From the Johns Hopkins Medical Institutions, Department of Anesthesiology
and Critical Care Medicine, Baltimore, Md.
Correspondence to Dr Patricia D. Hurn, Department of Anesthesiology/Critical Care, Blalock 1404, Johns Hopkins Hospital, 600 N Wolfe St, Baltimore. MD 21287-4963. E-mail phurn{at}welchlink.welch.jhu.edu
Methods—Dogs anesthetized with pentobarbital-fentanyl
were treated with 30 minutes of global incomplete cerebral
ischemia produced by intracranial pressure elevation. Cerebral
energy metabolites (ATP, phosphocreatine) and intracellular pH
(pHi) were measured by 31P magnetic resonance
spectroscopy. Preischemic plasma glucose level was
manipulated to titrate end-ischemic pHi. After
ischemia, brains were perfused with cold phosphate-buffered
saline solution; then 16 different brain areas were sampled, filtered
to separate the LMW fraction (<30 000 D), and assayed by rapid
colorimetric assay for tissue iron. Total iron, LMW
iron, and protein in each sample were measured in sham-operated (no
ischemia, n=8), normoglycemic ischemia (ISCH [glucose
7±4 mmol/L], n=7), and hyperglycemic (GLU-ISCH [glucose
31±3 mmol/L], n=9) groups.
Results—High-energy phosphates fell to near zero values in both
ISCH and GLU-ISCH groups by 30 minutes but remained unchanged in the
sham-operated group. As expected, pHi decreased during
ischemia but to a greater extent in GLU-ISCH (6.20±0.05 in
ISCH, 6.08±0.04 in GLU-ISCH, P<.05). Iron could be
detected in all areas of the brain in sham-operated animals, with the
highest amounts obtained from subcortical areas such as the
hippocampus, pons, midbrain, and medulla. Total iron was higher in ISCH
relative to sham-operated animals and higher in cortex and pons
relative to GLU-ISCH. Regional LMW (as a percentage of total iron;
LMW/total iron) was elevated in numerous brain areas in ISCH, including
cortical gray matter, cerebellum, hippocampus, caudate, and midbrain.
LMW/total iron was higher in GLU-ISCH versus ISCH in cortical gray
matter only. In other brain areas, ischemic LMW/total iron was
equivalent in glucose-treated or normoglycemic animals (white matter,
thalamus, pons, medulla) or lower in the glucose-treated group
(cerebellum, hippocampus, caudate, midbrain).
Conclusions—These data demonstrate that levels of total and LMW
iron increase with global cerebral ischemia in the majority of
cortical and subcortical regions of normoglycemic brain. However,
exacerbation of ischemic acidosis via glucose administration
does not increase tissue iron and produces a greater increase in the
LMW fraction in cortical gray matter only. In other brain regions,
total and LMW iron availability is similar to that of
nonischemic animals.
We have previously shown that reducing intracellular pH
(pHi) to <6.0 and bicarbonate ion to <1 to
2 mmol/L during hyperglycemic cerebral ischemia produces a
marked secondary deterioration of brain ATP and cerebral blood flow
during reperfusion.9 This rapid and profound
secondary injury can be ameliorated with antioxidant treatments,
including the iron chelator and superoxide scavenger,
deferoxamine.10 11 Therefore, we
wished to further examine the role of iron in ischemic injury
and quantify the increase in nonprotein bound iron that ordinarily
occurs during global cerebral ischemia in vivo. The goal of the
present study was to evaluate regional distribution of cerebral
iron during ischemic conditions compared with the
nonischemic brain. Furthermore, we tested the hypothesis that
reduction of tissue pH during hyperglycemic ischemia amplifies
free iron, some component of which may participate in oxidant-mediated
brain injury.
Arterial and sagittal sinus blood samples were
analyzed for PO2,
P[scap]co2, and pH levels with a
Radiometer ABL electrode system. Oxygen content was measured with a
CO-Oximeter (No. 282, Instrumentation Laboratories).
Arterial blood pressure and intracranial pressure were
measured with Statham transducers. Serum glucose levels were measured
on a YSI model 23A Glucose Analyzer.
