![]() |
|
|
Nutrition Department, College of Human Development, The Pennsylvania State State University, University Park, PA 16802
| ABSTRACT |
|---|
|
|
|---|
KEY WORDS: iron deficiency anemia immune system central nervous system exercise
| INTRODUCTION |
|---|
|
|
|---|
|
| CHEMISTRY |
|---|
|
|
|---|
Four major classes of iron-containing proteins carry out these
reactions in the mammalian system: 1) iron-containing
nonenzymatic proteins (hemoglobin and myoglobin), 2)
iron-sulfur enzymes, 3) heme-containing enzymes and
4) iron-containing enzymes that are noniron-sulfur,
nonheme enzymes (Fig. 2
). In the principal oxygen transport nonenzymatic proteins, hemoglobin
and myoglobin, iron functions as a critical ligand for the binding of
dioxygen. In iron-sulfur enzymes, iron participates in
single-electron transfer reactions primarily in energy metabolism.
In the third category, iron is bound to various forms of heme and
participates again in electron transfer reactions when associated with
various cofactors (e.g., cytochrome P450 complexes). The final group of
iron-containing enzymes is a catch-all grouping in which iron
is not bound to a porphyrin ring structure or in iron-sulfur
complexes.
|
The movement of oxygen from the environment to terminal oxidases is one of the key functions of iron in which oxygen is bound to porphyrin ring iron-containing molecules either as part of the prosthetic group of hemoglobin within red blood cells or as the facilitator of oxygen diffusion in tissue, myoglobin.
Hemoglobin is a tetrameric protein with two pairs of identical subunits
(
2, ß2; MW 64,000). Each subunit has one prosthetic group,
Fe-PP-IX, whose ferrous iron reversibly binds dioxygen. The synthesis
of erythroid heme is controlled in part by the availability of iron to
these maturing erythroblasts because iron regulates the initial
rate-limiting step in heme biosynthesis by altering the stability
of the specific [
-amino-levulinic acid synthetase] mRNA. The four
subunits are not covalently attached to each other but do react
cooperatively with dioxygen with specific modulation by pH,
pCO2, organic phosphates and temperature. These
modulators of the affinity of hemoglobin for iron determine the
efficiency of transport of oxygen from the alveoli capillary interface
in the lung to the red cellcapillary-tissue interface in peripheral
tissues. The allosteric effect of decreasing pH, the well-known
Bohr effect, decreases the binding affinity of heme-Fe for dioxygen
via protonation of His-146 on ß chains and Val-1 on
chains in the
presence of Cl-1 and
CO2. CO2 forms a Schiff
base with the terminal amino acids of each chain and decreases dioxygen
affinity. This favors the unloading of oxygen in tissues in which the
pH is lower and pCO2 is higher than in arterial
blood. 2,3-Diphosphoglycerate is a product of a side pathway within
erythrocytes and binds to a specific region of the ß chain to
decrease Hb-O2 binding affinity. Homeostasis with
respect to oxygen transport is evident in iron-deficient anemic
individuals. There is usually a right shift of the dissociation curve
with anemia, in which the blood content of hemoglobin is significantly
reduced, and an increased cardiac output that is only partially
compensatory. The increase in cardiac output is the result of an
increase in both stroke volume and heart rate, with a resulting
hypertrophy of the ventricular muscle wall that is concentric in nature
(Mederios and Beard 1998
). Physiological conditions that
increase the demand for oxygen transport such as high physical exertion
rates (discussed later in this review and by others at this meeting)
may exist, with a resulting decreased maximal aerobic capacity because
of limitations in oxygen transport ability. It is uncommon, however, to
find significant arterial desaturation unless very high rates of oxygen
transport are required.
Myoglobin is the single-chain hemoprotein in cytoplasm (MW 17,000)
that increases the rate of diffusion of dioxygen from capillary red
cells to cytoplasm and mitochondria. The concentration of myoglobin in
skeletal muscle is drastically reduced (4060%) in tissue iron
deficiency, thus limiting the rate of diffusion of dioxygen from
erythrocytes to mitochondria (Beaton et al. 1989
). The
sensitivity of myoglobin concentration in other tissues to iron status
has not been established.
Electron transport.
The second group of iron-containing proteins comprises heme enzymes
that use Fe within the porphyrin ring structure and are usually coupled
with other enzymes in integrated electron transport processes. The
overall scheme of the electron transport chain and the forms of iron
that participate in electron transport are summarized elsewhere
(Beard and Dawson 1996
). The family of cytochromes
contains heme as the active site, with the Fe-porphyrin ring
functioning to reduce ferrous iron to ferric iron with the acceptance
of electrons.
