The effects of iron fortification on the gut microbiota in African children:
a randomized controlled trial in Co ˆte d’Ivoire1–4
Michael B Zimmermann, Christophe Chassard, Fabian Rohner, Elie ´zer K N’Goran, Charlemagne Nindjin, Alexandra Dostal,
Ju ¨rg Utzinger, Hala Ghattas, Christophe Lacroix, and Richard F Hurrell
Background: Iron is essential for the growth and virulence of many
pathogenic enterobacteria, whereas beneficial barrier bacteria, such
as lactobacilli, do not require iron. Thus, increasing colonic iron
could select gut microbiota for humans that are unfavorable to the
Objective: The objective was to determine the effect of iron forti-
fication on gut microbiota and gut inflammation in African children.
Design: In a 6-mo, randomized, double-blind, controlled trial, 6–
14-y-old Ivorian children (n = 139) received iron-fortified biscuits,
which contained 20 mg Fe/d, 4 times/wk as electrolytic iron or
nonfortified biscuits. We measured changes in hemoglobin concen-
trations, inflammation, iron status, helminths, diarrhea, fecal calpro-
tectin concentrations, and microbiota diversity and composition (n =
60) and the prevalence of selected enteropathogens.
Results: At baseline, there were greater numbers of fecal entero-
bacteria than of lactobacilli and bifidobacteria (P , 0.02). Iron
fortification was ineffective; there were no differences in iron status,
anemia, or hookworm prevalence at 6 mo. The fecal microbiota was
modified by iron fortification as shown by a significant increase in
profile dissimilarity (P , 0.0001) in the iron group as compared
with the control group. There was a significant increase in the
number of enterobacteria (P , 0.005) and a decrease in lactobacilli
(P , 0.0001) in the iron group after 6 mo. In the iron group, there
was an increase in the mean fecal calprotectin concentration (P ,
0.01), which is a marker of gut inflammation, that correlated with
the increase in fecal enterobacteria (P , 0.05).
Conclusions: Anemic African children carry an unfavorable ratio of
fecal enterobacteria to bifidobacteria and lactobacilli, which is in-
creased by iron fortification. Thus, iron fortification in this population
produces a potentially more pathogenic gut microbiota profile, and
this profile is associated with increased gut inflammation. This trial
was registered at controlled-trials.com as ISRCTN21782274.
J Clin Nutr 2010;92:1406–15.
Iron fortification can be an effective strategy to control iron-
deficiency anemia in developing countries (1, 2), and the food-
stuffs most often used for mass fortification are cereal flours.
Worldwide, the most commonly used fortificants for flours are
elemental iron powders such as hydrogen-reduced iron or
electrolytic iron despite their low bioavailability (absorption of
these poorly soluble forms of iron is often as low as ,2–3%) (1).
Flour-fortification programs are in place or in the planning
stages in 78 countries (3), including in one-quarter of the pop-
ulation of sub-Saharan Africa (4). Co ˆte d’Ivoire has mandated
the addition of electrolytic iron to wheat flour (decree 025 issued
on 18 January 2007). Low absorption of iron fortificants results
in .90% of the iron passing unabsorbed into the colon. Most
iron in the human body is tightly bound to various proteins that
limit iron supply to potential pathogens, and during infection,
iron supply is sharply reduced in the extracellular compartment
(5). But there is no similar system for the sequestration of di-
etary iron in the gut lumen. Iron is a growth-limiting nutrient for
many gut bacteria, and multiple strains vigorously compete for
unabsorbed dietary iron in the colon because colonization de-
pends on the ability of the bacteria to acquire iron and other
essential growth nutrients (6).
Although iron is an essential nutrient for most of the gut
microbiota, some beneficial barrier bacteria, such as lactobacilli,
pathogens butdo not require iron (7). In contrast, for most enteric
gram-negative bacteria (eg, Salmonella, Shigella, or pathogenic
Escherichia coli), iron acquisition plays an essential role in the
virulence and colonization of most pathogenic strains (8). In
animal studies, increasing doses of iron produced a linear in-
crease in diarrhea incidence and coliform number, and the ratios
1From the Institute of Food, Nutrition and Health, Swiss Federal Institute of
Technology (ETH) Zurich, Zurich, Switzerland (MBZ, CC, FR, AD, CL, and
RFH); the Division of Human Nutrition, Wageningen University, Wageningen,
Netherlands (MBZ); the Centre Suisse de Recherches Scientifiques, Abidjan,
Co ˆte d’Ivoire (EKN and CN); the UFR Biosciences, Universite ´ de Cocody-
Abidjan, Abidjan, Co ˆte d’Ivoire (ENK); the Department of Public Health and
Epidemiology, Swiss Tropical Institute, Basel, Switzerland (JU); and the Lon-
don School of Hygiene and Tropical Medicine, London, United Kingdom (HG).
