Regulation of phagosomal iron release from murine macrophages by nitric oxide.

Abstract

The role of NO in macrophage iron turnover was studied in macrophages from inducible nitric oxide synthase (iNOS)-deficient mice. Interferon gamma/lipopolysaccharide (IFNgamma/LPS)-activated bone marrow-derived macrophages from iNOS-deficient mice, following phagocytosis of 59Fe-labelled transferrin-anti-transferrin immune complexes, showed reduced iron release compared with cells from wild-type iNOS littermates. Uptake of the complexes by macrophages was similar in iNOS-deficient and wild-type mice. Ferritin was up-regulated by IFNgamma/LPS treatment, but NO exercised a modest opposing down-regulatory effect. No effect of iNOS deficiency was seen when iron was taken up from iron citrate, which enters via a non-phagocytic route. These results suggest that NO plays a key role in regulating iron turnover in macrophages acquiring iron by phagocytosis of erythrocytes or cell debris, and thus the supply to peripheral tissues, such as to the bone marrow for erythropoiesis.

Full text

Page 1

Biochem. J. (2002) 365, 127–132 (Printed in Great Britain)
127
Regulation of phagosomal iron release from murine macrophages by
nitric oxide
Victoriano MULERO*†1, Xiao-qing WEI*, Foo Y. LIEW* and Jeremy H. BROCK*
*Department of Immunology and Bacteriology, Western Infirmary, University of Glasgow, Glasgow, U.K., and †Department of Cell Biology, Faculty of Biology,
University of Murcia, 30100 Murcia, Spain
The role of NO in macrophage iron turnover was studied in
macrophages from inducible nitric oxide synthase (iNOS)-defi-
cient mice. Interferon γ?lipopolysaccharide (IFNγ?LPS)-acti-
vated bone marrow-derived macrophages from iNOS-deficient
mice, following phagocytosis of ??Fe-labelled transferrin–anti-
transferrin immune complexes, showed reduced iron release
compared with cells from wild-type iNOS littermates. Uptake
of the complexes by macrophages was similar in iNOS-deficient
and wild-type mice. Ferritin was up-regulated by IFNγ?LPS
treatment,but NO exerciseda modest opposing down-regulatory
INTRODUCTION
Macrophages play a critical role in iron metabolism. They are
responsible for the processing of haemoglobin iron from sen-
escent erythrocytes and subsequent iron supply to the bone
marrow for erythropoiesis [1]. In addition, iron is essential in
macrophage-mediated cytotoxicity by contributing to the pro-
duction of highly toxic hydroxyl radicals via the Fenton reaction
[2] and by controlling the production of NO after activation by
immunological stimuli [3].
Intracellular iron homoeostasis is usually controlled by cyto-
plasmicironregulatoryproteins(IRP1andIRP2),whichregulate
expression of several proteins by binding to iron-responsive
elements (IREs) on their mRNA [4,5]. IRP binding to the IREs
in the 5?-untranslated regions of ferritin (Ft) and 5-aminolaevu-
linate synthase mRNAs represses their translation, whereas
binding of IRPs to multiple IREs in the 3?-untranslated region of
transferrin (Tf) receptor (TfR) mRNA confers stability against
targeted endonucleolytic degradation.
IRP binding activity is normally regulated by cellular iron
levels [6,7], but can also be modulated by NO [5,8,9] and H?O?
[5,8,10], both of which are produced by activated macrophages.
NO was shown originally to activate IRE-binding activity of
both IRP1 and IRP2 [8], but we [11] and others [12] have shown