Using a Vivospec spectrometer (Otsuka Electronics with a 1.89-T
horizontal superconducting magnet, 25-cm bore; Oxford
Instruments),31 P MRS spectra were obtained
as previously described.9 10 12 Spectral
areas were analyzed via planimetry for B-ATP and
phosphocreatine and expressed as a percentage of the respective
baseline areas. pHi was determined by
methods previously described by Petroff et
al,13 using constants derived from our
titration data:
pHi=6.73+log[(a-3.07)/5.68-a]
where a is the chemical shift of inorganic phosphate
relative to phosphocreatine. An external standard
(dimethyl[2-oxopropyl]-phosphonate) placed over the coil served as a
marker for spectral position when the phosphocreatine peak disappeared.
The Henderson-Hasselbalch equation was used to calculate intracellular
[HCO3-] using a
pKa of 6.12, pHi as
measured by MRS, and sagittal sinus
PCO2 with a solubility coefficient of
0.0314 mmol/L per mm Hg. Differences in sagittal sinus
P[scap]co2 were assumed to reflect
changes in tissue P[scap]co2.
The dogs were divided into three treatment groups: sham-operated
without ischemia, normoglycemic (ISCH), or hyperglycemic
ischemia (GLU-ISCH). In GLU-ISCH, plasma glucose was raised to
Brain samples were evaluated for levels of LMW and total iron by the
rapid colorimetric method as described by
Fish.14 All assay solutions were made from
double-chelated water to minimize iron contaminants (Chelex-100,
BioRad). All chemicals were purchased from Sigma Chemical Co. Samples
were weighed and then homogenized in 1-mmol/L EDTA to a
final concentration of 40 mg tissue/mL 1-mmol/L EDTA. An aliquot was
obtained for subsequent analysis of total iron and protein. The
LMW aliquot was centrifuged in a three-step process: (1) at
3000 rpm for 20 minutes, (2) at 19 000 rpm for 40 minutes, and (3) the
supernatant was then spun at 6000 rpm for 60 minutes through a 30 000
molecular weight nylon filter collection system (Vanex VG8504–00,
Vangard International) to separate the LMW fraction. Total and LMW iron
samples were assayed by incubating 400-µL aliquots with 200 µL 1.2
mol/L hydrochloric acid–0.285 mol/L potassium permanganate solution
for 2 hours at 60°C to release tissue iron. Next, 40 µL of 2 mol/L
ascorbic acid–5 mol/L ammonium acetate–6.5 mmol/L
ferrozine–13.1 mmol/L neocurpine solution was added to chelate
ferric iron species and other potentially interfering metals. To remove
protein precipitates, total iron samples were centrifuged for 1
minute on an Allied-Fisher Scientific Micro-Centrifuge model
235C. A standard curve was generated for the range of 0.01 to 2.0
µg/mL, and all samples were measured at 562 nm with a Milton Roy
Spectronic 601 spectrophotometer. Total protein levels were determined
using the Bradford method.
Because minimal differences were apparent in the eight cortical
regions, the data are presented as summed gray matter and
individually represented deep brain structures. Data are
normalized to total iron within the sample and to the sample wet
weight. To evaluate reproducibility and asses intra-animal variability,
a brain region from one sham-operated, one ISCH, and one GLU-ISCH
animal each was homogenized, and several samples from the
same region per animal were assayed using study methods. The
coefficient of variation for this repeated analysis was
3.6±0.01%. To these known samples, predetermined amounts of iron were
added. Measurements were accurate to within 0.06 µg iron.
Because cerebral blood flow was not measured in the present study,
severity of ischemia was assessed by the lack of any residual
phosphocreatine signal and severe depression of ATP by
end-ischemia as previously observed with this ischemic
model and duration.9 10 11 12 Therefore, any
animal with measurable phosphocreatine (>5%) by 30 minutes of
ischemia was excluded from the study. Eight animals were
excluded on this basis (5 normoglycemic and 3 hyperglycemic dogs).
The data are presented as mean±SEM, and significance was set
at P<.05 for all tests.
