| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 1996;27:957-964.)
© 1996 American Heart Association, Inc.
Articles |
23Na Nuclear Magnetic Resonance Spectral Changes During and After Forebrain Ischemia in Hypoglycemic, Normoglycemic, and Hyperglycemic Rats
From the Departments of Chemistry (R.L.T., J.P.), Pharmacology and Therapeutics (J.P.), and Radiology (J.P.), University of Manitoba; the Department of Chemistry, University of Winnipeg (Manitoba) (J.P.); and the Department of Neurosurgery, Foothills Hospital, Calgary, Alberta (G.R.S.), Canada.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Background and Purpose The severity of brain injury in animal models of forebrain ischemia increases with blood glucose level. During ischemia, energy failure is slower and maintenance of ion gradients is prolonged as the level of glycemia increases. It is not clear how the level of glycemia influences recovery of ion homeostasis on reperfusion. It has been shown that changes in the intensity of the multiple-quantum 23Na nuclear magnetic resonance (NMR) signals reflect changes in intracellular Na+ levels. We have used 23Na NMR spectroscopy to evaluate the influence of the level of glycemia on changes in Na+ concentration during and after forebrain ischemia in rats.
Methods Single-quantum (SQ) and double-quantum (DQ) 23Na NMR spectra were measured before and during 10-minute forebrain ischemia and during reperfusion in hypoglycemic, normoglycemic, and hyperglycemic rats.
Results The DQ 23Na NMR signal increased to 210% of preischemia intensity in all rats, but a delay in this increase was observed in normoglycemic and hyperglycemic animals. The rate of the DQ 23Na NMR signal increase was fastest in hypoglycemic (apparent first-order rate constant 0.673±0.046 min-1, P<.002 compared with normoglycemic animals) and slowest in hyperglycemic (0.285±0.024 min-1, P<.03) rats. During reperfusion, the signal intensity recovered rapidly in hypoglycemic (0.385±0.050 min-1) and normoglycemic (0.464±0.047 min-1) rats, whereas in hyperglycemic animals recovery was slow (0.108±0.044 min-1, P<.0001 compared with normoglycemic animals). The SQ 23Na NMR signal intensity increased to 117% of preischemia level in hypoglycemic (P<.05 compared with normoglycemic animals) and to 107% in normoglycemic and hyperglycemic animals during reperfusion.
Conclusions The slower increase in the 23Na DQ NMR signal intensity during forebrain ischemia in rats with higher blood glucose levels suggests that Na+ homeostasis is maintained longer in these animals. On reperfusion, the slower recovery of the DQ 23Na NMR signal intensity in hyperglycemic animals likely indicates a slower recovery of Na+ homeostasis, perhaps contributing to the increased neuronal injury after cerebral ischemia in hyperglycemic animals.
Key Words: cerebral ischemia, transient • glucose • sodium • spectroscopy, nuclear magnetic resonance • rats
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Approximately 40% of anaerobic metabolism in resting brain is dedicated to the maintenance of ion homeostasis,1 which enables the transmission of neuronal impulses and the maintenance of cell volume, substrate transport, and other processes. The Na+ gradient, maintained by Na+/K+ ATPase, is of particular importance because this form of potential energy is used to maintain Ca2+ homeostasis through the Na+/Ca2+ exchanger, to maintain optimal intracellular pH through the Na+/H+ antiport, and to cotransport glucose and amino acids across the cell membrane. During cerebral ischemia, high-energy phosphates are rapidly depleted, so ion gradients can no longer be maintained and efflux of K+ into the extracellular space and influx of Na+, Ca2+, Cl-, and HCO3- into the intracellular space occurs.
The length of time that ion gradients are maintained during ischemia increases with the level of glycemia,2 3 probably because of the larger amounts of glucose available for anaerobic ATP production. The level of glycemia also affects the extent of brain injury, which is increased by hyperglycemia in models of focal4 5 6 7 8 9 10 and global11 12 13 14 15 ischemia and decreased by moderate hypoglycemia.5 16 17 18 19 Increased ischemic damage is therefore not due to events triggered by faster failure of ion gradients during short-duration ischemia. However, the extent of injury may be related to the recovery of ion homeostasis in the reperfusion period. No differences are observed in the reperfusion recovery of extracellular K+, Ca2+, and H+ concentrations after ischemia in normoglycemic and hyperglycemic animals, although electroencephalographic recovery is slower after hyperglycemic ischemia.3 Recovery of intracellular ion concentrations during reperfusion in relation to blood glucose remains unstudied.