more recently that activation of macrophages with interferon γ
(IFNγ) causes a down-regulation of IRP2 activity, despite
production of NO. In addition, we have reported that IFNγ?
lipopolysaccharide (LPS) stimulation of murine J774 macro-
phages decreases TfR expression and Fe uptake from Tf [11].
Therefore, the regulation of iron uptake from Tf in activated
macrophages seems to be very complex and does not depend
solely upon NO-mediated activation of IRPs. Furthermore, NO
can cause release of iron from Ft [13] and may thus be able to
mobilize intracellular stores.
Abbreviations used: BMMφ, bone marrow-derived macrophages; FCS, fetal calf serum; Ft, ferritin; IFNγ, interferon γ; iNOS, inducible nitric oxide
synthase; IRE, iron-responsive element; IRP, iron regulatory protein; LPS, lipopolysaccharide;
nitrilotriacetate; Tf, transferrin; TfR, Tf receptor.
1To whom correspondence should be addressed, at the Department of Cell Biology, Faculty of Biology, University of Murcia, 30100 Murcia, Spain
(e-mail vmulero?um.es).
L-NMMA, NG-monomethyl-L-arginine; NTA,
effect. No effect of iNOS deficiency was seen when iron was taken
up from iron citrate, which enters via a non-phagocytic route.
These results suggest that NO plays a key role in regulating iron
turnover in macrophages acquiring iron by phagocytosis of
erythrocytes or cell debris, and thus the supply to peripheral
tissues, such as to the bone marrow for erythropoiesis.
Keywords:ferritin,interferon,lipopolysaccharide,phagocytosis,
transferrin.
Although most previous studies of iron metabolism in macro-
phages have used Tf as the carrier to load the cells with iron, the
most important function of macrophages in iron metabolism is
to process effete erythrocytes and return the iron to the plasma
for subsequent re-utilization [1]. In this case, iron enters the
macrophage not via the TfR, but by phagocytosis, and the main
intracellular compartment involved in iron processing is the
phagolysosome, rather than the early endosome. Impaired iron
release by macrophages probably contributes to the hypo-
ferraemia of inflammation, and we have shown previously that
tumour necrosis factor α can reduce release of iron taken up
previously by phagocytosis [14]. However, the way in which the
cells handle iron acquired by this route has not been studied in
detailandtheroleofNO,whichisalsoproducedbymacrophages
during immune and inflammatory responses, has not been
investigated. The aim of this work was therefore to clarify the
role of NO in the handling of iron taken up by activated
macrophages using a phagocytic route. To this end, we examined
cellular iron uptake, release and distribution, and Ft expression
in activated macrophages from inducible nitric oxide synthase
(iNOS)-deficient mice using a model based on phagocytosis of
??Fe-labelled Tf–anti-Tf immune complexes [15].
EXPERIMENTAL
Reagents
RPMI 1640 culture medium, antibiotics, PBS and fetal calf
serum (FCS) were purchased from Gibco (Paisley, Scotland,
U.K.). Human apotransferrin was obtained from Sigma (Poole,
Dorset, U.K.) and, when required, was saturated with ??Fe using
Fe-nitrilotriacetate (NTA; Fe?NTA molar ratio, 1:4) [16] pre-
pared from Na-NTA and ??FeCl?(NEN, Boston, MA, U.S.A.;
specific radioactivity 3–10 mCi?mg). Tf-free human serum al-
bumin, NG-monomethyl--arginine (-NMMA), LPS, leupeptin,
? 2002 Biochemical Society

