Iron metabolism in heterozygotes for hemoglobin E (HbE),

-thalassemia 1, or -thalassemia and in compound heterozygotes

for HbE/-thalassemia
13
Michael B Zimmermann, Suthat Fucharoen, Pattanee Winichagoon, Pornpan Sirankapracha, Christophe Zeder,
Sueppong Gowachirapant, Kunchit Judprasong, Toshihiko Tanno, Jeffery L Miller, and Richard F Hurrell
ABSTRACT
Background: Despite large populations carrying traits for thalas-
semia in countries implementing universal iron fortification, there
are fewdata onthe absorptionandutilizationof ironinthese persons.
Objective: We aimed to determine whether iron absorption or uti-
lization (or both) in women heterozygous for -thalassemia,
-thalassemia 1, or hemoglobinE(HbE) differedfromthat incontrol
subjects and compound HbE/-thalassemia heterozygotes.
Design: In Thai women (n 103), red blood cell indexes, iron
status, non-transferrin-bound iron, and growth differentiation factor
15 were measured, and body iron was calculated. Fractional iron
absorption was measured from meals fortified with isotopically la-
beled (
57
Fe) Fe sulfate, and iron utilization was measured by the
infusion of (
58
Fe) Fe citrate.
Results: Iron utilization was 15% lower in -thalassemia 1 or
-thalassemia heterozygotes than in controls. When corrected for
differences in serum ferritin, absorption was significantly higher in
the - and -thalassemia groups, but not the HbE heterozygotes,
than in controls. HbE/-thalassemia compound heterozygotes had
lower iron utilization and higher iron absorption and body iron than
did controls. Nontransferrin-bound iron and growth differentiation
factor 15 were higher in the compound heterozygotes, but not in the
other groups, than in the controls.
Conclusions: In -thalassemia 1 and -thalassemia heterozygotes
with ineffective erythropoesis, dietary iron absorption is not ade-
quately down-regulated, despite a modest increase in body iron
stores. In populations with a high prevalence of these traits, a pro-
gram of iron fortification could include monitoring for possible
iron excess and for iron deficiency. Am J Clin Nutr 2008;88:
102631.
INTRODUCTION
Globally, iron excess occurs mainly in persons with geograph-
ically specific genetic mutations that permit the absorption from
the diet of more iron than is physiologically needed. Two main
types of hereditary iron overload are well recognized: 1) hered-
itary hemochromatosis, which is seen in populations derived
from northern Europe, and 2) the thalassemias and related he-
moglobinopathies of South and Southeast Asia, the Middle East,
and the Mediterranean (1). Thalassemia mutations are extremely
common (2): up to 25% of Thai people are carriers of
-thalassemia, and, in regions of Thailand, Laos, and Cambodia,
up to 60% of people are carriers of hemoglobin E (HbE), a
hemoglobinopathy caused by a mutation of the -globin gene. In
southern China, which has a population of 350 million, 5% of
people are carriers for -thalassemia and 4% are carriers for
-thalassemia or HbE (3).
Heterozygotes for -thalassemia 1, -thalassemia, and HbE
typically are asymptomatic and have mild microcytic, hypochro-
mic anemia. In contrast, in thalassemia homozygotes and com-
pound heterozygotes such as HbE/-thalassemia compound het-
erozygotes, ineffective erythropoiesis in an expanded marrow
stimulates iron absorption even if iron stores are adequate, and
this stimulation increases the risk of iron excess when transfu-
sions are given (4, 5). If heterozygotes have some degree of
ineffective erythropoesis andabsorbmore dietaryiron, theymay,
to a lesser degree than homozygotes and compound heterozy-
gotes, also be at risk of iron excess. In heterozygotes for hemo-
chromatosis, one study reported that iron absorption froma meal
with added iron was 3-fold that in controls (6); another study
found no differences in absorption (7).
Despite large populations that are heterozygous for thalasse-
mia in countries implementing iron fortification, there are few
data on the absorption and utilization of iron in these persons.
Earlier studies in -thalassemia heterozygotes are difficult to
interpret because they used oral solutions of
59
Fe given with
ascorbic acid that overestimate dietary iron absorption and made
comparisons without adjustment for differences in iron status (4,
811). In countries where thalassemias are common, public
1
From the Laboratory for Human Nutrition, Swiss Federal Institute of
Technology Zurich, Zurich, Switzerland (MBZ, CZ, and RFH); the Thalas-
semia ResearchCenter, Institute of Science andTechnologyfor Researchand
Development (SF and PS) and the Institute of Nutrition (PW, SG, and KJ),
Mahidol University, Salaya, Nakon Pathom, Thailand; and the Molecular
Medicine Branch, National Institute of Diabetes and Digestive and Kidney
Diseases, National Institutes of Health, Bethesda, MD (TT and JLM).
2
Supported by SUSTAIN (Washington, DC), The Global Alliance for
Improved Nutrition (Geneva, Switzerland), the International Atomic Energy
Agency (Vienna, Austria), and the Swiss Federal Institute of Technology
(ETH) Zurich (Zurich, Switzerland). SF is a senior research scholar of The
Thailand Research Fund.
3
Reprints not available. Address correspondence to MB Zimmermann,
Laboratory for Human Nutrition, Swiss Federal Institute of Technology
(ETH) Zurich, Schmelzbergstrasse 7, LFV E19, CH-8092 Zurich, Switzer-
land. E-mail: michael.zimmermann@ilw.agrl.ethz.ch.
Received May 14, 2008.
Accepted for publication July 3, 2008.
1026 Am J Clin Nutr 2008;88:102631. Printed in USA. 2008 American Society for Nutrition