Physiological data were analyzed with a
repeated measures ANOVA, with treatment as the between-subject factor
and time as the within-subject factor. If group or time interactions
were significant, a Newman-Keuls test was used to distinguish
individual groups at specific time points. Neurochemical data were
analyzed using a one-way ANOVA and Newman-Keuls test.
As previously shown in this model, high-energy phosphates declined
sharply during the 30-minute period of intracranial pressure elevation
(Fig 1
Levels of total and LMW iron were determined in 16 different brain
regions. Because minimal differences were apparent in the eight
cortical regions, the data are presented as summed gray matter
(Table 2
Given the regional and intergroup variability in total iron, Table 2
The role of iron in oxidant injury as a consequence of ischemia
has been inferred from numerous studies that show benefit from
treatment with strong iron chelators or from apotransferrin. We found
previously that pretreatment with deferoxamine, but not
conjugated deferoxamine or iron-saturated feroxamine,
ameliorates the secondary metabolic deterioration and
severe hypoperfusion observed with ischemia complicated by
extreme lactic acidosis.10 However,
deferoxamine has significant radical scavenging properties
in vitro15 via hydrogen-donating free hydroxamate
groups,16 so its mechanism of action relative to
iron chelation is not entirely clear. Oxidative injury has been
ameliorated by inhibiting iron-dependent lipid peroxidation with
aminosteroids after trauma and
ischemia.11 17
Consequently, whether ischemia-induced release of substantial
amounts of reactive iron occurs in vivo is of interest. LMW ferrous
iron has been demonstrated to increase at various reperfusion time
points after cardiac arrest, with the large increases observed at 1 to
2 hours after arrest, and to be accompanied by increases in levels of
lipid peroxidation.18 19 20 Our results with
incomplete global cerebral ischemia are consistent with
these earlier findings in cortical homogenates, and we now
show that increased free iron availability is widespread in brain,
including regions of the brain that are known to be highly vulnerable
to ischemic insults. Measurable levels of LMW iron were
present in the nonischemic brain, with greatest levels of
iron in the white matter of all groups. Others have demonstrated iron
concentration in white matter, midbrain, pons, and
medulla.21 Our baseline nonischemic LMW
iron values may be overestimated because of our fractionation procedure
with serial centrifugations and because 1-mmol/L EDTA,
a common chelating agent, was used in the
homogenization procedure. However, major increases
were apparent in postischemic tissue when compared with
nonischemic samples subject to the same potential for
homogenization and fractionation artifacts. The
lower limit of resolution of the assay is 0.01 µg iron, and
differences in iron seen in our study are well above the lower limit of
the assay, supporting the validity of our findings.
A 30 000-D filter membrane was used to separate tissue
homogenates; the fraction passing though the membrane was
defined as LMW iron. This cutoff is comparable to that previously used
to examine tissue harvested from ischemic
animals.19 20 A 30 000-D filtering capacity
allowed separation of delocalized iron from the biologically inactive
form, ordinarily complexed to large proteins such as transferrin
(80 000 D) and ferritin (440 000 D).22 We were
specifically interested in fractionation at this level because of the
pH-dependent binding properties for transferrin23
and possibly ferritin, the latter of which is a major source of total
tissue iron. It seemed likely that pH-associated delocalization of iron
would be through release and reuptake of iron from carrier/storage
proteins. Nevertheless, our measurements do not determine the precise
chemical identity or catalytic activity of the LMW iron; small
iron-complexed moieties and heme-containing proteins less than 30 000
D are likely present as LMW species and potential nonparticipants
in iron-catalyzed oxidative injury.