Page 2

128
V. Mulero and others
PMSF, benzamidine and pepstatin A were all obtained from
Sigma. Desferrioxamine was obtained from Novartis (Horsham,
W. Sussex, U.K.). Nitrocellulose membranes and enhanced
chemiluminescence(ECL) reagents
Amersham Bioscience (Little Chalfont, Bucks., U.K.). Rabbit
anti-human Tf antibody was obtained from Dako (Glostrup,
Denmark), and rabbit anti-mouse Ft antibody (αFt) was
produced in our laboratory [16]. Mouse recombinant IFNγ was
provided kindly by Dr G. Adolf (Boehringer Ingelheim Austria
GmbH, Vienna, Austria).
wereobtainedfrom
Preparation of immune complexes
??Fe-NTA was added in a quantity sufficient to saturate the iron-
binding capacity of human apotransferrin. Then optimal propor-
tions of Tf and antiserum to Tf (normally 1 mg of Tf?2.5 ml of
antiserum) were mixed and incubated at 50 ?C for 1 h [15]. The
insoluble immune complexes (??Fe-Tf–anti-Tf) were centrifuged
at 500 g for 5 min, washed twice and resuspended in PBS.
Macrophage cultures and treatments
Bone marrow-derived macrophages (BMMφ) were obtained
from wild-type (129sv?129sv strain) and iNOS-deficient mice,
constructed as described previously [17]. It is recognized that the
iNOS-deficient mice retain some residual iNOS activity which
can be enhanced by treating cells with IFNγ?LPS [18], but NO
production is always substantially lower than in wild-type iNOS
littermates, and appropriate controls containing the NOS in-
hibitor -NMMA (500 µM) were always included. Cells from
mouse femurs were harvested, washed twice, grown for 5–7 days
in Petri dishes in the presence of 10–20% (v?v) L929 cell-
conditioned medium at 37 ?C in 5% CO?until confluent, plated
as required and incubated overnight.
Macrophages were then stimulated overnight with 50 units?ml
IFNγ and 10 ng?ml LPS in RPMI 1640 medium supplemented
with 10% heat-inactivated FCS, 100 units?ml penicillin and
100 µg?ml streptomycin, pulsed for 2 h with 10 µg?ml ??Fe-
Tf–anti-Tf immune complexes, washed twice with PBS to remove
uningested complexes, then incubated for 48 h in the presence of
the same stimulants and subjected to further procedures as
indicated below. In some experiments FCS was replaced by BSA
(0.2%) and the cells instead pulsed for 2 h with ??Fe-citrate
(1 µM Fe). Cells were fed with fresh medium containing the
appropriate stimulants every 24 h and viability checked by
Trypan Blue exclusion. Confocal microscopy showed that the
immune complexes co-localized with Lamp-1, indicating that
they had entered the late endosome and?or phagosome [19].
Iron-uptake and -release assays
Release of iron into the medium was monitored by determining
??Fe activity in the culture medium at 24 and 48 h intervals using
a LKB Compugamma counter. At the end of the experiments,
macrophages were lysed in 2% SDS and the remaining cell-
associated radioactivity determined.
Intracellular distribution of iron
Cells were lysed in 200 µl of cytoplasmic lysis buffer (1% Triton
X-100, 40 mM KCl, 25 mM Tris?HCl, pH 7.4, 50 µg?ml leu-
peptin, 200 µg?ml PMSF, 1 mM benzamidine and 50 µg?ml pep-
statin A) and the intracellular iron distribution determined as
described previously [11]. Briefly, this consisted of centrifugation
at 10000 g for 5 min to sediment the iron in insoluble material,
followed by immunoprecipitation of Ft with a rabbit anti-Ft
antibody. This allows separation of intracellular iron into three
compartments, containing insoluble material (consisting mainly
of un-degraded immune complexes and iron bound to intra-
cellular organelles), Ft-bound iron and soluble non-Ft iron.
Estimation of Ft
Macrophages stimulated as described above and pulsed if re-
quired with unlabelled Tf–anti-Tf immune complexes were lysed
with 50–100 µl of boiling lysis solution (1% SDS?10 mM Tris?
HCl,pH 7.4). Aliquots containing 10 µg ofprotein were analysed
by Western blotting using a 15% acrylamide gel and overnight
transfer at 150 mA to nitrocellulose membranes. The blots were
developed using rabbit anti-mouse Ft antibody and ECL (Amer-
sham Bioscience) according to the manufacturer’s protocol.
Membraneswerethenstainedwitha0.1%PonceauRedsolution
(Sigma) to confirm similar protein loading in all lanes.
Measurement of nitrite and protein determination
Nitrite in the culture medium was determined by the Griess
reagent [20]. The protein concentration of cell lysates was
estimated by the BCA protein assay reagent (Pierce, Chester,
U.K.) using BSA as a standard.
Statistical analysis
Data were analysed by ANOVA and unpaired Student’s t test to
determine the difference between each group. Differences were
considered statistically significant when P?0.05.
RESULTS
Uptake of iron from59Fe-Tf–anti-Tf immune complexes by BMMφ
To determine the effect of NO produced by activated macro-
phages on cellular iron handling, BMMφ from wild-type and
iNOS-deficient mice were stimulated with IFNγ and LPS,
and then pulsed with ??Fe-Tf–anti-Tf immune complexes.
Stimulation of BMMφ with IFNγ and LPS enhanced uptake
Figure 1
BMMφ
Effect of NO on uptake of59Fe-Tf–anti-Tf immune complexes by
BMMφ from iNOS-deficient mice (KO) or wild-type littermates (WT) were stimulated with
IFNγ/LPS (I/L), with or without L-NMMA (N), or left unstimulated. They were then pulsed
with59Fe-Tf–anti-Tf immune complexes for 2 h and uptake determined after removal of
uningested complexes. NO in the supernatant was measured 24 h after pulsing with the
immune complexes.
? 2002 Biochemical Society