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

health officials are hesitant to adopt iron fortification because of
concerns about possible iron excess in thalassemia carriers.
Therefore, our study aim was to determine whether iron metab-
olismabsorption and utilization (incorporation into erythro-
cytes)in heterozygotes for -thalassemia, -thalassemia 1, or
HbE differs from that in controls with normal hemoglobin A
(HbA) or in symptomatic compound HbE/-thalassemia het-
erozygotes.
SUBJECTS AND METHODS
Subjects
The study was carried out in Thai women recruited from the
western suburbs of Bangkok. Inclusion criteria were age 1850
y; premenopausal status; body weight 65 kg; self-reported
nonpregnancy and no plans for pregnancy; no chronic medical
illnesses; and desired Hb type: heterozygotes for -thalassemia;
heterozygotes for -thalassemia 1; HbE heterozygotes; com-
pound HbE/-thalassemia heterozygotes; and normal HbA.
None of the subjects were taking medicinal iron at entry into the
study. Twelve percent of the women, distributed through the 5
groups, were taking oral contraceptives. Sample size calcula-
tions indicated that 24 women should be included in the control
group and the groups heterozygous for -thalassemia,
-thalassemia 1, andHbE. This number was basedon80%power
to detect a 50% difference in iron absorption compared with the
control subjects, an SD of 8.2% for log-transformed absorption
data from previous absorption studies with the same meal and
iron compound in a similar population of Thai women, and a type
I error rate of 5%. We also studied 9 compound heterozygotes for
HbE/-thalassemia, a disorder with a clinical phenotype; these
persons exhibit clearly ineffective erythropoesis and were there-
fore studied as positive controls (4, 5).
Written informed consent was obtained fromall subjects. Eth-
ical approval for the study was given by ethical review commit-
tees at Mahidol University (Salaya, Nakon Pathom, Thailand)
and the Swiss Federal Institute of Technology (ETH, Zurich,
Switzerland).
Iron absorption and utilization study
Iron absorption was estimated by using stable-isotope tech-
niques in which the incorporation of
57
Fe and
58
Fe into erythro-
cytes is measured 14 d after administration (12, 13). If the sub-
jects were taking vitamin or mineral supplements or any other
medications, administration was stopped for 2 wk before the
study and until the final venipuncture. On day 1, a baseline
venous blood sample was drawn after an overnight fast for de-
termination of isotopic composition and confirmation of iron
status. The subjects then received a test meal (rice with vegetable
soup; see below) labeledwith
57
Fe as ferrous sulfate infishsauce,
which was fed under standardized conditions and close supervi-
sion. Each test meal contained 3 mg labeled
57
Fe. One hour later,
2 mL of an aqueous solution containing 100 g
58
Fe as iron
citrate was taken into a syringe containing 10 mL of 0.9% saline
solution and, via a 100-mL infusion bag leading into a 0.9%
saline drip, slowly infused over 50 min (7, 14). The rate of
intravenous infusionof ironwas basedonthe estimated2g/min
plasma appearance of iron normally absorbed from the gastro-
intestinal tract (7). No intake of food and fluids was allowed for
