:
273-290,
1975[Medline].
47.
Shechter, M.,
H. Hod,
N. Marks,
S. Behar,
E. Kaplinsky,
and
B. Rabinowitz.
Beneficial effect of magnesium sulfate in acute myocardial infarction.
Am. J. Cardiol.
66:
271-274,
1990[Medline].
48.
Shine, K. I.,
and
A. M. Douglas.
Magnesium effects on ionic exchange and mechanical function in rat ventricle.
Am. J. Physiol.
227:
317-324,
1974.
49.
Silverman, H. S.,
F. D. Lisa,
R. C. Hui,
H. Miyata,
S. J. Sollott,
R. G. Hansford,
E. G. Lakatta,
and
M. D. Stern.
Regulation of intracellular free Mg2+ and contraction in single adult mammalian cardiac myocytes.
Am. J. Physiol.
266 (Cell Physiol. 35):
C222-C233,
1994[Abstract/Free Full Text].
50.
Teague, W. E.,
and
G. P. Dobson.
Effect of temperature on the creatine kinase equilibrium.
J. Biol. Chem.
267:
14084-14093,
1992[Abstract/Free Full Text].
51.
Terada, H.,
H. Hayashi,
N. Noda,
H. Satoh,
H. Katoh,
and
N. Yamazaki.
Effects of Mg2+ on Ca2+ waves and Ca2+ transients of rat ventricular myocytes.
Am. J. Physiol.
270 (Heart Circ. Physiol. 39):
H907-H914,
1996[Abstract/Free Full Text].
52.
Trosper, T. L.,
and
K. D. Philipson.
Effects of divalent and trivalent cations on Na+-Ca2+ exchange in cardiac sarcolemmal vesicles.
Biochim. Biophys. Acta
731:
63-68,
1983[Medline].
53.
Tsukube, T.,
J. D. McCully,
E. A. Faulk,
M. Federman,
J. LoCicero,
I. B. Krukenkamp,
and
S. Levitsky.
Magnesium cardioplegia reduces cytosolic and nuclear calcium accumulation and DNA fragmentation in the senescent myocardium.
Ann. Thorac. Surg.
58:
1005-1011,
1994[Abstract].
54.
Tsukube, T.,
J. D. McCully,
K. R. Metz,
C. U. Cook,
and
S. Levitsky.
Amelioration of ischemic calcium overload correlates with high-energy phosphates in senescent myocardium.
Am. J. Physiol.
273 (Heart Circ. Physiol. 42):
H418-H425,
1997[Abstract/Free Full Text].
55.
White, R. E.,
and
H. C. Hartzell.
Effects of intracellular free magnesium on calcium current in isolated cardiac myocytes.
Science
239:
778-780,
1988[Abstract/Free Full Text].
56.
Williams, J. P.,
G. P. Dobson,
and
J. P. Headrick.
Acute changes in cytosolic magnesium in the in situ rat heart during adrenergic stimulation (Abstract).
FASEB J.
8:
A534,
1994.
57.
Yamazaki, Y.,
V. L. Truong,
and
J. M. Lowenstein.
5'-Nucleotidase I from rabbit heart.
Biochemistry
30:
1503-1509,
1991[Medline].
Am J Physiol Heart Circ Physiol 275(3):H917-H929
0002-9513/98 $5.00
Copyright © 1998 the American Physiological Society
Top
Abstract
Introduction
![[Mg<SUP>2+</SUP>]<SUB>i</SUB> = <IT>K</IT><SUP>Mg − ATP</SUP><SUB>D</SUB> (&phgr;<SUP>−1</SUP> − 1)](/content/vol275/issue3/fulltext/H917/img002.gif)
![&phgr; = <FR><NU>[ATP]<SUB>free</SUB></NU><DE>[ATP]<SUB>total</SUB></DE></FR> = <FR><NU>&dgr;<SUB>&agr;&bgr;</SUB> − &dgr;<SUP>Mg − ATP</SUP><SUB>&agr;&bgr;</SUB></NU><DE>&dgr;<SUP>ATP</SUP><SUB>&agr;&bgr;</SUB> − &dgr;<SUP>Mg − ATP</SUP><SUB>&agr;&bgr;</SUB></DE></FR>](/content/vol275/issue3/fulltext/H917/img003.gif)
![[ADP] = ([Cr] × [ATP])/([PCr] × <IT>K</IT><SUB>ck</SUB>′)](/content/vol275/issue3/fulltext/H917/img005.gif)
![[5′ − AMP] = <IT>K</IT> ′<SUB>ak</SUB> × [ADP]<SUP>2</SUP>/[ATP]](/content/vol275/issue3/fulltext/H917/img006.gif)
![&Dgr;<IT>G</IT><SUB>ATP</SUB> = &Dgr;<IT>G</IT><SUP>o</SUP><SUB>ATP</SUB> + <IT>RT </IT>ln([ADP][P<SUB>i</SUB>]/[ATP])](/content/vol275/issue3/fulltext/H917/img007.gif)
![<IT>E</IT><SUB>[ion]<SUB>o / i</SUB></SUB> = <FR><NU><IT>RT</IT> ln([ion]<SUB>o</SUB> /[ion]<SUB>i</SUB>)</NU><DE><IT>nF</IT></DE></FR>](/content/vol275/issue3/fulltext/H917/img009.gif)
![&Dgr;<IT>G</IT><SUB>[ion]<SUB>o / i</SUB></SUB> = (<IT>E</IT><SUB>[ion]<SUB>o / i</SUB></SUB> − <IT>E</IT><SUB>[K<SUP>+</SUP>]<SUB>o / i</SUB></SUB>) × <IT>nF</IT>](/content/vol275/issue3/fulltext/H917/img010.gif)