2The sponsors of the study played no role in the design of the trial, analyses
of data, or preparation of the manuscript.
3Supported by the Medicor Foundation (Vaduz, Liechtenstein), the Swiss
National Science Foundation (Bern, Switzerland), the Swiss Foundation for
Research in Nutrition (Zurich, Switzerland), and the Swiss Federal Institute
of Technology (ETH) Zurich (Zurich, Switzerland). Escherichia coli
O157H45 was provided by Roger Stephan.
4Address correspondence to M Zimmermann, Laboratory for Human Nutri-
tion, Swiss Federal Institute of Technology (ETH) Zurich, Schmelzbergstrasse
7, LFV E19, CH-8092 Zurich, Switzerland. E-mail: michael.zimmermann@ilw.
Received April 29, 2010. Accepted for publication September 24, 2010.
First published online October 20, 2010; doi: 10.3945/ajcn.110.004564.
Am J Clin Nutr 2010;92:1406–15. Printed in USA. ? 2010 American Society for Nutrition
by guest on June 6, 2011
Supplemental Material can be found at:
of bifidobacteria and lactobacilli to enterobacteria were mod-
ified (9). These ratios can be an index of gut health because
higher ratios provide greater resistance to infection (10).
Therefore, an increase in unabsorbed dietary iron in humans
through fortification or supplementation could modify the co-
lonic microbiota equilibrium and favor the growth of pathogenic
strains over barrier strains. This would be an important adverse
effect because diarrheal disease is the cause of death of ’1 in 6
children ,5-y-old in sub-Saharan Africa (11). A recent sys-
tematic review suggested that iron supplementation was asso-
ciated with a small increase in diarrheal disease (12). A World
Health Organization (WHO) Consultation, which interpreted the
Pemba study in which untargeted iron supplementation in-
creased mortality in children (13), cautioned that it is unclear
whether risks of iron are specific to malaria or whether they
apply to other infections, including sepsis from enteric bacteria
(14). Thus, our study objective was to determine the effect of
a poorly absorbed iron fortificant on gut microbiota, gut hel-
minthes, and gut inflammation in African children in an area
with high rates of diarrheal disease. Specifically, our hypotheses
were that 1) iron fortification would increase the ratio of fecal
enterobacteria to bifidobacteria and/or lactobacilli, 2) favor the
colonization by potentially pathogenic strains, and 3) increase
SUBJECTS AND METHODS
This study was nested within a larger 2 · 2 · 2 intervention
trial that tested the interactions of the intermittent treatment of
malaria, antihelminthic treatment, and iron fortification (15).
The study population consisted of 6–14-y-old school children in
rural central Co ˆte d’Ivoire with a high infectious disease burden
(16). The main dietary staples in the study region are rice and
yam. The usual wet season is from March to November. The
study was done during one school year from November 2006 to
July 2007. For the effect of iron on hemoglobin concentrations,
with the assumption of a mean (6SD) hemoglobin concentra-
tion of 117 6 12 g/L (17), to detect a hemoglobin concentration
increase of 8 g/L, 65 children were needed in each group to
achieve a power of 90% at a 5% level of significance. For the
microbiota substudy, we estimated 30 subjects per group would
be adequate for comparisons of the dominant bacteria on the
basis of previous studies (18, 19). Ethical approval was given
by the Swiss Federal Institute of Technology (ETH) Zurich
(Zurich, Switzerland) and the Ministry of Health in Co ˆte
d’Ivoire. Written informed consent was obtained from parents or
legal guardians of children who participated in the study. The
larger trial in which this study was nested was registered at
controlled-trials.com as ISRCTN21782274. Dietary iron intake
was assessed by 3-d weighed food records at the midpoint of the
intervention in 24 households and compared with recommended
dietary intakes (20).