4hafter the test meal intake. Fourteendays later, a secondvenous
blood sample was drawn.
Preparation of isotopically labeled iron
57
Fe was prepared fromisotopically enriched
57
Fe by dilution
in 0.1 mol H
2
SO
4
/L (12). The solution was kept refrigerated in
plastic containers, under argonatmosphere. Ironcitrate, enriched
with
58
Fe, was prepared for intravenous infusion fromelemental
58
Fe according to the method previously described (14). The
solution was divided in ampoules containing 100 g Fe, steril-
ized, and checked for pyrogens. Enrichment of isotopic labels
was 95.5%for
57
Fe and93.1%for
58
Fe. The isotopic composition
of the stable-isotope labels was measured by using negative
thermal ionizationmass spectrometry (13). Iron-isotope solu-
tions were divided into aliquots and sterilized at the Zurich Uni-
versity Hospital Pharmacy (Zurich, Switzerland).
Test meal composition
The test meal was composed of steamed white rice (50 g dry
weight; Jasmine perfume rice; Dragon Phoenix Brand, Bangkok,
Thailand), which was served with a vegetable soup containing
white cabbage, Chinese cabbage, Thai mushrooms, and steamed
carrots in 120 mL of water and seasoned with fish sauce. All
ingredients were purchased in bulk and used for the entire study.
The food portions were kept frozen until use, and each portion
was microwaved on the day of feeding.
Laboratory analysis
Whole blood was transported on ice to the thalassemia labo-
ratory at Mahidol University. Complete blood count and reticu-
locyte count was done by using the Advia 120 Hematology
System with 3-level Advia 120 TESTpoint hematology controls
(Bayer, Singapore). Hemoglobin typing for -globin abnormal-
ity was done by using HPLC (Variant Hemoglobin Testing Sys-
tem; BioRad, Hercules. CA) with calibrators and controls pro-
vided by the manufacturer. DNA analysis for -globin
abnormalities was done by using a GeneAmp PCR System (Ap-
plied Biosystem, Foster City, CA) and a Gel Doc 2000 Gel
Documentation System (BioRad, Hercules, CA). Serum ferritin
(SF) and serum transferrin receptor (TfR) measurements were
made by using an enzyme-linked immunosorbent assay (ELISA;
Ramco Laboratories Inc, Stafford, TX) on microplates (CO-
DATM Open Microplate System; BioRad); normal ranges: SF,
12300 g/L; TfR, 8.2 mg/L. Serum erythropoietin was mea-
sured by using an ELISA (R&D Systems, Minneapolis, MN)
with controls provided by the manufacturer; the normal range is
3.316.6 IU/L. Serumiron (SI), total ironbinding capacity, and
transferrin saturation were measured by using colorimetric spec-
trophotometry (UV-Vis Spectrophotometer 8453; Hewlett-
Packard, PaloAlto, CA); the normal range is 929mol/Lfor SI,
4570 mol/L for total ironbinding capacity, and 3050% for
transferrin saturation. High-sensitivity C-reactive protein was
measured by using latex nephelometry (BNProSpec System;
Dade Behring Limited, Tokyo, Japan); the normal range is 3
mg/L. SF data from one subject with a CRP concentration 10
(a HbE/-thalassemia compound heterozygote) were excluded
from the data analysis. For measurement of plasma NTBI,
plasma (450 L) was incubated with 800 mmol nitrilotriac-
etate/L, pH7.0 solution (50 L), at roomtemperature for 30 min.
Plasma proteins were then removed by centrifugation of the
IRON METABOLISM IN THALASSEMIA 1027