Page 3

129
Iron metabolism in macrophages
Figure 2 Release of iron by BMMφ following uptake of59Fe-Tf–anti-Tf immune complexes
BMMφ from iNOS-deficient mice (KO) or wild-type littermates (WT) were stimulated with IFNγ/LPS (I/L), with or without L-NMMA (N), or left unstimulated, pulsed with59Fe-Tf–anti-Tf immune
complexes for 2 h and release of59Fe determined 24 or 48 h later. The inset shows the release of iron at 48 h, expressed as a percentage of ingested iron (the bars are the same as in the main
figure).
of immune complexes (P?0.001) in both wild-type and iNOS-
deficient macrophages (Figure 1), indicating that uptake was
independent of NO production. Furthermore, addition of the
iNOS inhibitor -NMMA did not affect uptake of ??Fe by
activated BMMφ.
Release of iron by BMMφ
The IFNγ?LPS-activated wild-type macrophages showed a 2–3-
fold increase (P?0.0001) in cellular iron release after 24 or 48 h
of incubation when compared with non-activated wild-type
macrophages (Figure 2). However, unlike the increased uptake
of ??Fe-Tf–anti-Tf immune complexes, this effect was blocked by
-NMMA after 24 and 48 h of incubation (P?0.01 and P?
0.0001 respectively). Furthermore, activation of iNOS-deficient
macrophages caused a much smaller increase (P?0.001) in Fe
release when compared with wild-type macrophages. This effect
was again reversed by -NMMA addition (P?0.05), suggesting
thatresidual iNOS activity was responsible for this small increase
of iron release by iNOS-deficient macrophages after activation.
If iron release is expressed as the percentage of ingested iron
rather than the absolute amount (Figure 2, inset), the results
show that in the absence of NO activated macrophages retain a
greater proportion of the ingested iron than non-activated
macrophages, but that in the presence of NO the proportion of
iron released is restored to the level in non-activated macro-
phages. Similar results were obtained with peritoneal exudate
macrophages (results not shown).
Intracellular distribution of59Fe in BMMφ following ingestion of
59Fe-Tf–anti-Tf immune complexes
We next investigated the intracellular distribution of the retained
iron in non-activated and IFNγ?LPS-activated BMMφ from
wild-type and iNOS-deficient mice, stimulated and pulsed as
Figure 3
59Fe-Tf–anti-Tf immune complexes
Intracellular distribution of iron in BMMφ following uptake of
BMMφ from iNOS-deficient mice (KO) or wild-type littermates (WT) were stimulated with
IFNγ/LPS (I/L), with or without L-NMMA (N), or left unstimulated. They were then pulsed
with59Fe-Tf–anti-Tf immune complexes for 2 h, and the uningested complexes removed. After
24 h the cells were lysed and the intracellular distribution of iron determined as described
in the Experimental section. NO in the supernatant was measured 48 h after pulsing with
the immune complexes.
above. After 24 h of incubation post-pulse, the majority (75–
85%) of iron was in all cases present in the soluble non-Ft
fraction (Figure 3). Likewise, the proportion of insoluble iron
? 2002 Biochemical Society