NMR relaxation of 23Na nuclei (nuclear spin 3/2) in biological environments occurs in a biexponential fashion through the interaction of these nuclei with charged polyelectrolytes (proteins); the application of a DQ filter allows the observation of Na+ in environments that contain a high concentration of polyelectrolytes.20 Because such high concentrations of proteins occur in the intracellular compartment, the observed 23Na DQ NMR signal of tissue arises mainly from intracellular sources, and a growing body of literature suggests that changes in the intensity of multiple-quantum 23Na NMR signals parallel changes in the intracellular Na+ concentration.21 22 23 24 25 26 27 In the present study, the influence of the level of glycemia on changes in the SQ and DQ 23Na NMR spectra during and after transient forebrain ischemia were evaluated.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All animals were cared for in accordance with the guidelines of the Canadian Council on Animal Care. Male Sprague-Dawley rats weighing 250 to 320 g were fasted for 24 hours before use in all experiments. Each rat was treated with atropine (0.5 mg/kg), anesthetized with sodium pentobarbital (60 mg/kg), intubated, and mechanically ventilated throughout the experiment. Brain temperature was monitored throughout the experiment with a tympanic membrane thermocouple probe28 (Cole-Parmer) and maintained at 38.2±0.2°C. The tail artery was exposed and catheterized to allow samples of blood to be taken for blood gas, osmolarity, and blood glucose measurement and for blood pressure recording throughout the experiment. To allow administration of glucose and insulin solutions, the femoral vein was catheterized. The carotid arteries were exposed through a midline neck incision, and snares were placed around them in preparation for occlusion during ischemia. The scalp and temporalis muscles were removed (bleeding being prevented by electrocauterization), and a surface coil for the reception of NMR signals was positioned on the exposed cranium.
Each rat was placed into an animal holder, the femoral catheter was connected to an infusion pump (Sage syringed pump, model 355, Cole-Parmer), and blood glucose was altered to the desired level (see below). The animal holder was placed into the magnet, and preischemia 23Na NMR spectra were obtained as described below.
After the acquisition of the preischemia spectra, blood pressure was lowered to a mean of 45 mm Hg through aspiration of blood from the tail catheter into a heparinized syringe, at which point the carotid arteries were occluded. Acquisition of 31P or 23Na NMR spectra was begun at the time of occlusion. After 10 minutes of ischemia, cerebral perfusion was restored, and the aspirated blood was reinfused. In normoglycemic animals, forebrain ischemia of this duration results (after a 7-day interval) in extensive neuronal necrosis in the CA1 sector of the hippocampus and the dorsal-lateral striatum, and in a mild to moderate degree of necrosis among pyramidal neurons in midneocortical layers.29
Blood glucose was monitored with a blood glucose meter (One Touch,
Lifescan Canada Ltd), with measurements based on the glucose oxidase
method for whole blood.30 Before ischemia, blood
glucose was adjusted to one of three groups: hypo (blood glucose 
2
mmol/L, n=7); normo (blood glucose 6 mmol/L, n=7); or hyper (blood
glucose 20 mmol/L, n=7). Blood glucose levels were altered by
administration of boluses and infusions (titrated against glucose
measurements taken every 10 minutes) as described below. The hypo and
normo groups were made iso-osmolar with the hyper group by
including 25% wt/vol mannitol in the solutions. A fourth group (blood
glucose 7 mmol/L, n=7), used as a control to enable the observation
of possible effects due to changes in osmolarity, was kept
normoglycemic without the addition of mannitol. Animals were stabilized
at their preischemic blood glucose levels for 20 minutes
before ischemia.
In the hypo group, hypoglycemia was induced with a bolus of 0.20 mL·100 g-1 of 7 IU/mL insulin and 25% wt/vol mannitol/saline (delivered over a 30-second interval to minimize physiological effects, eg, to the blood-brain barrier) followed by an infusion of 0.5% wt/vol glucose and 25% wt/vol mannitol/saline at a rate of 0.98±0.15 mL·100 g-1·h-1. Rats in the normo group were given a 0.20-mL·100 g-1 bolus of 5% wt/vol glucose and 25% wt/vol mannitol/saline, followed by infusion of the same solution (mean rate, 1.10±0.07 mL·100 g-1·h-1, titrated to a blood glucose concentration of 6 mmol/L). Hyperglycemia in the hyper group was produced by giving each animal a 0.20-mL·100 g-1 bolus of 65% wt/vol glucose/saline followed by infusion of the same solution (mean rate, 1.00±0.28 mL·100 g-1·h-1, titrated to a blood glucose concentration of 20 mmol/L). The control group was given a 0.20-mL·100 g-1 bolus and infusion (mean rate, 1.16±0.18 mL·100 g-1·h-1, titrated to a blood glucose concentration of 6 mmol/L) of 5% wt/vol glucose/saline with no mannitol for adjustment of osmolarity.
To determine the dependence of the intensity of the 23Na DQ NMR signal on pH, samples of homogenized brain tissue (n=5) were prepared and titrated with small amounts of 1 mol/L NaOH/deuterated saline or 1 mol/L HCl/deuterated saline over the range of pH 7.3 to pH 5.5 at 38°C. The titration was started at pH 5.5 for three samples and at pH 7.3 for two samples so that any systematic changes in samples over time could be detected. At each pH sampled, 31P and SQ and DQ 23Na NMR spectra were obtained.
All in vitro 79.39-MHz 23Na NMR experiments were performed
at 38.0°C with a Bruker AM300 NMR spectrometer (Bruker Analytische
Messtechnik GMBH) with a commercial multinuclear broadband probe.
Shimming was performed using the 1H NMR signal from water.