The third group of iron-containing proteins, the iron sulfur
enzymes, also acts as electron carriers via the action of iron bound to
either 2 or 4 sulfur atoms and cysteine side chains. The 40 different
proteins that constitute the respiratory chain contain 6 different heme
proteins, 6 iron sulfur centers, 2 copper centers, as well as
ubiquinone to connect NADH to oxygen. There are numerous examples of
these proteins and enzymes, which are summarized in Table 1
. Although the sensitivity of many of these enzymes to iron depletion
has been observed in one study or another, the systematic evaluation of
tissue sensitivity for this modest list is lacking.
|
| MODEL OF INTERORGAN IRON EXCHANGE |
|---|
|
|
|---|
|
Iron turnover and redistribution.
Iron turnover is mediated primarily by destruction of senescent
erythrocytes by the reticuloendothelial system (Finch et al. 1970
). For example, in a 70-kg individual with a normal iron
status,
35 mg/d of iron is turned over in the plasma (Finch et al. 1970
). Erythrocytes, which contain
80% of the
bodys functional iron, have a mean functional lifetime of 120 d
in humans. About 85% of the iron derived from hemoglobin degradation
is rereleased to the body in the form of iron bound to transferrin or
ferritin. Each day, 0.66% of the bodys total iron content is
recycled in this manner (Finch et al. 1970
). Smaller
contributions are made to plasma iron turnover by the degradation of
myoglobin and iron-containing enzymes.
Storage and mobilization of iron.
The concentration of iron in the body is
3040 mg/kg body. However,
that concentration varies as a function of the age and gender of the
individual and the specific tissues and organs being examined. About
8590% of nonstorage iron is found in the erythroid mass
(Bothwell et al. 1979
). The storage iron concentration
in the body varies from 0 to 15 mg/kg body, depending on the gender and
the iron status of the individual. The distribution of this stored iron
is not uniform because the liver contains
60% of the ferritin in
the body. The remaining 40% is found in muscle tissues and cells of
the reticuloendothelial system. Normally, 95% of the stored iron in
liver tissue is found in hepatocytes as ferritin. Hemosiderin
constitutes the remaining 5% and is found predominately in Kupffer
cell lysosomal remnants. However, during iron overload, the mass of
hemosiderin in the liver accumulates at 10 times the rate of ferritin
(Green et al. 1968
).
The overall structure of ferritin is conserved among higher eukaryotes
and, in humans, is composed of 24 polypeptide subunits. At least two
distinct isoforms of the polypeptide subunits exist, and combinations
of these subunits allow for considerable heterogeneity in the structure
of the full protein. The isoform designated H ferritin is a 22-kDa
protein composed of 182 amino acids. The L isoform is a 20-kDa protein
containing 174 amino acids. The subunit composition of ferritin seems
to be tissue specific. Theoretically, up to 4500 ferric iron atoms can
be stored in ferritin (Fishbach and Andreregg 1965
). In
vivo, ferritin is normally 20% saturated (800 of 4500 iron sites
occupied) with a variable ratio of H to L subunits (Crichton and Ward 1992
). The structure and composition of the mineralized
core is analogous to a polymer of ferrhydrite
(5Fe2O3 · 9H2O)
with a variable amount of phosphate (Fishbach and Andreregg 1965
). The
H-chain possesses a distinct ferroxidase site, which leads to the
current hypothesis that high H-Lratio ferritin functions primarily to
quickly move iron into and out of the core (Levi et al. 1988
). The L-chain of ferritin lacks this ferroxidase site,
and L-chainpredominant ferritin is viewed as a longer-term
storage pool of iron (Casey et al. 1988
, Levis et al. 1989
). Pertinent to this discussion of the role of ferritin
in the storage of iron is the proposed role of H-ferritin in growth
and development and perhaps a role in nuclear regulation because
H-ferritin receptors have been identified in several tissues and
characterized in the nucleus of neurons (Hulet et al. 1999
).
Iron losses.
The low solubility of iron precludes excretion as a major mechanism of
maintaining iron homeostasis. Thus, in contrast to most other trace
minerals whose homeostasis in maintained by excretion, the primary
mechanism of maintaining whole-body iron homeostasis is to regulate
the amount of iron absorbed so that it approximates iron losses
(Hallberg and Hulthen 2000
). Iron losses can vary
considerably with the gender of the individual and pathologies that
have blood loss as a significant component. In male humans, total iron
losses from the body have been calculated to be
0.80.9 mg/d. For
premenopausal female humans, this loss is slightly higher. The
predominant route of loss is from the gastrointestinal tract and
amounts to 0.6 mg/d in adult males (Finch et al. 1970
).
Fecal iron losses result from shed enterocytes, extravasated red blood
cells and biliary heme breakdown products that are poorly absorbed.