Page 4

130
V. Mulero and others
Figure 4Ferritin synthesis by BMMφ
BMMφ from iNOS-deficient mice (KO) or wild-type littermates (WT) were stimulated as in
Figure 1, then pulsed, if required, with unlabelled Fe-Tf–anti-Tf immune complexes (IC) for 2 h.
After a further 24 h NO release was measured in the supernatants, the cells were lysed and
Ft content determined by Western blotting. A similar protein loading in all lanes was confirmed
by staining the membranes with 0.1% Ponceau Red solution. Data are representative of two
independent experiments.
did not vary greatly, constituting about 10–13% of the total
intracellular iron. However, the proportion of iron in Ft, while
only a minor fraction in all cases, varied more noticeably
(P?0.001 by ANOVA), constituting about 10% of total
intracellular iron in non-activated macrophages but only about
4% in activated macrophages from wild-type mice. This re-
duction of Ft-bound iron in activated macrophages was not
totally due to NO, as it was not reversed completely by -
NMMA and was also observed in activated macrophages from
iNOS-deficient mice. This suggests a diversion of iron away
from the storage compartment in the more highly metabolic
activated macrophages. Therefore, treatment of macrophages
with IFNγ?LPS not only affected uptake and release of iron
acquired by phagocytosis, but also altered iron incorporation
into Ft.
Ferritin expression
Since treatment of macrophages with IFNγ and LPS altered the
proportion of iron bound to Ft (Figure 3), their effect on Ft
expression was investigated. Stimulation of both wild-type and
iNOS-deficient macrophages with IFNγ and LPS up-regulated
Ft expression (Figure 4). In the case of the wild-type macro-
phages, -NMMA increased Ft expression further, whereas it
had less effect on the iNOS-deficient cells. This suggests that
stimulation of macrophages with IFNγ and LPS has two
opposingeffectsonFtexpression:aNO-independentstimulatory
effect that is probably transcriptional [21] and an opposing
down-regulation by NO when the latter is produced in sufficient
quantity, which we found to be associated with up-regulation of
IRP1 activity (results not shown). When iron was added as Fe-
Tf–anti-Tf immune complexes there was a further modest in-
crease in Ft synthesis by wild-type macrophages. However,
overall, activation was more effective than uptake of Fe-Tf–anti-
Tf immune complexes at up-regulating Ft.
Iron uptake and release of iron from59Fe citrate by BMMφ
The ability of NO to promote Fe release by macrophages
following uptake of labelled immune complexes contrasts with
earlier findings with iron taken up from Tf via the TfR, where
NO did not increase iron release [11]. This suggests that the effect
of NO on macrophage iron is influenced by the way in which iron
is acquired by the cell. It is known that macrophages can also
acquire iron bound to low-molecular-mass chelators such as
NTA and citrate [22]. Since such complexes enter the cells by a
mechanism that probably involves direct transmembrane trans-
port, rather than by phagocytosis or via the TfR, it was of
Figure 5 Effect of NO on uptake and release of59Fe-citrate by BMMφ
(A) BMMφ from iNOS-deficient mice (KO) or wild-type littermates (WT) were stimulated as in
Figure 1, then pulsed with59Fe-citrate for 2 h in serum-free medium and uptake determined.
(B) Subsequent release of iron was then measured after a further 24 or 48 h of incubation. NO
in the supernatant was measured 48 h after incubation with59Fe-citrate.
interest to determine whether NO could influence the handling of
iron taken up from Fe-citrate. IFNγ?LPS-stimulated BMMφ
from wild-type and iNOS-deficient mice were pulsed with ??Fe-
citrate in FCS-free medium (to avoid binding of ??Fe to Tf
present in FCS). As can be seen in Figure 5, iron flux (uptake and
release) was unaffected by NO, nor directly by IFNγ?LPS
stimulation. Therefore, NO seems to affect macrophage iron
release only when iron is taken up by a phagocytic mechanism.
DISCUSSION
It is well known that mononuclear phagocytes of the liver and
spleen acquire iron as a result of erythrophagocytosis during
the normal process of removal of effete red cells. However, the
mechanisms involved in the liberation of iron by these cells in
order for it to be returned to the circulation remain unclear [1].
Furthermore, during inflammatory disease there is a tendency
for macrophages to retain more iron, which can eventually lead
to development of anaemia, i.e. the anaemia of chronic disease.
The reasons for this increased iron retention are not fully
understood, although increased Ft synthesis is probably im-
portant [23–25], and it has also been suggested that NO contri-
butes to this condition through activation of IRP1 and a
subsequent increase in TfR expression [26]. However, it is
? 2002 Biochemical Society