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

treated plasma with the use of an ultracentrifugation filtration
device (30-kDa cutoff, polysulfone type, NanoSep; Pall Life
Sciences, Ann Arbor, MI) at 12 000 rpm (10620 g; Hettich
Centrifugation, Bach, Germany) and 15 C for 45 min. The ul-
trafiltrate was then analyzed by HPLC (15). Nontransferrin-
bound iron (NTBI) concentrations in healthy subjects using this
method are 0 mol/L. Serum growth differentiation factor 15
(GDF-15) was measured in blinded samples by using a DuoSet
ELISA(R&DSystems, Minneapolis, MN) accordingtothe man-
ufacturers protocol. Whole blood was mineralized by micro-
wave digestion, and iron was separated by anion-exchange chro-
matography and a subsequent solvent-solvent extraction step
into diethylether. Iron was analyzed by negative thermal ioniza-
tionmass spectrometry with a magnetic sector field mass spec-
trometer (Finnigan MAT 262; Thermo Finnigan, Bremen, Ger-
many) equipped with a multicollector system for simultaneous
ion beam detection; isotopic dilution calculations were done as
described previously (13).
Statistical analysis
Data were analyzed by using PRISM (version 3; GraphPad,
San Diego, CA) and EXCEL (XP 2002; Microsoft, Redmond,
WA) software. Body iron stores (BFe) were calculated from the
ratio of TfR to SF (TfR:SF) (16). The amount of
57
Fe and
58
Fe
label present in the blood was calculated from isotope dilution.
Circulating iron was calculated from the blood volume and he-
moglobin concentration. The amount of stable isotope adminis-
tered was used to calculate the fractional iron incorporation into
erythrocytes (13). The absorption of the oral iron was calculated
by dividing the percentage of erythrocyte incorporation of the
oral dose by the fractional erythrocyte incorporation of the in-
travenous dose (7, 14). Results were presented as geometric
means SDs. Analysis of variance with unpaired t tests and a
Bonferroni correction were used to compare the incorporation
and absorption data between the groups, and values for iron
incorporation and absorption were logarithmically transformed
before statistical analysis. In the overall analysis of variance and
in the Bonferroni correction, P 0.05 and P 0.005, respec-
tively, were considered significant.
RESULTS
Age, anthropometric data, and hematologic indexes by group
are shown in Table 1. There were no significant between-group
differences in age, height, or weight, except that the HbE group
was significantly heavier than the normal hemoglobin and HbE/
-thalassemia groups (P0.05). Comparedwithcontrols, the -
and -thalassemia and HbE heterozygote groups had higher
RBC concentrations and lower mean corpuscular volume and
mean corpuscular hemoglobin (P 0.05). The - and
-thalassemia groups had significantly lower hemoglobin than
did the control subjects (P 0.02), and most of these heterozy-
gotes were mildly anemic. The HbE/-thalassemia group had
significantly lower RBC, hemoglobin, and mean corpuscular
hemoglobin and significantly greater RBC distribution width
than did the other groups (P 0.01 for all).
The iron status indicators are shown in Table 2. The HbE
heterozygotes did not differ from the control subjects in any of
the iron indexes. The -thalassemia 1 heterozygotes had signif-
icantly (P 0.05) higher SF, BFe, and erythropoietin (but not
NTBI) than did control subjects. The -thalassemia heterozy-
gotes had significantly (P 0.05) higher SF, TfR, and BFe (but
not NTBI) than did control subjects. The -thalassemia het-
erozygotes had slightly higher NTBI than did control subjects,
but this difference was not significant. The HbE/-thalassemia
group had significantly higher serumiron, transferrin saturation,
SF, TfR, BFe, erythropoietin, NTBI, andGDF-15concentrations
than did the other groups (P 0.01 for all).
As shown in Table 3, the incorporation of iron into erythro-
cytes (utilization) from the intravenous iron dose was signifi-
cantly (P 0.05) lower in the - and -thalassemia groups (but
not the HbE group) than in control subjects. Uncorrected frac-
tional iron absorption from the test meal was slightly but signif-
icantly (P0.05) lower in the -thalassemia heterozygotes than
in the other 4 groups. When corrected for differences in SF,
absorption was significantly (P 0.05) higher in the - and
-thalassemia groups (but not the HbE heterozygotes) than in
control subjects. The HbE/-thalassemia group had significantly
higher absorption and lower incorporation than did the other
TABLE 1
Age, anthropometric variables, and hematologic indexes in Thai women by hemoglobin (Hb) type
1
Normal Hb HbE trait
-Thalassemia 1
trait
-Thalassemia
trait HbE/-thalassemia
Subjects (n) 25 26 18 27 9
Age (y) 32.7 7.8
2
32.1 8.4 32.8 11.3 36.7 9.8 36.0 10.3
Weight (kg) 47.8 7.3
a
53.5 6.3
b
50.8 8.8
a,b
50.4 8.4
a,b
45.4 10.0
a
Height (cm) 154.5 5.4 154.1 5.7 155.8 4.4 153.5 5.7 152.6 10.0
Red blood cells (10
6
cells/mL) 4.47 0.33
a
4.78 0.46
b
5.31 0.44
c
5.31 0.34
c
3.74 0.85
d
Hb (g/dL) 13.0 1.0
a
12.4 1.2
a
11.6 1.0
a,b
11.1 0.7
b
6.9 1.4
c
Range (g/dL) 9.714.1 9.415.8 10.113.8 9.812.6 4.58.8
Subjects with Hb 12.0 g/dL (n) 3 7 13 23 9
Mean corpuscular volume (fL) 83.8 5.8
a
74.0 6.1
b
66.6 3.4
c
63.2 4.0
c
61.2 6.7
c
Mean corpuscular Hb (pg) 29.2 2.4
a
26.2 2.6
b
21.7 1.0
c
21.2 1.6
c
18.5 1.8
d
Red blood cell distribution width (%) 14.4 1.4
a
15.2 1.4
a,b
15.4 0.7
a,b
15.8 0.8
b
25.8 2.7
c
1
Values in a row without common superscript letters are significantly different: P 0.001 (ANOVA) and P 0.05 (unpaired t tests with Bonferroni
correction).
2
x SD (all such values)
1028 ZIMMERMANN ET AL