Urogenital and integumental iron losses have been estimated to be
> 0.1 and 0.3 mg/d, respectively, in adult males (Green et al. 1968
). Menstrual iron loss, estimated from an average blood
loss of 33 mL/mo, equals 1.5 mg/d but may be as high as 2.1 mg/d
(Cole et al. 1971
). Oral contraceptives reduce this loss
(Cole et al. 1971
, Frassinelli-Gunderson et al. 1985
), and intrauterine devices increase it (Cole et al. 1971
, Guillebaud et al. 1979
, Kivijarvi et al. 1986
).
| GENERAL CLINICAL MANIFESTATIONS |
|---|
|
|
|---|
|
| IMPAIRED IMMUNE FUNCTION |
|---|
|
|
|---|
Immunity during iron deficiency.
Experimental and clinical data suggest that there is an increased risk
of infection during iron deficiency, although a small number of reports
indicate otherwise. Hershko (1996)
urges caution in the
interpretation of many studies because the confounding issues of
poverty, generalized malnutrition and multimicronutrient deficiencies
are often present in those studies. The molecular and cellular defects
responsible for immune deficiency are complex because almost every
effector of the immune response is limited in number or action by
experimental iron deficiency. As mentioned previously, iron is
essential for proper cell differentiation and cell growth. In addition,
iron is a critical component of peroxide-generating enzymes and
nitrous oxidegenerating enzymes that are critical for the proper
enzymatic functioning of immune cells. Finally, iron is likely involved
in the regulation of cytokine production and mechanism of action
through its influence on second-messenger systems (Hershko 1996
). In one of few studies of role of iron nutrition in the
development of the immune system, a delay was noted in the development
of cell-mediated immunity (Kochanowski and Sherman 1985
).
In adult animals or humans with intact immune systems, nonspecific
immunity is affected by iron deficiency in several ways. Macrophage
phagocytosis is generally unaffected by iron deficiency, but
bactericidal activity of these macrophages is attenuated
(Hallquist et al. 1992
). Neutrophils have a reduced
activity of the iron-containing enzyme, myeloperoxidase, which
produces reactive oxygen intermediates responsible for intracellular
killing of pathogens (Spear and Sherman 1992
). There is
a decrease in both T-lymphocyte number and T-lymphocyte
blastogenesis and mitogenesis in iron deficiency in response to a
number of different mitogens. This alteration is largely correctable
with iron repletion (Kuvibidila et al. 1999
,
Spear and Sherman 1992
). Recent studies of T lymphocytes
in iron deficiency noted that protein kinase C activity and
translocation of both splenic and purified T cells were altered by iron
deficiency (Kuvibidila et al. 1999
). On the other hand,
others have found a normal T-lymphocyte proliferative response to
mitogens (Canonne-Hergaux et al. 1999
).
Humoral immunity appears to be less affected by iron deficiency than is
cellular immunity. In iron-deficient humans, antibody production in
response to immunization with most antigens is preserved
(Hallquist et al. 1992
, Spear and Sherman 1992
).
Mediators of immune functioning.
There are several possible mechanisms that could explain the effects of
iron deficiency on the immune system. DNA synthesis, initiated by the
iron-containing enzyme ribonucleotide reductase, is a
rate-limiting factor in cellular replication and may be limited by
iron deficiency. Control of differentiation of cells is influenced by
the available iron and iron transport into cells via the transferrin
receptor. Galan and colleagues (1992)
reported a
reduction in interleukin-2 production by activated lymphocytes in
iron-deficient subjects. The release of interleukin-2 is
fundamental to communication between lymphocyte subsets and natural
killer cells but it does not appear to be the only cytokine that is
altered by iron status (Sussman 1974
).
Acute-phase reactants.
Tumor necrosis factor, interleukin-1 and interferon-
all work as
effectors of iron movement. These cytokines operate in a coordinated
fashion to reduce the size of the intracellular labile iron pool by
reducing the amount of transferrin receptor on the cell surface,
increasing the synthesis of ferritin for iron storage and activating
nitric oxide systems (Damsdaran et al. 1979
,
Fishbane 1999
, Hallquist et al. 1992
,
Ike et al. 1992
, Kochanowski and Sherman 1985
, Murray et al. 1978
, Sussman 1974
). Regulation of gene transcription is a likely mechanism.
It is less apparent that the iron status of the individual can modify
the acute-phase response system (i.e., does the ability to mount an
acute-phase response depend in part on iron nutritional status?).
Nonetheless, there is a well-known decrease in the plasma iron
concentration as well as an increase in the plasma ferritin
concentration. Careful studies of the iron content of this newly
released ferritin are not well established; thus the role of plasma
ferritin in the sequestration of plasma iron remains uncertain.