Inclusion criteria were, for girls, nonpregnant (self-reported)
and, for boys and girls, no major chronic illnesses, no use of iron-
containing supplements, and anticipated local residence for the
study duration. Children were randomly assigned to either the
iron-fortification group or the control, no-iron group. The iron
group received 2 fortified biscuits (Midor AG, Meilen, Swit-
zerland) [electrolytic Fe (A-131; Dr Lohmann GmbH, Emmer-
thal, Germany), 20 mg Fe/d per child] 4 times/wk; the control
group consumed identical but unfortified biscuits. The biscuits
were made from low-extraction wheat flour and contained (per
100 g) 454 kcal, 76 g carbohydrate (16.5 g sucrose), 1.5 g total
fiber, 14 g fat, and 6.3 g protein. The iron content of biscuits was
confirmed at the Swiss Federal Institute of Technology Zurich by
using atomic absorption spectrometry (Spectra AA-50; Varian,
Palo Alto, CA).
Active and passive case detections for diarrhea and other
gastrointestinal illness were carried out throughout the study. For
active detection, teachers administered a health questionnaire
monthly with a 2-d recall period. If an illness was reported, the
child was examined by the study physician. Diarrhea was defined
as ?3 loose, watery stools in a day (21). For the passive re-
porting, parents and guardians were encouraged to refer children
free of charge to the local health center as soon as they presented
a symptom or illness.
Blood and stool variables were assessed at baseline and after 6
mo. Hemoglobin concentrations were measured in whole blood
with an AcT8 Counter (Beckman Coulter; Krefeld, Germany) on
the day of blood sampling. Plasma was divided into aliquots,
transported frozen, and stored at –25?C. Zinc protoporphyrin
(ZPP) concentrations were measured on washed red blood cells
with a hematofluorometer (Aviv Biomedical; Lakewood, NJ) ?7 d
after sampling. Plasma ferritin (PF) and C-reactive protein
(CRP) concentrations were measured with an automated chemi-
luminescent immunoassay system (IMMULITE; Diagnostic
Products Corporation, Los Angeles, CA). a1-Acid glycoprotein
(AGP) concentrations were measured by immunoturbidimetry
(Cobas Mira; Roche Diagnostics, Rotkreuz, Switzerland). Soluble
transferrin receptor (TfR) concentrations were measured by
using an automated immunonephelometric assay (Cobas Integra
800; Roche Diagnostics). Anemia was defined according to the
WHO (22); iron deficiency was defined as a PF concentration
,30 lg/L or TfR concentration .8.2 mg/L and ZPP concen-
tration .40 lmol/mol heme (22, 23). Systemic inflammation
was defined as an AGP concentration .1.2g/L or CRP con-
centration .10 mg/L.
A single, fresh, morning stool samplewas collected at baseline
and after 6 mo. Two Kato-Katz thick smears (41.7 mg) were
prepared from each stool sample according to standard protocols
(24), and hookworm eggs were counted microscopically. The
slides were reexamined, and the number of eggs of Ascaris
lumbricoides, Schistosoma mansoni, and Trichuris trichiura
were counted. For the conversion to eggs per gram of feces,
a multiplication factor of 24 was used. The remaining stool was
divided into aliquots and stored at 270?C until further analyses.
Fecal calprotectin concentrations were measured by using an
immunoassay (Eurospital SpA, Trieste, Italy) and expressed as
micrograms per gram.
Stratified by sex, 30 children were randomly selected from
each group for the gut microbiota analyses. None of the children
had unusual diet habits or received antibiotics in the 3 mo before
of antibiotics prescribed did not differ between the 2 groups (P =
0.942); no child included in the gut microbiota analyses was
IRON AND GUT MICROBIOTA
by guest on June 6, 2011
given antibiotics in the 3 wk before the endpoint assessment.
Fecal samples were collected from 0900 to 1130 at the school;
children were given prelabeled beakers with lids and were asked
to provide a stool sample. The beakers were immediately placed
in an ice chest with cooling elements and taken to the local
laboratory at the end of each morning. Stool samples were split
into aliquots, and the aliquot for gut microbiota analysis was
frozen immediately at 230?C until analyses.