Page 5

131
Iron metabolism in macrophages
uncertain whether iron uptake from Tf is relevant to iron
acquisition by macrophages in ?i?o, where phagocytosis of effete
erythrocytes is likely to be the main source of iron. In an attempt
to better understand how macrophages handle scavenged iron,
we have investigated the role of NO in this process, using a
system based on phagocytosis of ??Fe-Tf–anti-Tf immune com-
plexes as a model for erythrophagocytosis. This model has the
advantage of permitting uptake of radioactive iron of high
specific activity by a phagocytic route, analogous to erythro-
phagocytosis, but without the need for haem oxygenase to
release iron from its carrier. The phagosomal?secondary lysoso-
mal routing of the immune complexes was confirmed by their co-
localization with the Lamp-1 marker [19].
To study the role of NO in macrophage iron metabolism, we
used BMMφ from mice defective in the expression of the iNOS
[17]. It was found that cells from normal mice showed an increase
in iron release following activation with IFNγ and LPS and a
pulse of ??Fe-Tf–anti-Tf immune complexes, which could be
reversed by the NOS inhibitor -NMMA. In contrast, cells from
the iNOS-deficient mice showed a much smaller increase in iron
release following activation, this residual effect probably reflect-
ing the fact that the cells from the iNOS-deficient mice can still
produce small amounts of NO. In contrast, when the cells were
loaded with ??Fe-citrate instead of immune complexes, NO had
no effect on iron release. Thus NO appears to act only when iron
is delivered by a phagocytic route, as we have shown previously
that release of iron acquired from Tf by the J774 macrophage cell
line is not increased by NO [11]. These results would suggest that
NO, instead of contributing to the hypoferraemia of inflamma-
tion, may actually have a counterbalancing effect by promoting
iron release from macrophages. Moreover, the ability of NO to
increase uptake of iron from Tf, suggested as an alternative
mechanism by which NO might contribute to the hyperferraemia
of inflammation [26], may not be tenable as there is now evidence
that TfR expression is regulated mainly by IRP2 rather than
IRP1 [27], and that IRP2 activity is decreased rather than in-
creased when macrophages are stimulated with IFNγ and LPS
[9,11,12].
Although the mechanisms involved in cellular iron release
from macrophages are not fully understood, it has generally been
assumed that iron released by macrophages must first enter the
cytoplasm from the phagosome for subsequent export across
the cytoplasmic membrane. Iron-containing material phagocy-
tosed by macrophages is very likely routed to the lysosomes,
where a combination of proteolytic activity and low pH would
release iron from carrier molecules. NO may assist in this process
by promoting formation of low-molecular-mass iron nitrosyl
complexes [28]; it is known that nitrite?nitrate can permeate into
lysosomes [29] where the low pH will reconvert the anion to NO.
In addition, haem oxygenase can remove iron from haem, which
willbepresentafterphagocytosisoferythrocytes.Theinteraction
of NO with iron-containing phagocytosed material will be
enhanced in IFNγ?LPS-activated macrophages due to slowed
membrane trafficking [30]. Iron-nitroso compounds may subse-
quently decompose [31] or perhaps could themselves serve as a
vehicle for iron release from the cell. Otherwise, subsequent
events probably involve exit of iron from the phagolysosome to
the cytoplasm, an event that may be mediated by the phagosomal
metal transporter Nramp1, although the direction of cation
transport by Nramp1 is controversial [32–34]. Once in the
cytoplasm iron may be released across the cytoplasmic mem-
brane, a process now thought to be mediated by the ferroportin
membrane iron transporter [35]. However, this model would not
explain why NO did not enhance release of iron acquired from
Fe-citrate (Figure 5B) or Fe?Tf [19].
NO can cause release of iron from Ft [13] and this might also