Each in vitro SQ 23Na NMR spectrum consisted of the sum of
64 transients (1024 data points) with spectral width of 5000 Hz,
acquisition time of 0.050 s, and relaxation delay of 0.085 s. The
spectra were collected using CYCLOPS phase cycling. Both in vitro and
in vivo (see below) DQ-filtered 23Na NMR spectra were
obtained using the sequence31
90°
–
/2–180°±90°–/2–90°+90°–
–90°–acquire,
where is the axis along which the excitation field B1
is applied in the rotating frame of reference, is the echo delay
time, and is the DQ evolution time. The DQ 23Na NMR
signal was separated from SQ and triple-quantum signals also
generated by this pulse sequence, using a 32-step phase cycling
procedure.32 No significant changes in the signal
intensity of the 23Na SQ resonance occurred in any sample
throughout the experiment, so changes in intensity of the
23Na DQ resonance directly reflected the effect of changes
in pH. Spectral acquisition parameters for the in vitro DQ
23Na NMR spectra were similar to those used for the in
vitro SQ 23Na NMR spectra, with selected to be 8
milliseconds (the optimized value of used in the in vivo studies as
described below).
Sample pH was determined from the difference in chemical shift of inorganic phosphate and phenylphosphonate (added to the NMR samples to serve as an internal reference) from 121.3-MHz 31P NMR spectra, and this value was compared with the corresponding pH from a previously prepared calibration curve (data not shown). Spectra were obtained for 8 to 15 pH values for each sample.
A linear fit to the plot of 23Na DQ peak height against pH yielded the change in the intensity of the 23Na DQ resonance per pH unit, defined as fpH. The value of fpH was used to correct the in vivo DQ 23Na NMR signal intensity for the effects of pH determined as outlined below during and after ischemia.
Using published in vivo 31P NMR data,33 we
fitted plots of tissue pH against time during ischemia for each
blood glucose group to the data with an equation of the
form34
 | (1) |
where the superscript i represents
ischemia, the subscripts 0 and 
indicate the values of pH at
the start of ischemia and at infinity, respectively, and
kpHi is the apparent
first-order rate constant for the exponential decrease in pH during
ischemia. The data obtained after ischemia were fit to
the equation
 | (2) |
where the superscript p represents after
ischemia, pH0 and pH are the values
of pH at the beginning of reperfusion and at an infinite time of
reperfusion, and kpHp is
the apparent first-order rate constant for the recovery of pH
during reperfusion. Thus, at any time point the change in pH, 
pH,
can be calculated and used to correct the DQ signal intensity for the
effects of pH changes.
Interleaved in vivo SQ and DQ 23Na NMR spectra were obtained at 79.455 MHz with a Bruker Biospec 7/21 spectrometer (Bruker Analytische Messtechnik GMBH). For transmission, a 5-cm diameter saddle coil tuned to both the 1H (300.13 MHz) and 23Na resonance frequencies was used, while reception was accomplished through the use of a 1.8x2.3-cm elliptical surface coil tuned to the 23Na frequency and orthogonalized with respect to the saddle coil. We have shown previously that a correctly positioned surface coil of this size effectively samples the forebrain with minimal detection of signals from adjacent muscle.29
The SQ 23Na NMR spectra were obtained using a one-pulse sequence, each consisting of the sum of 24 transients (512 data points) collected with CYCLOPS phase cycling, with preacquisition delay of 100 µs, relaxation delay of 0.085 s, spectral width of 2500 Hz, and acquisition time of 0.1024 s.
Each DQ 23Na NMR spectrum (512 data points) consisted of
128 scans and acquisition parameters identical to those for
the SQ 23Na NMR spectra, and with the evolution time
set to 25 µs and the echo delay time set to 8 ms. The DQ
23Na NMR signal was found to be at a maximum for this value
of . The 90° pulse was 280 µs. Acquisition of an SQ and DQ
23Na NMR spectral pair required 30 s; for each rat, 10
pairs of interleaved SQ and DQ 23Na NMR spectra were
obtained before ischemia, 20 pairs during ischemia, and
120 pairs after ischemia.
A 5-Hz line-broadening function was applied to the SQ 23Na NMR free-induction decays after zero-filling to 2048 data points. The free-induction decays were Fourier transformed, and the resulting spectra were phased. Baseline distortion in each spectrum was removed using a deconvolution baseline correction procedure. Peak integrals were measured relative to the mean preischemia level.
Each DQ 23Na NMR free-induction decay was multiplied
with a 25-Hz line-broadening function to optimize
signal-to-noise ratio after zero-filling to 2048 data
points. After Fourier transformation of each free-induction decay,
the resulting spectrum was phased, and the peak height relative to the
mean preischemia peak height was used in the
analysis. Signal-to-noise ratio was insufficient to
apply the methods of line-shape simulation to the DQ
23Na spectra. Any changes in the relaxation
parameters of the DQ resonance would be negligible to the
line-broadening function, and measurement of peak height is
affected less than integrated intensity by errors due to phase and
baseline correction.35 The raw DQ 23Na NMR
peak heights were then multiplied by
1/(1-fpH·pH), where fpH is the
change in the 23Na DQ NMR peak height per pH unit (above)
and pH is the difference in pH relative to the
preischemia pH at each time point during ischemia
and reperfusion, to obtain the pH-corrected 23Na DQ NMR
peak heights.
Equation 3
was used to fit the pH-corrected 23Na DQ peak
height data obtained during ischemia:
 | (3) |
where ti is the time from the start of
ischemia, Ii and
Ii
are the DQ 23Na NMR signal peak heights at
ti=0 and ti=,
respectively, and kDQi is
the apparent first-order rate constant for the increase in the peak
heights during ischemia. Because a delay of approximately 30 s
was observed before an increase in the DQ 23Na NMR peak
height occurred in the normo and hyper groups,
ti=0 was set at 30 s of ischemia for
these groups.