Sequestration of iron seems to be an important part of the host
response to infection. Administration of a potent iron chelator,
deferoxamine, to humans was examined to determine the potential
antimalaria effect (Byrd and Horwitz 1989
, Fahmay and Young 1993
, Konijn and Hershko 1977
,
Lane et al. 1991
).
| MENTAL FUNCTION DURING IRON DEFICIENCY |
|---|
|
|
|---|
The brain obtains iron primarily via transferrin receptors expressed on
endothelial cells on the brain microvasculature (Connor and Benkovic 1992
, Fishman et al. 1987
). The
regulation of iron movement across this barrier is not well understood,
although regulation in response to organ iron status clearly occurs.
That is, the rate of iron uptake into the brain is increased when the
iron status of the subject is low and is decreased when the iron status
is higher (Taylor et al. 1991
). In addition, the process
is highly selective and not reflective of overall blood-brain
permeability (Crowe and Morgan 1992
, Morris et al. 1992
). The uptake of iron is reported to be homogeneous and
to be followed by redistribution to the basal ganglia, although there
are other possibilities that depend on local regulation of uptake
(Dwork et al. 1988
). The choroid plexus is a rich source
of transferrin mRNA, and transferrin secreted by this organ presumably
is used for the distribution of iron to glia and neurons for use or
storage. It is important to remember that plasma transferrin can move
across the blood-brain barrier and become part of the circulating
pool of transferrin in cerebrospinal fluid. The adaptive regulation of
transferrin synthesis and action in response to iron-deficiency
anemia in the brain is unknown.
Regions of the brain rich in iron in adulthood, i.e., the substantia
nigra, globus pallidus, nucleus accumbens, are not the regions that
have a high iron content in early life. In addition, they are far less
affected by dietary iron deficiency than are other regions such as the
cortex or the striatum that have less iron content (Erikson et al. 1997
). This iron is located primarily in microglia and
oligodendrocytes and functions in numerous metabolic activities
(Beard et al. 1993
, Epstein and Connor 1999
, Hill et al. 1985
, Mash et al. 1990
). The regional heterogeneity in the deposition of iron in
the brain is remarkably similar across many species with the basal
ganglia, substantia nigra and deep cerebellar nuclei particularly rich
in iron (Aoki et al. 1989
, Benkovic and Connor 1993
). Recent studies from our laboratory and those of
associates showed that iron accumulation in different brain regions is
a function of the stage of brain development occurring at the time of
the investigation (Fig. 4A
,
B) (Pinero et al. 2000
). For example, when brain iron
distribution is studied in a rodent model of lactational iron
deficiency, the pattern of iron loss that emerges is entirely different
from that for the effects of iron deficiency instituted during the
postweaning period.
|
Iron and transferrin levels have been reported to be high in
cerebrospinal fluid, especially in perinatal brains (Erikson et al. 1997
). The actual levels of iron, however, are poorly
described in conditions of iron overload and iron deficiency and during
active growth and development. Atomic absorption spectrophotometry
reveals iron concentrations of
1525 µg/L in humans
and monkeys and 520 µg/L in mice (Bradbury 1997
). These concentrations are
5- to 10-fold lower than the
corresponding plasma concentrations. Some unpublished data from our
laboratory indicate that chronic diseases and pathologies (hepatitis,
liver failure, diabetes) may alter these concentrations dramatically.
In patients with restless-legs syndrome, the ratio of plasma iron
to cerebrospinal fluid iron is abnormal, suggesting that the movement
of iron across the blood-brain barrier is perturbed in these
individuals (Earley et al. 2000a
). The normal
circulating level of transferrin in cerebrospinal fluid is also poorly
described, although some reviews suggest that the total
iron-binding capacity is barely even with the circulating iron
concentrations, resulting in the apparent availability of free iron
(Bradbury 1997
). The role of the cerebrospinal fluid in
the delivery of iron to various brain cells is not well understood
(Dwork 1995
, Malecki et al. 1998
).
The brain is richer in the H subunit than in the L subunit of ferritin,
and its localization is specific to cell type (Connor and Benkovic 1992
). In rats and humans, microglia and
oligodendrocytes contain ferritin, whereas in mice astrocytes contain
ferritin. Ferritin levels correlate with brain iron content; they are
highest at birth and decline thereafter in newborn rats (Miller et al. 1996
). Moreover, the concentration can be directly
affected by the body iron burden (Chen et al. 1995b
).
Ferritin isoforms are heterogeneously distributed in brain, and not all
regions of the brain seem to be equally sensitive to an alteration in
body iron status (Erikson et al. 1997
, Kivijarvi et al. 1986
). Studies in postnatal iron deficiency involving
ferritin ratios (H:L) in pig and rodent brain reveal a dramatic effect
of iron deficiency and the expression of these two subunits of the
ferritin molecule (Erikson et al. 1998
, Han et al. 2000
). The changes in the protein expression appear to be
regulated by the iron-responseelement system as described previously.
The developmental roles of the two subunits relative to iron storage or
use and detoxification are unknown, although accumulation of iron in
certain brain regions is believed to play a role in a number of
neuropathologies.