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

groups (P 0.02 for both). The relation between GDF-15 and
erythrocyte incorporation of
58
Fe is shown in Figure 1.
In a stepwise regression of all the hematologic variables in
Tables 1 and 2 on erythrocyte incorporation of
58
Fe, only hemo-
globin was a significant positive predictor, and SF and GDF-15
were negative predictors (Table 4). In a stepwise regression of
these variables on fractional absorption of
57
Fe, significant neg-
ative and positive predictors were BFe and the red blood cell
distribution width, respectively (Table 4).
DISCUSSION
Early ferrokinetic studies in -thalassemia homozygotes
reported that erythrocyte incorporation of iron was as low as
15%, whereas it was 7590% in control subjects (4,811). In
subjects with -thalassemia intermedia, iron absorption from
labeled meals varied from 26% to 54%, a range that is sharply
higher than values in control subjects (4). These studies and
others (17) suggested that greater iron absorption and lower
utilization in -thalassemia homozygotes resulted in large
increases in body iron that exceeded iron-binding capacity
(18, 19). At the same time, there appeared to be a certain
component of functional iron deficiency in subjects with
thalassemia, because excess iron stores were not entirely
available for erythropoiesis (2022).
If thalassemia heterozygotes partiallyexpress the homozygote
phenotype (ineffective erythropoiesis and higher plasma iron
turnover) and absorb significantly more dietary iron, they may
accumulate iron and be susceptible to damage from excess body
iron. In previous studies reporting SF or serum iron concentra-
tions (or both) in persons who were heterozygous for
-thalassemia, most foundthat SFconcentrations were similar to
or modestly higher than those in healthy persons (2329). Het-
erozygotes for -thalassemia alsohave higher erythropoietinand
TfR concentrations (30, 31) than do control subjects, which in-
dicates a modest increase in erythropoietic drive. Whereas TfR
concentrations are higher than those in control subjects, they are
lower than those in subjects with iron-deficiency anemia, which
suggests ineffective erythropoiesis (32). Previous studies of iron
absorption in thalassemia heterozygotes from the 1960s and
1970s were done by using oral solutions of
59
Fe given with
ascorbic acid (4, 811), a method that significantly overesti-
mates true dietary iron absorption (33, 34). Those studies pro-
duced mixed results, and a review(8) concluded that mild eryth-
ropoetic hyperplasia in heterozygotes for -thalassemia does not
influence iron absorption, and, if iron absorption increases, that
change is due to concurrent iron deficiency. The findings of the
present study differ. In women heterozygous for - or
-thalassemia, erythrocyte iron incorporation was 15%lower,
whereas serum TfR or erythropoietin concentrations (or both)
TABLE 2
Iron indexes and biochemical variables in Thai women by hemoglobin (Hb) type
1
Normal Hb HbE trait -Thalassemia 1 trait -Thalassemia trait HbE/-thalassemia
Serum iron (mmol/L) 15.6 4.8
2,a
18.4 7.0
a
13.1 4.5
a
14.4 4.1
a
28.3 6.8
b
Total ironbinding capacity
(mmol/L)
50.4 4.9
a
52.6 7.1
a
42.3 6.6
b
43.9 5.3
b
36.3 3.3
c
Transferrin saturation (%) 30.7 7.8
a
34.1 10.2
a
30.4 6.8
a
32.5 7.0
a
77.8 16.6
b
Serum ferritin (g/L)
3
15 (1148)
a
23 (3112)
a,b
28 (1142)
b
52 (2226)
c
877 (2064040)
d
Serum transferrin receptor
(mg/L)
3
6.2 (4.315.9)
a
6.7 (4.013.2)
a
7.5 (5.010.2)
a,b
9.0 (5.413.9)
b
16.3 (32.656.5)
c
Body iron status (mg Fe/kg
body wt)
4
2.1 (11.4 to 10.5)
a
2.9 (5.0 to 8.5)
a
5.3 (9.8 to 9.6)
b
5.6 (7.8 to 11.0)
b
10.8 (3.918.4)
c
Serum erythropoetin (IU/L)
3
4.8 (0.930.6)
a
5.6 (1.822.1)
a,b
7.0 (2.926.6)
b
6.2 (1.033.6)
a,b
58.9 (19.2154.9)
c
Nontransferrin-bound iron
(mol/L)
4
0.0 (0.8 to 3.1)
a
0.0 (0.8 to 2.4)
a
0.0 (0.4 to 0.8)
a
0.2 (0.4 to 5.5)
a
2.5 (1.26.6)
b
GDF-15 (pg/mL)
4
398 (2211132)
a
321 (157872)
a
454 (2011328)
a
414 (190596)
a
15 059 (186361 779)
b
1
GDF-15, growth differentiation factor 15. Values in a row without common superscript letters are significantly different: P 0.01 (ANOVA) and P
0.05 (unpaired t tests with Bonferroni correction).
2
x SD (all such values).
3
Geometric x; range in parentheses (all such values).
4
Median; range in parentheses (all such values).
TABLE 3
Iron absorption and erythrocyte iron utilization in Thai women by hemoglobin (Hb) type
1
Normal Hb HbE trait -Thalassemia 1 trait -Thalassemia trait HbE/-thalassemia
%
Erythrocyte
58
Fe incorporation 93.1 (82.7, 103.5)
a
90.2 (79.5, 100.9)
a,b
78.8 (68.4, 89.2)
b,c
80.1 (70.0, 90.3)
c
21.2 (9.3, 33.1)
d
Uncorrected fractional
57
Fe absorption 8.7 (3.2, 23.4)
a
7.5 (3.4, 16.7)
a
5.6 (2.8, 11.3)
a,b
4.5 (2.0, 9.7)
b
10.5 (4.6, 23.9)
a
Corrected fractional
57
Fe absorption 3.0 (0.3, 5.8)
a
4.0 (4.6, 12.6)
a,b
6.3 (1.9, 10.8)
b
6.6 (4.2, 15.9)
b
100
c
1
All values are geometric x; SD, SD in parentheses. Iron absorption in each subject was corrected to a value corresponding to 40 g serum ferritin/L
(30). Values in a row without common superscript letters are significantly different: P 0.01 (ANOVA) and P 0.05 (unpaired t tests with Bonferroni
correction).
IRON METABOLISM IN THALASSEMIA 1029