The DQ peak height fell to below the preischemia value
early in reperfusion in the hypo and normo groups, with subsequent
recovery. Therefore, to fit the data obtained during reperfusion, a
third term was added to give an equation of the form
 | (4) |
where tp is the time from the
start of reperfusion, Ip is the DQ
23Na NMR peak height relative to the
preischemia peak height at time
tp,
I0p
and Ip are the
peak heights at tp=0 and
tp=, respectively,
Irecp represents
the extent of the undershoot in the DQ 23Na NMR peak height
early in reperfusion,
kDQp is the apparent rate
constant for the initial fall in the DQ peak height, and
kDQrec is the apparent rate
constant for the recovery from the undershoot.
Physiological variables (levels of blood gases, osmolarity, hematocrit, blood glucose, and blood pressure) were compared among groups using ANOVA, with post hoc comparisons made using Scheffé's method of intergroup comparison (level of significance, P<.02).
The temporal curves of SQ and DQ 23Na NMR intensity data from the control, hypo, normo, and hyper groups were compared using repeated measures least-squares analysis (level of significance, P=.05). For those curves showing significant interaction and P>.02, data points were compared (with Bonferroni correction) at each time point. Parameter estimates from Equations 3 and 4 were obtained using a Levenberg-Marquardt nonlinear least-squares minimization algorithm. The level of statistical significance corresponded to the probability value giving largest confidence limits without overlap of the intervals for parameters being compared.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1
gives the physiological
data measured during this study. The difference in
preischemia plasma osmolarity between the control and hyper
groups was on the verge of significance (P=.089) and was
significant at 20 (P=.0061) and 60 minutes
(P=.0056) of reperfusion. No significant differences in
preischemic or postischemic hematocrit levels,
blood pH, PCO2,
PO2, or
HCO3- were observed.
|
Fig 1 shows the SQ 23Na NMR integrated
signal intensity for the hypo, normo, and hyper groups during and after
transient forebrain ischemia. No significant differences were
observed between the control and normo groups either during
ischemia (P=.53) or reperfusion (P=.73),
so the control group is not shown. During ischemia, the
apparent decrease in the SQ 23Na NMR intensity was not
statistically significant. After reperfusion, the signal intensity of
the SQ 23Na NMR resonance in the normo and hyper groups
increased to 107% of preischemia signal intensity at the
end of the 1-hour reperfusion period. The hypo group showed a
significantly greater increase in signal intensity than the normo and
hyper groups (117% relative to preischemia;
P=.0060 compared with the hyper group, P<.05
from 20 to 60 minutes of reperfusion compared with the normo
group).
|
The in vitro experiments using homogenized brain showed that the peak height of the 23Na DQ NMR resonance decreased 23±3% per unit decrease in pH over the range studied. Thus, the intensity of the observed 23Na DQ NMR signal during ischemia was significantly altered, and to different extents in different blood glucose groups, because pH fell below 5.8 in the hyper and to only 6.7 in the hypo group in this model after 10 minutes of ischemia.33
The time-series raw (pH-uncorrected) DQ 23Na NMR data
for the hypo, normo, and hyper groups are given in Fig 2; the pH-corrected results are shown in Fig 3. The data curves displayed marked differences among
blood glucose groups (P<.001). Data for the control group
are not included because 23Na DQ NMR spectra from the normo
and control groups showed no differences, either during or after
ischemia. Any physiological perturbations
that might be introduced by the use of mannitol to maintain uniform
osmolarity among the groups were apparently insufficient over the
course of the 90-minute experiment to affect the NMR spectra.
|
|
The apparent first-order rate constant
kDQi was
significantly different among groups and graded with respect to the
level of glycemia (Table 2), being greatest in the hypo
group (P<.002 compared with normoglycemia and
P<.0005 compared with hyperglycemia) and least in the hyper
group (P<.03 compared with normoglycemia). In the hypo
group, the DQ signal intensity rose within the first 30 s of
ischemia, reaching 200% of the preischemic level
by 120 s and slowly increasing to 225% by the end of 10 minutes of
ischemia. In both the normo and hyper groups during
ischemia, a delay of about 30 s was observed before the DQ
signal intensity increased, reaching 210% of preischemic
level at 10 minutes of ischemia. The parameter
Ii shows that the DQ
23Na NMR signal intensity for all groups would reach about
220% of preischemia at long ischemia times.