Iron deficiency effects.
Dallman and co-workers (Dallman et al. 1975
,
Dallman and Spirito 1977
) demonstrated two decades ago
that young rats deprived of iron in early postnatal life have
significantly lower (27%) whole-brain iron content than do
controls 28 d postnatally and were quite resistant to restoration
of their normal complement of brain iron (still 20% lower) despite
aggressive dietary repletion for 45 d. Although these studies were
landmark investigations at the time, they were usually misinterpreted
to indicate that brain iron content was very static and not at all
sensitive to dietary iron deficiency. This concept of a protected organ
persisted for nearly two decades, until it was demonstrated by us and
others that the brain is quite sensitive to dietary iron depletion and
repletion and uses a host of mechanisms to regulate iron flux
homeostatically.
When animals are given low iron diets in postweaning life, there is a
significant decline in brain iron content and a rapid repletion with
refeeding (Chen et al. 1995b
, Erikson et al. 1997
). This is in contrast to neonatal or preweaning iron
deficiency in which the effects appear irreversible (Dallman and Spirito 1977
, Felt and Lozoff 1996
). From animal
studies across a number of species, we assume that human brain iron
content goes down with a decrease in body iron status, although there
is no published direct proof of this. Preliminary magnetic resonance
imaging data from several restless legs syndrome patients with
iron-deficiency anemia have weighted T2 relaxation times consistent
with depleted striatal and nigra iron contents (C. Earley, unpublished
data, personal communication). The areas of the brain that are quite
sensitive to iron depletion in early life often are located within
dopaminergic regions of the brain (Chen et al. 1995a
,
Erikson et al. 1997
). Hill (1988)
argues
that the areas of highest iron concentration in the adult are not
identical to those brain regions in which dopaminergic neurons either
originate or terminate, which is not in agreement with Youdim et al. (1989)
.
Iron is required for proper myelination of the spinal cord and white
matter of cerebellar folds (Larkin and Rao 1990
), and it
is a cofactor for a number of enzymes involved in neurotransmitter
synthesis, including tryptophan hydroxylase (serotonin) and tyrosine
hydroxylase (norepinephrine and dopamine). Iron is also a cofactor for
ribonucleotide reductase, the rate-limiting step in DNA synthesis.
The predominant cell type containing iron in the brains of mice, rats,
monkeys, pigs and humans is the oligodendrocyte (Hill 1988
). These cells are responsible for the production of
myelin; hence alterations in the functioning of these cells are
associated with hypomyelination. Oligodendrocytes are responsible for
the synthesis of fatty acids (Mackler et al. 1979
,
McKay et al. 1983
) and cholesterol for myelin; both of
these metabolic processes require iron. When oligodendrocyte maturation
is disrupted, as results from some gene mutations, iron accumulation is
only
50% of normal (Connor and Menzies 1990
). In
iron deficiency, oligodendrocytes appear immature (Erikson et al. 1997
). The failure to deliver iron to these cells during
particular periods of early brain development could be causally related
to delayed motor maturation and perhaps behavioral alterations in young
humans (Rocangliolo et al. 1996
). These investigators
demonstrated a slowed nerve conduction velocity during an auditory
evoked potential test. The reversibility of this finding remains under
investigation. Although there are no quantitative data showing that
iron deficiency leads to a lesser number of oligodendrocytes,
hypomyelination occurs as a result of postnatal iron deficiency
(Larkin and Rao 1990
, Wiesinger et al. 2000
). In addition, iron deficiency could block cholesterol
biosynthesis in these cells because at least three isoforms of P450 (a
protoheme monooxygenase) are found specifically in oligodendrocytes.
| IRON AND NEUROTRANSMITTER SYSTEMS |
|---|
|
|
|---|
Oxidation-reduction.
Iron-dependent electron transport alterations in the brain resulting
from iron deficiency are sparsely documented. Mackler and
colleagues (1979)
demonstrated that cytochrome concentrations
in mitochondria from brain of iron-deficient animals were not
different from those of controls. In addition, oxidative
phosphorylation in these mitochondria using pyruvate-malate,
succinate or
-glycerol phosphate was unaffected by iron deficiency.
In animals with similar severities of iron-deficiency anemia
(hemoglobin <60 g/L), skeletal muscle oxidative capacity was reduced
by 4050% (Sourkes 1973
). At least superficially, it
does not appear that there is a massive decrease in oxidative
metabolism in the iron-deficient brain. This must be viewed with
caution, however, because the brain is one of the most oxidative organs
of the body.
Synthesis and degradation.