Totalbacterial DNAwas extracted from fecal samples (200 mg
feces) with a Fast DNA SPIN kit (MP Biomedicals, Illkirch,
France) according to the manufacturer’s instructions. DNA
concentrations were measured with a NanoDrop ND-1000
Spectrophotometer (Witec AG, Littau, Switzerland) at a wave-
length of 260 nm, and samples were stored at 224?C until further
Analyses of gut microbiota by polymerase chain reaction
and temporal temperature gradient electrophoresis
Primers HDA1-GC and HDA2 were used to amplify the
variable regions 2 and 3 of the bacterial 16S ribosomal RNA
(rRNA) genes (Table 1) and investigate the whole bacterial di-
versity. A GC-rich sequence (5# CCC CCC CCC CCC CGC
CCC CCG CCC CCC GCC CCC GCC GCC C 3#) was added to
the 5# end of the reverse primer. Reaction tubes contained 1 lL
fecal DNA, 25 lL polymerase chain reaction (PCR) Master Mix
(2·) (Fermentas, Nunninnen, Switzerland), and 0.4 lL of each
primer in a final volume of 50 lL. One microliter of fecal DNA
was used for standardization because preliminary tests showed
only a small variability in DNA concentrations extracted from 1 lL
fecal DNA (in a concentration range of 100–300 ng/lL) and
that this variability did not affect the temperature gradient
electrophoresis (TGGE) profiles (see supplemental Figure 1
under “Supplemental data” in the online issue). PCR amplifi-
cations were performed by using the following conditions: an
initial DNA denaturation and enzyme activation at 94?C for 4 min,
30 cycles consisting of denaturation (30 s at 94?C), annealing
(30 s at 58?C), elongation (1 min at 68?C), and a final elongation
at 68?C for 7 min. PCR product concentrations and sizes were
estimated by using 2% agarose gel electrophoresis that con-
tained ethidium bromide (0.1 ng ethidium bromide/mL), in 1·
TBE. The Dcode universal mutation detection system (Bio-Rad
Laboratories, Reinach, Switzerland) was used for the sequence-
specific separation of amplicons. These amplicons were loaded
in a 1-mm polyacrylamide gel that consisted of 9% (vol:vol)
polyacrylamide (vol:vol, acrylamide-bisacrylamide, 37.5: 1) and
8 mol urea/L with 1.5 · TAE as the electrophoresis buffer. A
prerun of 15 min at a constant voltage of 20 V preceded a run at
65 V. The temperature of the gel system was programmed to
increase by 0.3?C/h from 66?C to 70?C. In addition, similarly
obtained PCR products from known bacterial strains were
loaded to allow standardization of the band migration and gel
curvature between different gels. This ladder consisted of DNA
of the following organisms listed in migration order: Bacter-
oides thetaiotaomicron [Deutsche Sammlung von Mikroorganismen
(DSM) 2079; http://www.dsmz.de/], Lactobacillus acidophilus
(DSM 20079), Roseburia intestinalis (DSM 14610), E. coli
(American Type Culture Collection 25288; http://www.atcc.org/),
and Bifidobacterium longum (DSM 20219).
The investigation of the Lactobacillus community was per-
formed by 2 successive nested PCRs (26); the first was per-
formed with the primers Bact-0011f and Lab-0677r (Table 1).
The reaction mixtures (50 lL) consisted of 25 lL PCR Master
Mix (2·) (Fermentas), 1 lmol primer Bact-0011f/L, 1 lmol
primer Lab-0677r/L, and ’1 lL bacterial DNA. The PCR
products, which were used as template for the second PCR, were
firstly purified with a QIAquick PCR Purification Kit (Qiagen
AG, Basel, Switzerland). To prevent the amplification of re-
sidual genomic DNA or low-yield aspecific amplicons formed
in the first PCR, purified samples were diluted to a concentra-
tion of 15 ng/lL. The second nested PCR was performed with
the universal primers Bact-0124-GCf and Univ-0515r. The re-
action mixtures (50 lL) consisted of 25 lL PCR Master Mix
(2·) (Fermentas), 1 lmol primer Bact-0124-GCf/L, 1 lmol
primer Univ-0515r/L, and 1lL of the previously amplified
eluted DNA samples. PCR product concentrations and sizes
were estimated by using 2% agarose gel electrophoresis that
contained ethidium bromide (0.1 ng ethidium bromide/mL) in
1· TBE. The Dcode universal mutation detection system was
also used for the sequence-specific separation of amplicons.