|
After ischemia, the DQ 23Na NMR signal intensity in
the hypo and normo groups decreased to 85% of the
preischemic level after about 7 minutes and recovered to
about 95% of preischemic level by the end of the
experiment. The total recovery, given by
Ip+Irecp,
for the hypo and normo groups was not significantly different from 1.0,
indicating that complete recovery would occur at a time longer than
that studied. Recovery of the DQ 23Na NMR signal was rapid
during early reperfusion in the hypo and normo groups, the apparent
first-order rate constant for recovery being greater in the normo
group (kDQp=0.385
min-1 for the hypo group,
kDQp=0.464
min-1 for the normo group;
P=.15). In marked contrast to the lower blood glucose
groups, recovery of the DQ signal intensity in the hyper group was much
slower (kDQp=0.108
min-1; P<.001 versus hypo and
normo groups). The peak height remained slightly elevated at the end of
1 hour of reperfusion, and the results of the curve fit in Table 2
indicate that recovery would not occur at longer times. Neither the
values of Irecp nor
kDQrec were significantly
different among the three groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The increase in Na+t concentration after 10 minutes of forebrain ischemia observed in the present study agrees well with increases found using a four-vessel occlusion model of global ischemia of 15-minute duration36 and using a temporary middle cerebral arterial occlusion model of 1-hour duration.37 The increase in Na+t after ischemia may be due to increased leakage across the blood-brain barrier due to stimulation of Na+/K+ ATPase in the endothelial cells of microvessels,38 thereby transferring Na+ (and water) from blood to the extracellular compartment. It has been suggested that free fatty acids and free radicals generated during ischemia may contribute to such a stimulation of Na+/K+ ATPase.39 Glucose, on the other hand, inhibits endothelial Na+/K+ ATPase,40 41 so postischemic stimulation may be more pronounced in hypoglycemic rats, leading to a greater increase in Na+t in these animals. This may explain the greater increase in the SQ 23Na NMR signal intensity after ischemia observed in hypoglycemic rats in this study.
The 23Na nucleus has a nuclear spin quantum number of 3/2.
Under conditions where the motional correlation time is of the order of
the inverse of the Larmor frequency, biexponential relaxation can
result.20 Restricted motion satisfying this condition can
be caused by interaction of Na+ with charged macromolecules
(polyelectrolytes) such as proteins. Biexponential relaxation of
quadrupolar nuclei then allows the detection of multiple-quantum
coherence signals in tissue.42 There is evidence that in
perfused organs changes in the multiple-quantum 23Na
NMR signal correspond to changes in intracellular sodium concentration.
Thus, the addition of acetylcholine in the presence of ouabain to
perfused rat salivary gland causes a large increase in intracellular
Na+, with a concurrent increase in the DQ
23Na NMR signal.26 In perfused kidney, both
anoxia and ischemia produce large increases in the DQ
23Na NMR signal,22 and addition of ouabain to
the perfusate produces large increases in the
triple-quantum 23Na NMR signal.23
Ischemia in the perfused heart produces large increases in
intracellular Na+ as measured by SQ 23Na NMR
spectroscopy in the presence of a shift reagent, which separates the
NMR signals from intracellular and extracellular
Na+, with a concurrent increase in the DQ
23Na NMR signal.25 Indeed, at short DQ
preparation times (
4 ms), in perfused liver the DQ
23Na NMR signal is due almost exclusively to intracellular
sodium.27 There is also evidence that changes in DQ
23Na NMR signal reflect changes in intracellular sodium
concentration in brain. Following the addition of veratridine (a
Na+-channel agonist) to brain slices, a large
increase in the DQ 23Na NMR signal intensity occurs due to
increased intracellular Na+ following the opening of
Na+ channels.24 In the presence of
tetrodotoxin, veratridine has no effect on the DQ 23Na NMR
signal.24 At death, a large increase in DQ
23Na NMR signal and a decrease in the DQ 39K
NMR signal are observed,21 consistent with the
movement of Na+ into and K+ out of the
intracellular space.
Changes in the DQ 23Na NMR signal due to ischemia
and reperfusion then reflect movement of Na+ between the
extracellular and the intracellular compartments. The values in Table 2
for the apparent first-order rate constants
kDQi and
kDQp demonstrate that there
is a clear dependence in the rates of movement of Na+ on
the level of glycemia. During ischemia,
kDQi is smallest in
hyperglycemic and largest in hypoglycemic animals, and in addition the
onset of the increase in the DQ 23Na NMR signal is delayed
in the higher blood glucose groups. These observations indicate that
there is a slower and delayed increase in the intracellular
Na+ concentration during ischemia in the setting of
elevated blood glucose levels. The concentrations of the
high-energy phosphates PCr and ATP similarly decrease at rates that
decrease with increasing blood glucose level.33 These
observations are consistent with prolonged maintenance
of Na+ homeostasis due to ongoing
Na+/K+-ATPase activity, supported by ATP
produced through anaerobic glycolysis of elevated brain
glucose stores, during ischemia in rats with higher blood
glucose levels. A similar glucose-dependent delay in the increase
of extracellular K+ during ischemia has been
observed using ion-selective microelectrodes.2 3
With reestablished blood flow, the DQ 23Na NMR signal rapidly decreases in intensity in the hypo and normo groups (kDQp=0.385 and 0.464 min-1, respectively), consistent with rapid movement of Na+ out of the intracellular space. However, under hyperglycemic conditions, recovery is significantly slower (kDQp=0.108 min-1). This parallels the much slower recovery of ATP during reperfusion in hyperglycemic animals,33 43 so the slower recovery of Na+ homeostasis may be due in part to prolonged suppression of Na+/K+-ATPase activity that follows hyperglycemic ischemia. However, in addition there is a large increase in the intracellular H+ concentration during ischemia in hyperglycemic animals10 33 43 44 and slower recovery of normal pH after reperfusion.33 The elevated intracellular H+ levels may contribute to the slow recovery of intracellular Na+ by stimulation of the Na+/H+ antiport,45 countering the removal of Na+ from the intracellular space by Na+/K+-ATPase.