Iron is a cofactor for tyrosine hydroxylase, tryptophan hydroxylase,
xanthine oxidase and ribonucleoside reductase (Youdim and Green 1978
). Thus, nutritional iron deficiency would be expected to
lead to decreased activities of these enzymes; however, this has not
been observed consistently. When brain iron levels are reduced by as
much as 40% with dietary restriction in postweanling rats, there was
no change in the activity of tyrosine hydroxylase, tryptophan
hydroxylase, monoamine oxidase, succinate hydroxylase or cytochrome c
oxidase (Ashkenazi et al. 1982
). Aminobutyric acid
transaminase and glutamate decarboxylase activities decreased, but
these observations have yet to be reproduced by others. Whole-brain
concentrations of norepinephrine and dopamine were unaltered by iron
deficiency (Ashkenazi et al. 1982
) and responded equally
well to priming doses of levodopa (50 mg/kg). Serotonin and
5-hydroxyindole acetic acid concentrations were reported to be
decreased. The turnover of norepinephrine, dopamine and serotonin in
brain homogenates was also unaffected by iron deficiency, but more
sophisticated approaches have not be used in recent years (Felt and Lozoff 1996
, Youdim and Green 1977
,
Youdim et al. 1989
).
Monoamine oxidase and aldehyde dehydrogenase are critical in the
catabolism of neurotransmitters in the dopaminergic, serotoninergic and
noradrenergic systems of the brain (Youdim et al. 1975
and 1980
). However, Mackler et al. (1979)
were
unable to demonstrate any decreased activity in brain of severely
iron-deficient rats. Youdim et al. (1980)
also
showed no effect of iron deficiency in whole-brain preparations,
although heart monoamine oxidase activity was decreased by >50%.
-Aminobutyric acid.
Hill (1988)
noted great similarity in the brain iron
distribution and brain regions that receive input from
-aminobutyric
acid (GABA). She argued that the regions that have the highest
concentrations of iron, the globus pallidus, substantia nigra, ventral
pallidus and the cerebellar nuclei, also are highly innervated by
GABA-mediated nerve tracts. Because GABA release will modulate the
activity of dopaminergic neurons, this is an important point to be
resolved. In an older study, Youdim and colleagues
(1980)
observed no effect of iron deficiency on the GABA
receptor population, amount of GABA or production rates. Other
investigators, however, report that iron deficiency in utero and
postweaning is associated with significant decreases in glutamate
decarboxylase, glutamate dehydrogenase and GABA transaminase activities
(Li 1998
, Taneja et al. 1986
). These
latter two enzymes are shunt enzymes responsible for the synthesis and
degradation of GABA. Thus, although concentrations of GABA may not be
changed, there are some indications that GABA metabolism is altered by
iron deficiency. Little substantive biochemical work has been done on
this interaction, but such studies would expand the range of possible
neurotransmitter systems that are altered by iron-deficiency
anemia. No distinction has been made between intrauterine effects of
iron deficiency and later effects that may occur during lactation and
after weaning.
Dopamine.
The dopaminergic system in the only neurotransmitter system in the
central nervous system that has been consistently sensitive to
experimental changes in iron status. As whole-brain iron content
drops 15% below normal, biological and behavioral alterations occur
that may result from changes in the dopaminergic system (Erikson et al. 1997
and 2000
, Morse et al. 1999a
,
Nelson et al. 1997
, Yehuda 1990
,
Youdim 1990
). Youdim and colleagues measured affinities
and densities for dopamine D1 and
D2 receptors, serotonin, GABA, benzodiazepine,
and
and ß adrenergic and muscarinic-cholinergic receptors in
brain regions after postweaning dietary iron deficiency. They observed
an effect of severe postweaning iron deficiency only on the dopamine
system. The more recent experiments demonstrate that striatum dopamine
D1 and D2 receptor
densities are significantly lower (2535%) in postweaning
iron-deficient rats and that the dopamine transporter is also
significantly lower in density in several brain regions (Erikson et al. 1997
, Pinero et al. 2000
).
Recent in vivo animal data demonstrate that extracellular dopamine is
elevated in striatum of postweaning iron-deficient rats and returns
to normal levels when brain iron content and iron status return to
normal (Chen et al. 1995a
, Nelson et al. 1997
). Pharmacologic experiments with cocaine, a dopamine
transporter inhibitor, demonstrate both in vivo and in vitro deficits
in dopamine transport from the intersynaptic space back into
presynaptic neurons (Erikson et al. 2000
). Attentional
processing of environmental information is highly dependent on
appropriate rates of dopamine clearance from the interstitial space,
which suggests that iron status may affect behavior through effects on
dopamine metabolism. Alterations in dopamine in the mesolimbic and the
nigrostriatal tracts are associated with changes in motor control as
well as altered perception, memory and motivation. However, lesions in
many other parts of the brain may also result in alterations in
perception, memory and motivation; thus the specificity of the
connection between striatal dopamine changes and impaired spatial
memory, attentional deficits and avoidance behavior remains to be
established.
Serotonin and norepinephrine.