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

were significantly higher, which indicated ineffective erythro-
poesis and increased erythropoetic drive. At the same time, iron
absorption was less well regulated by the increase in iron stores
and was approximately twice that expected for the level of iron
stores in persons with normal hemoglobin. This higher iron ab-
sorption resulted in modest increases in storage iron as reflected
by significantly higher concentrations of SF and BFe.
Both heterozygotes and homozygotes for HbE, the most com-
mon form of -thalassemia, are asymptomatic and minimally
anemic, and they have microcytic and hypochromic red blood
cells. However, when the
E
allele interacts with a -thalassemia
mutation in the compound heterozygous state, a variable and
often severe anemia is produced (35) that has evidence of inef-
fective erythropoiesis and shortened red blood cell survival (5,
36). In the women in the present study who were heterozygous
for HbE, erythrocyte incorporation of
58
Fe and dietary iron ab-
sorption were similar to those in control subjects. However, the
HbE/-thalassemia compound heterozygotes showed sharply
lower erythrocyte iron incorporation and significantly higher
iron absorption, body iron, and plasma NTBI.
Hepcidin concentrations should be high in iron-loaded per-
sons with -thalassemia (37); however, hepcidin concentrations
are low in these persons, unless they have recently received a
transfusion (30, 38, 39). Production of GDF-15 by the expanded
erythroid compartment contributes to iron overload in thalasse-
mia by inhibiting hepcidin gene expression (40). Compared with
the concentration in control subjects, The concentrations of se-
rumGDF-15 in persons with -thalassemia syndromes is 100-
foldthose incontrol subjects, but theyare not significantlyhigher
in heterozygotes for - and -thalassemia (40). Our data support
these findings in that, in the HbE/-thalassemia compound het-
erozygotes, serum GDF-15 was sharply higher (40-fold) than
in control subjects, whereas GDF-15 in heterozygotes for HbEor
- or -thalassemia was not higher than that in control subjects.
Circulating GDF-15 was a significant negative predictor of
erythrocyte incorporationof iron, whichsuggests that ineffective
incorporation of iron may stimulate the synthesis of GDF-15 by
the developing erythron.
The present study has several limitations. First, we did not
measure hepcidin concentrations in our subjects. Also, we stud-
ied only young Thai women, and our results may not be gener-
alizable to other population groups, such as older men or groups
with differing ethnicities or diets (or both). Finally, estimates of
body iron (16) in persons with ineffective erythropoesis may not
be valid; however, if we had used SF alone, we would have
obtained similar results and relations, and SF concentrations
were used to correct the iron absorption values in Table 3.
Although the prevention of iron deficiency through fortifica-
tion of foods is a recommended strategy (41, 42), there may be
potential health risks associated with an overabundant iron sup-
ply (43). Because of fears over possible iron overload, iron for-
tification of flour has been discontinued in Sweden and Denmark
(44, 45). Moreover, in developing countries with a high preva-
lence of hemoglobinopathies (eg, Egypt and Cambodia), public
health officials are reluctant to adopt iron fortification (P
Winichagoon, RHurrell, personal communication, 2008). In this
context, what is the significance of our findings? In HbE het-
erozygotes, the most common hemoglobinopathy in Thailand,
iron absorption and utilization do not differ significantly from
those in control subjects, which suggests that additional dietary
iron fromuniversal fortification may be beneficial in populations
with low dietary iron intakes. In HbE/-thalassemia compound
heterozygotes, iron absorption and body iron are sharply higher
andutilizationis impaired, whichsuggests that additional dietary
iron would not be beneficial. But these persons typically are
followed clinically, and potential iron excess is monitored and
treated. In heterozygotes for - or -thalassemia, iron utilization
is lower and absorption is not appropriately down-regulated,
despite modestly higher concentrations of SFand storage iron. In
y = 35.806x + 179.65
R
2
= 0.6553
0
20
40
60
80
100
120
2 2.5 3 3.5 4 4.5 5
log GDF-15 (pg/mL)
(
%
)
FIGURE 1. The relation between erythrocyte incorporation of
58
Fe and
log growth differentiation factor 15 (GDF-15) in Thai women (n 103) who
had normal hemoglobin A; who were heterozygous for -thalassemia,
-thalassemia 1, or hemoglobin E (HbE); or who were compound heterozy-
gous for HbE/-thalassemia.
TABLE 4
Stepwise regressions of the hematologic variables in Tables 1 and 2 on the dependent variables of erythrocyte incorporation of
58
Fe and fractional
absorption of
57
Fe
1
SE Standardized P
Erythrocyte incorporation of
58
Fe
Hemoglobin 4.423 0.749 0.393 0.0001
Serum ferritin 10.249 1.663 0.338 0.0001
GDF-15 14.220 2.789 0.336 0.0001
Fractional absorption of
57
Fe
Body iron stores 0.041 .007 0.563 0.0001
Red blood cell distribution width 0.041 .011 0.358 0.0001
1
GDF-15, growth differentiation factor 15. Subjects were Thai women (n 103) who had normal hemoglobin A; were heterozygous for -thalassemia,
-thalassemia 1, or hemoglobin E (HbE); or were compound heterozygous for HbE/-thalassemia. Unstandardized and standardized coefficients are shown.
1030 ZIMMERMANN ET AL