The Ii values for the
three blood glucose groups are similar, consistent with
intracellular Na+ reaching a similar concentration during
ischemia in all groups. The maximum change in the DQ signal
attained during ischemia is similar to that obtained in hypoxic
brain slices,24 and the 220% increase
(Ii in Table 2) agrees
well with a 205% increase in the intracellular concentration of
Na+ measured using microelectrodes in rat cortex during
global ischemia.46
After reperfusion, the DQ 23Na NMR signals in the hypo and
normo groups fall below the preischemic level before
complete recovery, as given by the value of
Ip in Table 2. The significance
of the Na+ undershoot in the hypo and normo groups early in
reperfusion is not clear, but similar undershoots involving
extracellular K+ concentrations have been observed after
ischemia47 48 and neuronal
excitation49 and have been attributed to enhanced
postischemic activity of
Na+/K+ ATPase50 51 in these
cases. After reperfusion in hyperglycemic animals, however, the DQ
Na23 NMR signal returns much more slowly to the
preischemic level with no undershoot, consistent
with the argument above that delayed recovery of ATP levels may impede
the resumption of Na+/K+ ATPase activity
in these rats. This slower recovery of intracellular Na+
following ischemia may be an important factor contributing to
the accentuated ischemia brain injury in the presence of
elevated blood glucose levels. For instance, Na+ carries
water as it is transferred across cell membranes, so the slower
removal of intracellular Na+ may contribute to the observed
increase in postischemic cytosolic edema in hyperglycemic
animals.52 53 54 Furthermore, slower recovery of
Na+ homeostasis may impede the function of the
Na+/Ca2+ exchanger,55 a
major transporter of Ca2+ from the intracellular space, and
contribute to the observed slower recovery of Ca2+ after
ischemia in hyperglycemic animals.56
Changes in the SQ 23Na NMR signal intensity have been used to follow Na+ movement between the intracellular and extracellular spaces during generalized seizure in cats57 and global cerebral ischemia in dogs.58 The present study shows that, whereas small changes in SQ 23Na NMR signal intensity occur during ischemia, the DQ 23Na NMR signal intensity is a much more sensitive monitor of changes in intracellular Na+ concentration during and after ischemia.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This study was supported by the National Research Council of Canada and the Heart and Stroke Foundation of Canada. The authors thank Maureen Donnelly, Fang Wei Yang, and Yan Fan for their assistance in animal preparation.
|
Footnotes |
|---|
Reprint requests to Dr James Peeling, Department of Pharmacology and Therapeutics, University of Manitoba, 770 Bannatyne Ave, Winnipeg, Manitoba, Canada R3E 0W3. E-mail jim@bionmr.mrrl.umanitoba.ca.
Received September 28, 1995; revision received January 30, 1996; accepted January 30, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Astrup J, Moller-Sorensen P, Rahbek-Sorensen H. Oxygen and glucose consumption related to Na+-K+ transport in canine brain. Stroke. 1981;12:726-730.
2. Hansen AJ. The extracellular potassium concentration in brain cortex following ischemia in hypo- and hyperglycemic rats. Acta Physiol Scand. 1978;102:324-329. [Medline] [Order article via Infotrieve]
3.
Siemkowicz E, Hansen AJ. Brain extracellular
ion composition and EEG activity following 10 minutes ischemia
in normo- and hyperglycemic rats. Stroke. 1981;12:236-240.
4. Nedergaard M. Transient focal ischemia in hyperglycemic rats is associated with increased cerebral infarction. Brain Res. 1987;408:79-85. [Medline] [Order article via Infotrieve]
5. Nedergaard M, Diemer NH. Focal ischemia of the rat brain, with special reference to the influence of plasma glucose concentration. Acta Neuropathol. 1987;73:131-137. [Medline] [Order article via Infotrieve]
6.
de Courten-Myers GM, Myers RE, Schoolfield L.
Hyperglycemia enlarges infarct size in cerebrovascular occlusion in
cats. Stroke. 1988;19:623-630.
7. Prado R, Ginsberg MD, Dietrich WD, Watson BD, Busto R. Hyperglycemia increases infarct size in collaterally perfused but not end-arterial vascular territories. J Cereb Blood Flow Metab. 1988;8:186-192. [Medline] [Order article via Infotrieve]
8.
de Courten-Myers GM, Kleinholz M, Wagner KR, Myers
RE. Fatal strokes in hyperglycemic cats.
Stroke. 1989;20:1707-1715.
9.
Yip PK, He YY, Hsu CY, Garg N, Marangos P, Hogan
EL. Effect of plasma glucose on infarct size in focal cerebral
ischemia-reperfusion. Neurology. 1991;41:899-905.
10. Wagner KR, Kleinholz M, de Courten-Myers GM, Myers RE. Hyperglycemic versus normoglycemic stroke: topography of brain metabolites, intracellular pH, and infarct size. J Cereb Blood Flow Metab. 1992;12:213-222. [Medline] [Order article via Infotrieve]
11.
Myers RE, Yamaguchi S. Nervous system effects of
cardiac arrest in monkeys. Arch Neurol. 1977;34:65-74.
12. Siemkowicz E, Hansen AJ. Clinical restitution following cerebral ischemia in hypo-, normo- and hyperglycemic rats. Acta Neurol Scand. 1978;58:1-8. [Medline] [Order article via Infotrieve]
13.