As mentioned earlier, tryptophan and tyrosine hydroxylase are
iron-containing enzymes that are essential for the production of
serotonin from tryptophan and norepinephrine and dopamine from tyrosine
(Webb 1992
, Youdim and Green 1977
and 1978
). As part of their initial survey of neurotransmitter
systems that may be affected by iron deficiency, Youdim and colleagues
measured the concentrations of serotonin, norepinephrine and their
primary metabolites, 5-hydroxy indole acetic acid and normetanephrine
(Youdim and Green 1977
and 1978
). They observed no
significant alterations in concentrations of these neurotransmitters or
metabolites in striatum of rats. Other studies of monoamine oxidase
activity in brain of iron-deficient rats showed no effect on this
enzyme responsible for the degradation of the monoamines, although
platelet monoamine oxidase activity was affected (Youdim et al. 1975
). Our laboratory has also examined these two
neurotransmitter systems with in vivo microdialysis and failed to
observe consistent alterations in their concentrations in extracellular
fluid in the brain (Chen et al. 1995a
, Nelson et al. 1997
). However, radioligand binding studies performed in
inbred strains of mice demonstrated significantly lower densities of
the serotonin transporter in striatum of iron-deficient mice
(Morse et al. 1999a
). This observation has some
significance because the dopamine, norepinephrine and serotonin
transporters have a high degree of homology. Our observations of a
consistent decrease in dopamine transporter density in striatum of
iron-deficient rats, in combination with the one study of serotonin
transporter, suggest a more general role of iron in the removal of
neurotransmitters from the synaptic cleft.
Developmental models.
Irreversible alterations in brain iron content have been shown in
animal studies by feeding rats low iron diets early in life before the
completion of the brain organization and myelination and the
establishment of the dopaminergic tracts (Dallman and Spirito 1977
, Felt and Lozoff 1996
). Felt and colleagues
demonstrated behavioral changes that could be associated with the
irreversible changes in brain iron content that occurred, but they did
not determine distribution of brain iron. This is in contrast to the
ability of postweanling or late-lactational iron-deficient rats to
recover brain iron content and functioning (Erikson et al. 1997
, Pinero et al. 2000
). In those studies, the
investigators demonstrated a prompt recovery of brain iron in nearly
all brain regions with feeding of a high iron diet. These studies, in
combination, suggest that important biological switches for the
acquisition of brain iron in early development may be irreversibly
altered. A significant caveat to the observations from these rodent
studies is that much of the rodent brain maturation occurs postnatally.
Peak myelination of the rodent brain is occurring between postnatal d 8
and 14, whereas in the human infant, this occurs between ages 8 and 15
mo. Thus, the timing of the nutritional studies relative to
interspecies comparisons must be carefully considered. Felt and Lozoff (1996)
developed a very nice paradigm for this
comparison that has been used very successfully by our laboratory to
produce appropriate periods of iron deficiency to mimic lactational and
early postlactational nutritional stress (Erikson et al. 1997
, Pinero et al. 2000
). Thus, much of the
early disconnect between data collected in animal models and
observations in human infants has been avoided. The next several years
will see the publication of a number of articles from active research
groups using rodent and primate models of iron deficiency that will
further delineate the biological underpinnings.
| IMPAIRED PHYSICAL PERFORMANCE |
|---|
|
|
|---|
Lethargy, apathy and listlessness are frequently observed symptoms of
severe iron-deficiency anemia and perhaps anemia in general.
However, for some time, iron deficiency has been known to be associated
with decreased physical activity (Baynes and Bothwell 1990
, Dallman 1982
and 1986
, Finch and Heubers 1982
). The consistency of evaluation of this effect on
exercise tolerance and work performance hinges on the definitions of
iron-deficiency anemia that have used by investigators. This has
led to some ambiguity in the results. However, the mechanisms of these
effects have been thoroughly investigated in rodent models, and clear
distinctions between the effects of diminished oxygen transport and
oxidative capacity of muscle have been established (Dallman 1986
). When rats are fed various contents of iron in purified
diets, it is straightforward to establish that muscle myoglobin and
cytochrome c are affected to a proportionally similar degree as
hemoglobin with decreased iron intakes in young growing animals. That
is, a severely anemic rat with a 50% decrease in hemoglobin
concentration also has
50% lower myoglobin and cytochrome c
concentrations. That deceased exercise capacity is related to
diminished oxygen transport, diminished oxygen diffusion within the
exercising tissue and decreased oxidative capacity of muscle
(Maguire et al. 1982
, McLane et al. 1981
). Although the efficiency of oxygen extraction is improved
by the hemoglobin-oxygen saturation curve shifts,
VO2max is still decreased 3050% in both
animals and humans.