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

regions with a high prevalence of these traits, iron should be
targeted to groups vulnerable to iron deficiency, such as women
and children. If universal iron fortification is implemented, it
may be useful to monitor iron stores in groups with lower iron
turnover, such as adult males and postmenopausal women.
We thank the subjects for their participation in the study and the nursing
staff at Mahidol University; we also thank Napaporn Reabroy (Nakon
Pathom, Thailand) and Adam Krzystek (Zurich, Switzerland) for laboratory
assistance.
The authors responsibilities were as followsMBZ, SF, PW, CZ, and
RFH: designed the research; all authors: performed the research and partic-
ipated in the editing of the manuscript; MBZ, SF, and PW: analyzed the data;
and MBZ: wrote the first draft of the manuscript. None of the authors had a
personal or financial conflict of interest.
REFERENCES
1. Gordeuk VR, McLaren GD, Samowitz W. Etiologies, consequences,
and treatment of iron overload. Crit Rev Clin Lab Sci 1994;31:89133.
2. Weatherall DJ, Clegg JB. Inherited haemoglobin disorders: an increas-
ing global health problem. Bull World Health Organ 2001;79:70412.
3. Angastiniotis M, Modell B. Global epidemiology of haemoglobin dis-
orders. Ann N Y Acad Sci 1998;850:25169.
4. Pippard MJ, Callender ST, Warner GT, Weatherall DJ. Iron absorption
and loading in beta-thalassaemia intermedia. Lancet 1979;2(8147):819-
21.
5. Pootrakul P, Sirankapracha P, Hemsorach S, et al. A correlation of
erythrokinetics, ineffective erythropoiesis, and erythroid precursor ap-
optosis in Thai patients with thalassemia. Blood 2000;96:260612.
6. Lynch SR, Skikne BS, Cook JD. Food iron absorption in idiopathic
hemochromatosis. Blood 1989;74:218793.
7. Roe MA, Heath AL, Oyston SL, et al. Iron absorption in male C282Y
heterozygotes. Am J Clin Nutr 2005;81:81421.
8. Heinrich HC, Gabbe EE, Oppitz KH, et al. Absorption of inorganic and
food iron in children with heterozygous and homozygous beta-
thalassemia. Z Kinderheilkd 1973;115:122.
9. Pippard MJ, Callender ST, Warner GT, Weatherall DJ. Iron absorption
in iron-loading anaemias: effect of subcutaneous desferrioxamine infu-
sions. Lancet 1977;2:7379.
10. Finch CA, Deubelbeiss K, Cook JD, et al. Ferrokinetics in man. Medi-
cine (Baltimore) 1970;49:17-53.
11. Pippard MJ, Weatherall DJ. Iron absorption in non-transfused iron load-
ing anaemias: prediction of risk for iron loading, and response to iron
chelation treatment, in beta thalassaemia intermedia and congenital sid-
eroblastic anaemias. Haematologia (Budap) 1984;17:17-24.
12. Kastenmayer P, Davidsson L, Galan P, Cherouvrier F, Hercberg S,
Hurrell RF. A double stable isotope technique for measuring iron ab-
sorption in infants. Br J Nutr 1994;71:41124.
13. Walczyk T, Davidson L, Zavaleta N, Hurrell R. Stable isotope labels as
a tool to determine the iron absorption by Peruvian school children from
a breakfast meal. Fresenius J Anal Chem 1997;359:4459.
14. Dainty FR, Roe MA, Teucher B, Eagles J, Fairweather-Tait SJ. Quan-
tification of unlabeled nonheme iron absorption in human subjects: a
pilot study. Br J Nutr 2003;90:5036.
15. Singh S, Hider RC, Porter JB. A direct method for quantification of
non-transferrin-bound iron. Anal Biochem 1990;186:3203.
16. Cook JD, Flowers CH, Skikne BS. The quantitative assessment of body
iron. Blood 2003;101:335963.
17. Fiorelli G, Fargion S, Piperno A, Battafarano N, Cappellini MD. Iron
metabolism in thalassemia intermedia. Haematologica 1990;75:8995.
18. Walter PB, Fung EB, Killilea DW, et al. Oxidative stress and inflam-
mation in iron-overloaded patients with beta-thalassaemia or sickle cell
disease. Br J Haematol 2006;135:25463.
19. Esposito BP, Breuer W, Sirankapracha P, Pootrakul P, Hershko C,
Cabantchik ZI. Labile plasma iron in iron overload: redox activity and
susceptibility to chelation. Blood 2003;102:26707.
20. Wickramasinghe SN, Hughes M, Wasi P, Fucharoen S, Litwinczuk
RAC. Ultrastructure and cell cycle distribution of erythropoietic cells in
heterozygotes and homozygotes for haemoglobin E. Br J Haematol
1984;57:68594.
21. Wickramasinghe SN, Fucharoen S, Wasi P. Studies of erythropoietic
cells in heterozygotes and homozygotes for haemoglobin Constant