The mechanisms of iron release during ischemia are likely
complex, potentially involving both intracellular stores and
intravascular iron transport and delivery to brain. Because of the
potential cytotoxicity of iron, physiological
systems have developed intricate mechanisms to protect cells from
oxidative injury. Brain iron is largely complexed to protein carriers
(transferrin, lactoferrin) for intravascular and transmembrane
transport and to ferritin for intracellular storage. Ischemic
acidosis may drive iron dissociation from those carrier proteins with
pH-dependent binding characteristics (eg, carbonate binding ion of
transferrin).24 Once mobilized, free iron likely
binds nonspecifically to a variety of small molecular moieties and
augments the ordinarily small LMW nonprotein-bound tissue pool. In
cortical homogenates, striking increases in LMW iron are
observed at pH 6.0 when pH is reduced from 7.0 by direct addition of
lactic acid. Furthermore, brain from decapitated hyperglycemic rats
shows elevated LMW iron relative to normoglycemic
controls.9
Surprisingly, we did not observe increased LMW iron in all brain
regions in hyperglycemic ischemia despite reduction in
pHi to values <6.0 in many glucose-treated
animals. Total iron was higher in many but not all brain regions in
ISCH; consequently, we evaluated regional LMW iron for each group as a
percentage of total iron. LMW/total iron was elevated in hyperglycemic
versus normoglycemic ischemia only in cortical gray matter. In
other brain areas, ischemic LMW/total iron was equivalent or
lower in brain of glucose-treated animals when compared with either
normoglycemic or nonischemic brain, suggesting that lower
end-ischemic pHi does not exacerbate
ischemia-induced iron availability for catalysis of oxidant
injury mechanisms, at least within the one time-point window of our
observations. We cannot exclude the possibility that lower pH in
GLU-ISCH accelerated iron dissociation, peaking during ischemia
or immediate reperfusion with subsequent early elimination of the LMW
tissue pool. In addition, the end-ischemic
pHi achieved in ISCH and GLU-ISCH spanned a
narrow range. We did not evaluate iron over a range of pH to determine
an optimum pH associated with LMW iron augmentation or whether there is
a threshold below which free iron begins to decrease because of other
pH-related factors. Nevertheless, it seems unlikely that the large
differences observed with ISCH versus GLU-ISCH are explained by these
factors. A more plausible explanation is that the total iron pool, and
consequently the LMW fraction, increases during
ischemia/reperfusion in an unanticipated manner that is
independent of pHi depression. However, with
hyperglycemic ischemia, total tissue iron and the LMW fraction
are regionally heterogenous and not uniformly higher
than in nonischemic animals.
One hypothesis is that ischemic brain iron loads are linked to
intravascular iron delivery and tissue transfer via normal or
pathophysiological mechanisms. The ischemic
insult, as determined by the near complete loss of ATP and
phosphocreatine, was different between ischemic groups in
plasma glucose level only (7 versus 31 mmol/L in ISCH and
GLU-ISCH, respectively) and the severity of end-ischemic
pHi (6.20 versus 6.08). Our previous measurements
with this model indicate that brain blood flow, as measured by
radiolabeled microspheres, is reduced uniformly throughout the
brain, with a residual flow of <10 to 12 mL/min per 100 g during
both normoglycemic and hyperglycemic
ischemia.9 10 11 12 However, we sampled tissue
after 30 minutes of ischemia and 15 minutes of reperfusion in
each animal. Postischemic hyperemia occurs in
normoglycemic animals during early
reperfusion9 10 11 12 ; hyperglycemic ischemia
results in heterogenous
reperfusion25 and significant
hypoperfusion.9 10 11 12 26 27 While it is uncertain
that extensive hypoperfusion would account for the failure to see
increased total and LMW iron in the GLU-ISCH group, de novo iron
delivery from circulating iron containing proteins could be reduced
during hyperglycemic reperfusion. Although the kinetics and mechanisms
of iron transport across the postischemic blood-brain
barrier are not known, iron transfer into cells via transferrin binding
and endosomic internalization is thought to occur within minutes in
healthy brain.6 If intravascular iron delivery to
tissue represents a significant contribution to the dynamic
pool in reperfused brain, then gross hypoperfusion would alter tissue
iron content.
An alternative hypothesis is that the low levels of iron in GLU-ISCH
are a function of the high plasma and tissue glucose levels per se.