be a mechanism whereby macrophage iron release is increased
following activation with IFNγ?LPS. Although the proportion
of iron bound to Ft was reduced in wild-type macrophages
following IFNγ and LPS stimulation, this effect was not fully
reversed by -NMMA and a similar decrease occurred in the
iNOS-deficient macrophages, suggesting that other mechanisms
may operate. Moreover, the proportion of Fe incorporated into
Ft was low and activation of the cells increased rather than
decreased Ft levels, especially in the iNOS-deficient cells and in
the wild-type cells stimulated in the presence of -NMMA. It
therefore seems rather unlikely that increased mobilization of Ft-
bound iron is the mechanism of enhanced Fe release by NO.
An alternative possibility is that iron is released from activated
macrophages directly via a lysosomal secretory pathway in-
volving Nramp1, as we [19] and others [36] have suggested
previously. Given the potentially damaging influence of high
cytoplasmic iron on mRNA stability [37], this may prove the
safest route for recycling of iron from effete red cells. NO is
known to enhance secretory and?or excretory mechanisms in
other cells [38], which could account for its ability to enhance
iron release here. Direct extracellular secretion of iron from
phagosomal contents would also imply that most if not all iron
entering a cell by phagocytosis never enters the cytoplasm.
Loading the cells with iron via Tf–anti-Tf immune complexes
had little effect on Ft expression (Figure 4), compatible with this
proposal. Although some radioactive iron was found in Ft,
suggesting that some ingested iron can reach the cytoplasm, the
amounts are relatively small compared with those seen in J774
cells loaded with Tf [11].
In conclusion, NO appears to participate in aiding the release
of iron taken up by macrophages through phagocytosis, and may
be involved in the maintenance of iron homoeostasis. Since NO
synthesis in macrophages is induced by IFN and?or LPS, it may
therefore be of particular importance during immune and inflam-
matory responses, when iron tends to accumulate in macro-
phages. Its role may be to counterbalance the effect of other
cytokines, such as tumour necrosis factor and interleukin-1,
which promote iron retention through increased Ft synthesis.
This work was supported by grants from the Arthritis and Rheumatism Council (to
J.H.B.) and the Wellcome Trust (to F.Y.L.), and an EMBO fellowship to V.M.
REFERENCES
1 Brittenham, G. M. (1994) The red cell cycle. In Iron Metabolism in Health and
Disease (Brock, J. H., Halliday, J. W., Pippard, M. J. and Powell, L. W., eds.),
pp. 31–62, Saunders, London
Alford, C. E., King, T. E. and Campbell, P. A. (1991) Role of transferrin, transferrin
receptors, and iron in macrophage listericidal activity. J. Exp. Med. 174, 459–466
Dlaska, M. and Weiss, G. (1999) Central role of transcription factor NF-IL6 for
cytokine and iron-mediated regulation of murine inducible nitric oxide synthase
expression. J. Immunol. 162, 6171–6177
Eisenstein, R. S. and Blemings, K. P. (1998) Iron regulatory proteins, iron responsive
elements and iron homeostasis. J. Nutr. 128, 2295–2298
Bouton, C. (1999) Nitrosative and oxidative modulation of iron regulatory proteins.
Cell. Mol. Life Sci. 55, 1043–1053
Kennedy, M. C., Mende-Mueller, L., Blondin, G. A. and Beinert, H. (1992) Purification
and characterization of cytosolic aconitase from beef liver and its relationship to the
iron-responsive element binding protein. Proc. Natl. Acad. Sci. U.S.A. 89,
11730–11734
Iwai, K., Drake, S. K., Wehr, N. B., Weissman, A. M., LaVaute, T., Minato, N.,
Klausner, R. D., Levine, R. L. and Rouault, T. A. (1998) Iron-dependent oxidation,
ubiquitation, and degradation of iron regulatory protein 2: implications for degradation
of oxidized proteins. Proc. Natl. Acad. Sci. U.S.A. 95, 4924–4928
Pantopoulos, K., Weiss, G. and Hentze, M. W. (1996) Nitric oxide and oxidative
stress (H2O2) control mammalian iron metabolism by different pathways. Mol. Cell.
Biol. 16, 3781–3788
2
3
4
5
6
7
8
? 2002 Biochemical Society