The manifestations of depletion of essential body iron also have
profound effects on skeletal muscle, with a significant decrease in
mitochondrial iron-sulfur content (Maguire et al. 1982
), mitochondrial cytochrome content (McKay et al. 1983
, McLane et al. 1981
, Willis et al. 1987
) and total mitochondrial oxidative capacity (Davies et al. 1982
and 1984
, McKay et al. 1983
,
Willis et al. 1987
). Pyruvate and malate oxidase were
decreased to 35% of normal in iron-deficient muscle and improved
to 85% of normal after 10 d of iron treatment. 2-Oxoglutarate
oxidase was decreased to 47% of normal and improved to 90%. In
contrast, succinate oxidase was only 10% of normal in iron deficiency
and improved to only 42% of normal after 10 d (Perkkio et al. 1985
). The cytoplasmic enzymes, hexokinase and lactate
dehydrogenase, were unaffected by iron status. The 5090% decrease in
both the iron-sulfur enzymes and in the heme-containing
mitochondrial cytochromes is consistent with many observations over the
past two decades (Dallman 1986
, Willis et al. 1987
). What seems to determine the amount of decline in
activity with iron deprivation is the rate of turnover of that
particular iron-containing protein in the time of cellular
deprivation of iron. Although this makes sense conceptually, the direct
testing of this hypothesis occurred in a cursory fashion.
Exercise performance at the time of severe iron-deficiency anemia
was only 20% of normal in a brief, intense exercise protocol
(Willis et al. 1987
). Treatment with iron dextran
corrected this exercise tolerance within 4 d, but mitochondrial
oxidation using either malate-pyruvate or succinate as substrates
was not improved. When
-glycerol phosphate was used as the
substrate, however, a very significant improvement in the rate of
oxidation was realized, suggesting that this is the rate-limited
step in iron-dependent oxidation. Longer-term repletion led to
the normalization of other enzyme activities (Willis et al. 1990
). The interpretation of these data is consistent with
rates of turnover of cellular enzymes that require iron and thus
support the premise that rates of loss of enzymatic function in muscle
are related to turnover of iron-containing enzymes within those
cells. A study using 31P nuclear magnetic
resonance spectroscopy to examine the functional state of bioenergetics
in iron-deficient and iron-replete rat gastrocnemius muscle at
rest and during 10 min of contraction at 2 Hz demonstrated an effect of
iron status (Thompson et al. 1993
). Iron-deficient
animals had a clear increase in phosphocreatine breakdown and a
decrease in pH compared with controls and a slower recover of
phosphocreatine and inorganic phosphate concentrations after exercise.
During repletion for 27 d with iron dextran, there was no substantial
improvement in these indicators of muscle mitochondrial energetics.
These authors concluded that "tissue factors" such as decreased
mitochondrial enzyme activity, decreased number of mitochondria and
altered morphology of the mitochondria might be responsible for these
observations.
Anemia vs. tissue iron deficiency.
Dallman and colleagues established the concept that anemia limits the
capacity of the individual to deliver oxygen to exercising muscle,
whereas tissue iron deficiency limits the capacity of the individual to
perform oxidative metabolism (Dallman 1982
and 1986
).
This was accomplished by iron-repletion and
exchange-transfusion studies and the observation of recovery of
sprint performance compared with endurance performance (Fig. 5
). There was a stronger correlation of sprint performance with anemia
than with tissue cytochrome c content, whereas there was a strong
correlation of tissue cytochrome c with recovery of endurance
performance. Exchange-transfusion experiments in which the
hemoglobin concentration of animals was manipulated by transfusion
further demonstrated that endurance performance is almost entirely
related to muscle metabolism and relatively independent of hemoglobin
concentration down to 100 g/L.
|
Iron-containing enzymes in skeletal muscle and liver are altered in
iron deficiency to promote an increased rate of lactate production in
muscle and use by liver (Davies et al. 1982
,
Finch et al. 1979
, Henderson et al. 1986
,
Ohira et al. 1986
, Thompson et al. 1993
).
These animal studies clearly demonstrated that the rates of plasma
disappearance of radioactively labeled glucose and lactate were both
increased in iron-deficiency anemia. Even at rest, there were
higher concentrations of glucose in the plasma of iron-deficient
animals than controls. Additional studies demonstrated that lactate
dehydrogenase is increased in activity in iron-deficient skeletal
muscle and that isozyme adaptations occur to maximize this capacity for
anaerobic metabolism (Ohira et al. 1986
). In contrast,
other more recent experiments suggest that a number of nonheme
iron-sulfur proteins, especially aconitase, may be very susceptible
to variations in cellular iron content (Willis et al. 1990
). Dallmans laboratory demonstrated that this key enzyme
in gluconeogenesis could be restored to full activity with 15 h of
an iron injection in the rat model, with a rapid return of lactate and
glucose concentrations toward normal. Rates of gluconeogenesis in
iron-deficiency anemia are increased to improve the provision of
blood glucose (Klempa et al. 198