Hyperglycemia glycates many protein moieties, including
transferrin28 and increases tissue autoxidation
products. Transferrin-bound iron is low in rat hyperglycemic sera
secondary to glycation, although the proteins remain redox-active, and
target cell binding is depressed.28 29 Therefore,
this source of iron dissociation may be inconsequential in GLU-ISCH
versus ISCH. The effect of acute hyperglycemia on intracellular storage
proteins such as ferritin is unknown; however, with intense oxidant
stress, ferritin can be transformed into a
hemosiderin-like protein,30 which
is less reactive to Fe2+-releasing
stimuli.31 32 The abnormal
ferritin-Fe2+ complex would not be captured in
our fractionation assay because of its relatively high molecular
weight. Therefore, it is feasible that hyperglycemia alters carrier
protein binding in normal or ischemic tissue and yields lower
brain iron during reperfusion. Our present method for quantifying
the total or LMW iron pool is not sufficiently sensitive to distinguish
a baseline effect of hyperglycemia in sham-operated animals,
independent of ischemia.
In conclusion, our findings strengthen the hypothesis that tissue iron
and its potentially reactive LMW fraction are elevated after global
incomplete cerebral ischemia and now demonstrate that the
regional distribution is widespread. However, when the fall in tissue
pH ordinarily observed with ischemia is exacerbated by systemic
hyperglycemia, total iron and subsequently the LMW iron species
available for catalysis of oxidant injury mechanisms are not elevated
during early reperfusion. Whether other means of altering
ischemic pH produce a similar lack of ischemia-induced
iron release remains to be tested. However, we speculate that
hyperglycemia alters ischemic tissue iron content in a
pH-independent manner, possibly by decreasing iron delivery, tissue
internalization, and binding protein function during severe but
incomplete cerebral ischemia and early reperfusion.
Received February 20, 1997;
revision received September 30, 1997;
accepted November 17, 1997.
Laboratory
of Cerebrovascular Biology and Stroke,
Department of Neurology University of Minnesota,
Minneapolis, Minnesota
The regulation of iron homeostasis in the postischemic
brain is poorly understood. Iron, a metal insoluble in water at
physiological pH, is transported in the blood
stream as Fe3+-bound transferrin (see Reference 4 for
review). The transferrin-iron complex enters the cell by endocytosis
via membrane-bound transferrin receptors. Once inside the cell, iron is
sequestered by the storage protein ferritin from which it is released
as Fe2+ (see Reference 5 for review). Expression of
critical proteins involved in iron metabolism, eg, ferritin
H and L chains, the transferrin receptor, and 5-aminolevulinate
synthetase (the enzyme that synthesizes heme), is controlled by the
iron response element–binding protein (IRE-BP).5 Thus,
the intracellular concentration of catalytically active iron is tightly
regulated by these proteins, the expression of which is controlled by
IRE-BP. The effect of cerebral ischemia and/or hyperglycemia on
this complex sequence of events remains to be elucidated. The work of
Lipscomb et al is a step in the right direction because it provides
much awaited in vivo data on the postischemic perturbation
in iron homeostasis and on the effect of hyperglycemia on iron
availability. This enlightening study constitutes a starting point for
future investigations focusing on the effect of cerebral
ischemia on the cellular and molecular factors regulating free
intracellular iron.
Received February 20, 1997;
revision received September 30, 1997;
accepted November 17, 1997.
© 1998 American Heart Association, Inc.
Original Contributions
Low Molecular Weight Iron in Cerebral Ischemic Acidosis In Vivo
Abstract
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
Background and
Purpose—Iron-catalyzed radical generation is a potentially
significant mechanism by which extensive tissue acidosis exacerbates
brain injury during ischemia/reperfusion. We hypothesized that
levels of low-molecular-weight (LMW) iron increase during in vivo
global cerebral ischemia in a pH-dependent manner, potentially
catalyzing oxidant injury. The present study quantified regional
differences in LMW iron during global cerebral incomplete
ischemia and determined whether augmenting the fall in
ischemic tissue pH with hyperglycemia also amplifies free
iron availability.