Pulsinelli WA, Waldman S, Rawlinson D, Plum F.
Moderate hyperglycemia augments ischemic damage: a
neuropathological study in the rat. Neurology. 1982;32:1239-1246.
14. Siemkowicz E. The effect of glucose upon restitution after transient cerebral ischemia: a summary. Acta Neurol Scand. 1985;71:417-427. [Medline] [Order article via Infotrieve]
15. Smith M, Kalimo H, Warner DS, Siesjö BK. Glucose pretreatment preceding forebrain ischemia causes substantia nigra damage. J Cereb Blood Flow Metab. 1987;7(suppl 1):S76.
16.
LeMay DR, Gehua L, Zelenock GB, D'Alecy LG.
Insulin administration protects neurologic function in cerebral
ischemia in rats. Stroke. 1988;19:1411-1419.
17.
Voll CL, Whishaw IQ, Auer RN.
Postischemic insulin reduced spatial learning deficit
following transient forebrain ischemia in rats.
Stroke. 1989;20:646-651.
18.
Strong AJ, Fairfield JE, Monteiro E, Kirby M, Hogg AR,
Snape M, Ross-Field L. Insulin protects cognitive function in
experimental stroke. J Neurol Neurosurg
Psychiatry. 1990;53:847-853.
19.
Voll CL, Auer RN. Postischemic
seizures and necrotizing ischemic brain damage: neuroprotective
effects of postischemic diazepam and insulin.
Neurology. 1991;41:423-428.
20. Pekar J, Leigh JS Jr. Detection of biexponential relaxation in sodium-23 facilitated by double-quantum filtering. J Magn Reson. 1986;69:582-584.
21. Lyon RC, Pekar J, Moonen CTW, McLaughlin AC. Double-quantum surface-coil NMR studies of sodium and potassium in the rat brain. Magn Reson Med. 1991;18:80-92. [Medline] [Order article via Infotrieve]
22.
Dowd TL, Gupta RK. Multinuclear NMR studies of
intracellular cations in perfused hypertensive rat kidney.
J Biol Chem. 1992;267:3637-3643.
23.
Dowd TL, Gupta RK. NMR studies of the effect of
hyperglycemia on intracellular cations in rat kidney.
J Biol Chem. 1993;268:991-996.
24. Brooks KJ, Pirttlla T-RM, Kauppinen RA. Delayed increase in intracellular Na+ in cerebral cortical slices during severe hypoxia as measured by double quantum filtered 23Na+ NMR. Neuroreport. 1993;4:139-142. [Medline] [Order article via Infotrieve]
25. Hutchison RB, Huntley JJA, Zhou DJ, Ciesla DJ, Shapiro JI. Changes in double quantum filtered sodium intensity during prolonged ischemia in the isolated perfused heart. Magn Reson Med. 1993;29:391-395. [Medline] [Order article via Infotrieve]
26. Seo Y, Rooney WD, Murakami M. An analysis of intracellular 23Na relaxation using the double-quantum filtered NMR signal from the perfused rat salivary gland. Biochim Biophys Acta. 1993;1177:111-116. [Medline] [Order article via Infotrieve]
27. Lyon RC, McLaughlin AC. Double-quantum-filtered 23Na NMR study of intracellular sodium in the perfused rat liver. Biophys J. 1994;67:369-376. [Medline] [Order article via Infotrieve]
28. Sutherland GR, Lesiuk H, Hazendonk P, Peeling J, Kozlowski P, Jazinski A, Saunders JK. Magnetic resonance imaging and 31P spectroscopy study of the effect of temperature on ischemic brain injury. Can J Neurol Sci. 1992;19:317-325. [Medline] [Order article via Infotrieve]
29. Sutherland GR, Peeling J, Lesiuk HJ, Brownstone RM, Rydzy M, Saunders JK, Geiger JD. The effects of caffeine on ischemic neuronal injury as determined by magnetic resonance imaging and histopathology. Neuroscience. 1991;42:171-182. [Medline] [Order article via Infotrieve]
30. Marks V, Dawson A. Rapid stick method for determining blood glucose concentration. Br Med J. 1965;1:293-294.
31. Bax A, Freeman R, Kempsell SP. Natural abundance 13C-13C coupling observed via double quantum coherence. J Am Chem Soc. 1980;102:4849-4851.
32. Piantini U, Sorensen OW, Ernst RR. Multiple quantum filters for elucidating NMR coupling networks. J Am Chem Soc. 1982;104:6800-6801.
33. Tyson R, Peeling J, Sutherland G. Metabolic changes associated with altering blood glucose levels in short duration forebrain ischemia. Brain Res. 1993;608:288-298. [Medline] [Order article via Infotrieve]
34.
Petroff OAC, Prichard JW, Behar KL, Alger JR, den
Hollander JA, Shulman RG. Cerebral intracellular pH by
31P nuclear magnetic resonance spectroscopy.
Neurology. 1983;33:222-233.
35. Bain AD, Fahie BJ, Kozluk T, Leigh WJ. Improvements in the quantitation of NMR spectra by the use of statistical methods. Can J Chem. 1991;69:1189-1192.