Spring and in heterozygotes for both haemoglobin Constant Spring and
-thalassaemia 1 trait: extent of globin chain precipitation and cell cycle
distribution. Clin Lab Haematol 1986;8:18798.
22. Pootrakul P, Wattanasaree J, Anuwatanakulchai M, Wasi P. Increased
red blood cell protoporphyrin in thalassemia: a result of relative iron
deficiency. Am J Clin Pathol 1984;82:28993.
23. Mazza U, Saglio G, Cappio FC, Camaschella C, Neretto G, Gallo E.
Clinical and haematological data in 254 cases of beta-thalassaemia trait
in Italy. Br J Haematol 1976;33:919.
24. Economidou J, Augustaki O, Georgiopoulou V, Vrettou H, Parcha S,
Loucopoulos D. Assessment of iron stores in subjects heterozygous for
beta-thalassaemia based on serum ferritin levels. Acta Haematol 1980;
64:2058.
25. White JM, Richards R, Jelenski G, Byrne M, Ali M. Iron state in alpha
and beta thalassaemia trait. J Clin Pathol 1986;39:2569.
26. Qureshi TZ, Anwar M, Ahmed S, Ahmed Khan D, Saleem M. Serum
ferritin levels in carriers of beta-thalassaemia trait. Acta Haematol 1995;
94:79.
27. ParfreyPS, Barnett M, Sachs JA, PollockDJ, Turnbull AL. Ironoverload
in beta-thalassaemia minor. Afamily study. Scand J Haematol 1981;27:
294302.
28. Fargion S, Piperno A, Panaiotopoulos N, Taddei MT, Fiorelli G. Iron
overload in subjects with beta-thalassaemia trait: role of idiopathic hae-
mochromatosis gene. Br J Haematol 1985;61:48790.
29. Mehta BC, Pandya BG. Iron status of beta thalassemia carriers. Am J
Hematol 1987;24:13741.
30. Origa R, Galanello R, Ganz T, Giagu N, Maccioni L, Faa G, Nemeth E.
Liver iron concentrations and urinary hepcidin in beta-thalassemia.
Haematologica 2007;92:5838.
31. Tassiopoulos T, Konstantopoulos K, Tassiopoulos S, Rombos Y,
Alevizou-Terzaki V, Kyriaki P, Loukopoulos D. Erythropoietin levels
and microcytosis in heterozygous beta-thalassaemia. Acta Haematol
1997;98:1479.
32. Demir A, Yarali N, Fisgin T, Duru F, Kara A. Serumtransferrin receptor
levels in beta-thalassemia trait. J Trop Pediatr. 2004;50:36971.
33. Hallberg L, Hulten L. Methods to study dietary iron absorption in
manan overview. In: Halberg L, Asp NG, eds. Iron nutrition in health
and disease. London, United Kingdom; John Libbey &Co, 1996:81-95.
34. Cook JD, Dassenko SA, Lynch SR. Assessment of the role of nonheme-
iron availability in iron balance. Am J Clin Nutr 1991;54:71722.
35. Fucharoen S, Ketvichit P, Pootrakul P, Siritanaratkul N, PiankijagumA,
Wasi P. Clinical manifestation of beta-thalassemia/hemoglobin E dis-
ease. J Pediatr Hematol Oncol 2000;22:5527.
36. Rees DC. Hemoglobin F and hemoglobin E/beta-thalassemia. J Pediatr
Hematol Oncol 2000;22:56772.
37. Ganz T. Hepcidina regulator of intestinal iron absorption and iron
recycling by macrophages. Best Pract Res Clin Haematol 2005;18:171
82.
38. Papanikolaou G, Tzilianos M, Christakis JI, et al. Hepcidin in iron
overload disorders. Blood 2005;105:41035.
39. Kattamis A, Papassotiriou I, Palaiologou D, et al. The effects of eryth-
ropoetic activity and iron burden on hepcidin expression in patients with
thalassemia major. Haematologica 2006;91:80912.
40. Tanno T, Bhanu NV, Oneal PA, et al. High levels of GDF15 in thalas-
semia suppress expression of the iron regulatory protein hepcidin. Nat
Med 2007;13:1096101.
41. WHO/UNICEF/UNU. Iron deficiency anaemia: assessment, preven-
tion, and control. A guide for programme managers. Geneva, Switzer-
land: World Health Organization 2001. (WHO/NHD/01.3, 2001.)
42. Hurrell R. How to ensure adequate iron absorption from iron-fortified
food. Nutr Rev 2002;60:S7-15.
43. Adams PC, Reboussin DM, Barton JC, et al. Hemochromatosis and Iron
Overload Screening (HEIRS) Study Research Investigators. Hemochro-
matosis and iron-overload screening in a racially diverse population.
N Engl J Med 2005;352:176978.
44. Hallberg L, Hulthen L. Perspectives on iron absorption. Blood Cells Mol
Dis 2002;29:56273.
45. Milman N, Byg KE, Ovesen L, Kirchhoff M, Jurgensen KS. Iron status
in Danish women, 1984-1994: a cohort comparison of changes in iron
stores and the prevalence of iron deficiency and iron overload. Eur J
Haematol 2003;71:5161.
IRON METABOLISM IN THALASSEMIA 1031

b
y

g
u
e
s
t

o
n

J
u
l
y

8
,

2
0
1
4
a
j
c
n
.
n
u
t
r
i
t
i
o
n
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m