Key Words: acidosis • iron • cerebral ischemia • spectroscopy, nuclear magnetic resonance
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
Iron-catalyzed oxygen
radical production has long been hypothesized to play a
prominent role in brain injury during ischemia/reperfusion. Via
Fenton-Haber chemistry, ferrous iron (Fe2+)
interacts with hydrogen peroxide or superoxide to produce hydroxyl
radical and other free radicals that damage cell protein and DNA
structure.1 2 3 Alternatively, iron is an
effective catalyst of tyrosine nitration by another cytotoxic oxidant,
peroxynitrite.4 Furthermore, radical-mediated
mechanisms may be particularly relevant to cerebral injury when the
ischemic insult is accompanied by extensive tissue acidosis,
eg, during ischemia in hyperglycemic subjects. Rehncrona et
al5 hypothesized that lactic acidosis increases
iron dissociation from carrier proteins that use carbonate binding,
whereascarbonic acidosis (eg, from tissue CO2)
acts to stabilize bicarbonate ion and carbonate-bound iron. For example, transferrin has
been shown to release bound iron during acidotic conditions in vitro,
particularly as pH falls to <6.0.6 Cortical
homogenates incubated in acidotic media release iron as pH
falls, as measured by an increase in LMW iron
species.7 Furthermore, acidosis increases TBARS
in anoxic brain slices.8 Although iron likely
plays an important role in acidosis-mediated mechanisms of
ischemic injury, the conditions of iron release, regional
distribution of prooxidant iron, and participation in lipid
peroxidation have not been clearly outlined.
Materials and Methods
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
This study was approved by the institutional animal care and use
committee and is in compliance with the guidelines of the National
Institutes of Health for care and handling of animals. A
well-characterized model of global incomplete cerebral ischemia
in dogs was used.9 12 Male dogs (10 to 15 kg)
were anesthetized with intravenous fentanyl (50
µg/kg) and pentobarbital (6 mg/kg, followed by 3 mg/kg per hour) and
mechanically ventilated with supplemental oxygen. The dogs were
paralyzed with pancuronium bromide (0.1 to 0.2 mg/kg). Systemic
arterial and venous catheters were placed for blood
pressure monitoring, blood gas sampling, and infusion of
intravenous fluids and medications. After complete
retraction of temporalis muscles, a midline burr hole was placed near
the junction of the coronal sutures for placement of a superior
sagittal sinus. A Silastic ventricular drain catheter
(Cordis) was placed into the ventricle via a second burr hole for
intracranial pressure monitoring and infusion of artificial
cerebrospinal fluid. An epidural thermistor was placed for continuous
temperature monitoring. To maintain normothermic epidural
temperature, animals were warmed as needed with a water blanket and
head insulation. 
500 mg/dL with intravenous 50% dextrose solution
immediately before and during ischemia. Baseline measurements
of arterial and venous blood gases, intracranial pressure,
blood pressure, glucose, and MRS spectra were obtained in all groups.
To create global incomplete cerebral ischemia, warmed
cerebrospinal fluid was infused into the lateral ventricle from a
reservoir, raising intracranial pressure and lowering cerebral
perfusion pressure to a controlled level of 10 to 15 mm Hg,
yielding a low residual cerebral blood flow. The ischemic
period was 30 minutes.9 10 11 12 MRS data were
collected throughout the experimental period in one 6-minute epoch and
three 8-minute epochs. To end ischemia, the reservoir was
disconnected and intracranial pressure rapidly returned toward
baseline. The animal was removed from the magnet, and a left lateral
thoracotomy was performed with ventricular cannulation and
perfusion of 3 L of phosphate-buffered 0.9% NaCl solution to clear
blood from the cerebral vasculature. The skull cap was removed at 15
minutes of reperfusion in each animal, and 16 different brain areas
were sampled and frozen in liquid nitrogen for neurochemical
analysis.
Results
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
Physiological data are contained in Table 1
. Arterial blood gases were
controlled to maintain constant oxygenation and
arterial CO2; however,
arterial pH was depressed in both ischemic groups
relative to sham-operated animals. Epidural temperature decreased in
both ischemic groups compared with sham but in an equivalent
manner. During ischemia, arterial blood pressure
initially rose in ISCH and GLU-ISCH, then decreased to baseline levels.