36. Nakao N, Itakura T, Yokote H, Nakai K, Komai N. Effect of atrial natriuretic peptide on ischemic brain edema: changes in brain water and electrolytes. Neurosurgery. 1990;27:39-43. [Medline] [Order article via Infotrieve]
37.
Ito U, Ohno K, Nakamura R, Suganuma F, Inaba Y.
Brain edema during ischemia and after restoration of blood
flow. Stroke. 1979;10:542-547.
38. Mrsulja BB, Djuricic BM, Cvejic V, Mrsulja BK, Abe K, Spatz M. Biochemistry of experimental ischemic brain edema. Adv Neurol. 1980;28:217-230. [Medline] [Order article via Infotrieve]
39. Asano T, Shigeno T, Johshita H. A novel concept on the pathogenetic mechanism underlying ischaemic brain oedema: relevance of free radicals and eicosanoids. Acta Neurochir Suppl (Wien). 1987;41:85-96. [Medline] [Order article via Infotrieve]
40. Knudsen GM, Jakobsen J. Blood-brain barrier permeability to sodium: modification by glucose or insulin? J Neurochem. 1989;52:174-178. [Medline] [Order article via Infotrieve]
41. Yorek MA, Stefani MR, Moore SA. Acute and chronic exposure of mouse cerebral microvessel endothelial cells to increased concentration of glucose and galactose: effect on myo-inositol metabolism, PGE2 synthesis, and Na+/K+-ATPase transport activity. Metabolism. 1991;40:347-358. [Medline] [Order article via Infotrieve]
42. Pekar J, Renshaw PF, Leigh JS Jr. Selective detection of intracellular sodium by coherence-transfer NMR. J Magn Reson. 1987;72:159-161.
43. Sutherland GR, Peeling J, Sutherland E, Tyson R, Dai F, Kozlowski P, Saunders JK. Forebrain ischemia in diabetic and nondiabetic BB rats studied with 31P magnetic resonance spectroscopy. Diabetes. 1992;41:1328-1334. [Abstract]
44. Haraldseth O, Nygård Ø, Grønås T, Southon T, Gisvold SE, Unsgård G. Hyperglycemia in global cerebral ischemia and reperfusion: a 31-phosphorous NMR spectroscopy study in rats. Acta Anaesthesiol Scand. 1992;36:25-30. [Medline] [Order article via Infotrieve]
45. Jakubovicz DE, Klip A. Lactic acid-induced swelling in C6 glial cells via Na+/H+ exchange. Brain Res. 1989;485:215-224. [Medline] [Order article via Infotrieve]
46.
Hansen AJ. Effect of anoxia on ion distribution
in the brain. Physiol Rev. 1985;65:101-148.
47. Sick TJ, Rosenthal M, LaManna JC, Lutz PL. Brain potassium ion homeostasis, anoxia, and metabolic inhibition in turtles and rats. Am J Physiol. 1982;243:R281-R288.
48. Roberts EL Jr, Sick TJ. Recovery of synaptic transmission predicted from extracellular K+ undershoots following brief anoxia in hippocampal slices. Brain Res. 1987;402:178-181. [Medline] [Order article via Infotrieve]
49. Heinemann U, Lux HD. Undershoots following stimulus-induced rises in extracellular potassium concentration in the cerebral cortex of cat. Brain Res. 1975;120:231-249.
50. Krnjevic K, Morris ME. Factors determining the decay of K+ potentials and focal potentials in the central nervous system. Can J Physiol Pharmacol. 1975;53:923-934. [Medline] [Order article via Infotrieve]
51. Cordingly GE, Somjen GG. The clearing of excess potassium from the extracellular space in spinal cord and cerebral cortex. Brain Res. 1978;151:291-306. [Medline] [Order article via Infotrieve]
52.
Berger L, Hakim AM. The association of
hyperglycemia with cerebral edema in stroke. Stroke. 1986;17:865-871.
53. Jakubovicz DE, Grinstein S, Klip A. Cell swelling following recovery from acidification in C6 glioma cells: an in vitro model of postischemic brain edema. Brain Res. 1987;435:138-146. [Medline] [Order article via Infotrieve]
54. Kraig RP, Chesler M. Astrocytic acidosis in hyperglycemic and complete ischemia. J Cereb Blood Flow Metab. 1990;10:104-114. [Medline] [Order article via Infotrieve]
55. Stys PK, Waxman SG, Ransome BR. Ionic mechanisms of anoxic injury in mammalian CNS white matter: role of Na+ channels and Na+-Ca2+ exchanger. J Neurosci. 1992;12:430-439. [Abstract]
56. Araki N, Greenberg JH, Sladky JT, Uematsu D, Karp A, Reivich M. The effect of hyperglycemia on intracellular calcium in stroke. J Cereb Blood Flow Metab. 1992;12:469-476. [Medline] [Order article via Infotrieve]
57. Schnall MD, Yoshizaki K, Chance B, Leigh JS Jr. Triple nuclear NMR studies of cerebral metabolism during generalized seizure. Magn Reson Med. 1988;6:15-23. [Medline] [Order article via Infotrieve]
58.
Eleff SM, Maruki Y, Monsein LH, Traystman RJ,
Bryan N, Koehler RC. Sodium, ATP, and intracellular pH
transients during reversible complete ischemia of dog
cerebrum. Stroke. 1991;22:233-241.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||