Cerebral perfusion pressure was maintained at approximately 10
mm Hg, and there were no differences between the ischemic
groups. Cerebral perfusion pressure and intracranial pressure remained
unchanged over time in the sham-operated group. Plasma glucose was
elevated throughout ischemia in GLU-ISCH relative to both ISCH
and sham groups. During ischemia, hemoglobin was lower than in
the sham-operated and ISCH groups.
View this table:
[in a new window]
Table 1. Physiological Variables
Before and During 30 Minutes of Cerebral Ischemia ). At end-ischemia,
phosphocreatine and ATP levels were similar between ISCH and GLU-ISCH.
As expected, pHi decreased in both
ischemic groups but to a greater extent in GLU-ISCH (Fig 2). End-ischemic
pHi was 6.20±0.05 in ISCH and 6.08±0.04 in
GLU-ISCH (P<.05).
View larger version (13K):
[in a new window]
Figure 1. Phosphocreatine and ATP during 30 minutes of
global cerebral ischemia. Data are expressed as percentage of
baseline values (mean±SEM). Sham-operated, n=8. Hyperglycemic
ischemia (GLU-ISCH), n=9. Normoglycemic ischemia
(ISCH), n=7.
View larger version (17K):
[in a new window]
Figure 2. Intracellular pH during 30 minutes of cerebral
ischemia. Data are mean±SEM. Sham-operated, n=8. Hyperglycemic
ischemia (GLU-ISCH), n=9. Normoglycemic ischemia
(ISCH), n=7. *P<.05 compared with GLU-ISCH. ). Iron could be detected in all
areas of the brain in sham-operated animals, with the highest amounts
obtained from subcortical areas such as the hippocampus, pons,
midbrain, and medulla. Total iron was higher per gram of wet weight in
ISCH animals than in the sham group in most brain areas and higher in
ISCH than in GLU-ISCH in three brain regions. In gray matter, LMW iron
was 3.06±1.09 µg/g wet wt in ISCH compared with 0.22±0.04 in
sham-operated animals and 0.6±0.35 µg/g wet wt in GLU-ISCH.
Similarly, white matter LMW iron was higher in ISCH relative to sham
but not to GLU-ISCH (2.86±0.98, 0.36±0.25, and 0.15±0.05 µg/g wet
wt in ISCH, GLU-ISCH, and sham, respectively).
View this table:
[in a new window]
Table 2. Iron Levels During Global Incomplete
Ischemia 
presents the LMW values referenced to the total iron obtained from
the region of interest. Regional LMW (percentage of total iron;
LMW/total iron) was elevated in numerous brain areas in ISCH, including
cortical gray matter, cerebellum, hippocampus, caudate, and midbrain.
LMW/total iron was higher in GLU-ISCH versus ISCH in cortical gray
matter only. In other brain areas, ischemic LMW/total iron was
equivalent in glucose-treated or normoglycemic animals (white matter,
thalamus, pons, medulla) or lower in the glucose-treated group
(cerebellum, hippocampus, caudate, midbrain).
Discussion
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
These data demonstrate two important findings. First, measurable
levels of total and LMW iron can be assayed throughout the brain, and
these levels increase in a region-specific manner during normoglycemic
ischemia. Elevated LMW iron relative to nonischemic
animals was observed in both cortical and subcortical brain regions,
most prominently in the cerebellum, midbrain, hippocampus, and cortical
gray matter. Second, increased total or free iron was not uniformly
observed in animals treated with hyperglycemic ischemia, which
sustained the lowest end-ischemic pHi.
These findings raise fundamental questions regarding the mechanisms and
timing of iron release during ischemic acidosis.
Selected Abbreviations and Acronyms
GLU-ISCH
=
hyperglycemic ischemia group
ISCH
=
normoglycemic ischemia group
LMW
=
low molecular weight
MRS
=
magnetic resonance spectroscopy
TBARS
=
thiobarbituric acid–reactive substances
Acknowledgments
This study was supported by grants from the National Institutes
of Health (NR03521 and NS33668). We would like to thank Judy Klaus for
outstanding technical support.
References
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
Editorial Comment
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
There is extensive evidence that hyperglycemia exacerbates
cerebral ischemic injury. Whereas hyperglycemic patients have a
worse neurological outcome after ischemic stroke,
preischemic hyperglycemia aggravates ischemic brain
damage in experimental animals.1 2 The mechanisms of this
effect have not been elucidated. It has been suggested that the
worsening of the brain damage is secondary to the greater pH reduction
associated with hyperglycemic ischemia.3 More
severe acidosis would in turn increase in the availability of free iron
for formation of reactive oxygen species through the Haber-Weiss
reaction. In the accompanying article, Lipscomb et al sought to test
this hypothesis in vivo using a well-established canine model of
temporary global cerebral ischemia. As anticipated,
hyperglycemic ischemia produced a more severe reduction in
brain pHi than normoglycemic ischemia. Furthermore,
normoglycemic ischemia elevated LMW iron in most brain regions.
Unexpectedly, however, hyperglycemic ischemia enhanced this
effect only in the neocortical gray matter and not in other brain
regions. Rather, in most regions, ischemic hyperglycemia
resulted in iron concentrations similar to those observed in normal
brain. The data suggest that in this model, increased iron availability
does not play a prominent role in the exacerbation of brain damage
associated with hyperglycemic ischemia, with the possible
exception of the cerebral cortex.
Selected Abbreviations and Acronyms
GLU-ISCH
=
hyperglycemic ischemia group
ISCH
=
normoglycemic ischemia group
LMW
=
low molecular weight
MRS
=
magnetic resonance spectroscopy
TBARS
=
thiobarbituric acid–reactive substances
References
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
References
This article has been cited by other articles:
 |
|
 |
|
  A. Siddiq, I. A. Ayoub, J. C. Chavez, L. Aminova, S. Shah, J. C. LaManna, S. M. Patton, J. R. Connor, R. A. Cherny, I. Volitakis, et al. Hypoxia-inducible Factor Prolyl 4-Hydroxylase Inhibition: A TARGET FOR NEUROPROTECTION IN THE CENTRAL NERVOUS SYSTEM J. Biol. Chem., December 16, 2005; 280(50): 41732 - 41743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Demougeot, M. Van Hoecke, N. Bertrand, A. Prigent-Tessier, C. Mossiat, A. Beley, and C. Marie Cytoprotective Efficacy and Mechanisms of the Liposoluble Iron Chelator 2,2'-Dipyridyl in the Rat Photothrombotic Ischemic Stroke Model J. Pharmacol. Exp. Ther., December 1, 2004; 311(3): 1080 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Kress, K. E. Dineley, and I. J. Reynolds The Relationship between Intracellular Free Iron and Cell Injury in Cultured Neurons, Astrocytes, and Oligodendrocytes J. Neurosci., July 15, 2002; 22(14): 5848 - 5855. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. B. Becker, M. L. Weisfeldt, M. H. Weil, T. Budinger, J. Carrico, K. Kern, G. Nichol, I. Shechter, R. Traystman, C. Webb, et al. The PULSE Initiative: Scientific Priorities and Strategic Planning for Resuscitation Research and Life Saving Therapies Circulation, May 28, 2002; 105(21): 2562 - 2570. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Davalos, J. Castillo, J. Marrugat, J. M. Fernandez-Real, A. Armengou, P. Cacabelos, and R. Rama Body iron stores and early neurologic deterioration in acute cerebral infarction Neurology, April 25, 2000; 54(8): 1568 - 1574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Zaman, H. Ryu, D. Hall, K. O'Donovan, K.-I Lin, M. P. Miller, J. C. Marquis, J. M. Baraban, G. L. Semenza, and R. R. Ratan Protection from Oxidative Stress-Induced Apoptosis in Cortical Neuronal Cultures by Iron Chelators Is Associated with Enhanced DNA Binding of Hypoxia-Inducible Factor-1 and ATF-1/CREB and Increased Expression of Glycolytic Enzymes, p21waf1/cip1, and Erythropoietin J. Neurosci., November 15, 1999; 19(22): 9821 - 9830. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||