These amplicons were loaded in a 1-mm polyacrylamide gel that
consisted of 8% (vol:vol) polyacrylamide (vol:vol, acrylamide-
bisacrylamide, 37.5: 1) and 7 mol urea/L with 1.5 · TAE as the
electrophoresis buffer. A prerun of 15 min at a constant voltage
of 20 V preceded a run at 76 V. The temperature of the gel
system was programmed to increase by 0.2?C/h from 64?C to
Gels were stained with ethidium bromide solution (2.5 mg
ethidium bromide/L; Fluka, Buchs, Switzerland) for 45 min,
system; Alpha Innotech Corporation, San Leandro, CA). Gel
patterns were analyzed with GelCompar II software (Applied
Primers used for polymerase chain reaction and temporal temperature gradient electrophoresis analyses of gut microbiota
Target group Primer and sequence (5#–3#) Reference
All bacteria HDA1-GC, GACTCCTACGGGAGGCAGCAG1
1GC clamp: CGC CGG GGG CGC GCC CCG GGC GGG GCG GGG GCA CGG GGG G.
ZIMMERMANN ET AL
by guest on June 6, 2011
Maths NV, Sint-Martens-Latem, Belgium). Each band was iden-
tified and normalized on the basis of the migration of the marker.
Baseline and endpoint samples from the same volunteer were
always comigrated side by side for accurate comparisons.
Analyses of gut microbiota by quantitative real-time PCR
PCR amplification and detection were performed with an ABI
PRISM 7500-PCR sequence-detection system (Applied Bio-
systems, Zug, Switzerland). Each reaction mixture of 25 lL was
composed of 2· SYBR Green PCR Master Mix (Applied Bio-
systems), with each of the specific primers at a concentration of
0.20 lM, and 1 lL template DNA diluted to DNA concen-
trations of 5 or 0.5 ng/lL depending on the targeted bacterial
group. The quantitative real-time PCR conditions were kept at
the presettings of the ABI PRISM 7500-PCR sequence-detection
system (Applied Biosystems) with an initial heating step of 2
min at 50?C and a denaturation step of 10 min at 95?C followed
by 40 amplification cycles at 95?C for 15 s and 60?C for 1 min.
The fluorescent products were detected during the second step of
each cycle. A melting-curve analysis was performed after am-
plification to evaluate the specificity of the primers and the
quality of the PCR. Primers used in this study are listed in Table
2. They were synthesized and purified by Microsynth (Micro-
synth, Balgach, Switzerland). Amplified 16S rRNA genes from
the following bacteria served as standard templates: pLME21
plasmid that contained Bifidobacterium lactis 16S ribosomal
DNA (rDNA) for universal primers (34), B. thetaiotaomicron
(DSM 2079) for Bacteroides spp. primers, Lactobacillus del-
brueckii (DSM20081) for Lactobacillus/Leuconostoc/Ped-
iococcus spp., E. coli O157H45 (Laboratory of Food
Biotechnology culture collection, www.bt.ilw.agrl.ethz.ch/) for
Enterobacteriaceae primers, and Salmonella Thyphimurium
N15 (Laboratory of Food Biotechnology culture collection) for
Salmonella primers. For each primer pair, a standard curve was
analyzed in the same run and used for the calculation of the
number of 16S rDNA or gene copies detected in each sample.
Because the copy number of 16S rDNA genes varies according
to the species considered, the data for Bacteroides, Lactobacil-
lus, and Enterobacteriaceae enumeration were normalized by
using the different mean copy numbers of 4.63, 5.54 and 7.0,
respectively, referenced in the Ribosomal RNA Database (35,
36) Normalization was not necessary for Bifidobacterium spe-
cies enumeration because they possess a single copy of the xfp
gene used as the target (37). Therefore, results were expressed in
cell numbers standardized by dividing absolute copy numbers
obtained by the mean copy number for that bacterial group.
Detection of bacterial gut pathogens by PCR
Pathogenic strains of E. coli, such as enteropathogenic E.
coli (EPEC), enteroaggregative E. coli (EAEC), Shiga toxin-
producing E. coli (STEC), and enteroinvasive E. coli (EIEC) as
well as Shigella spp. were detected by using the primers of
Aranda et al (38) for multiplex PCRs (Table 2). Each PCR
mixture of 25 lL contained PCR Master Mix (2·) (Fermentas),
1 lmol/L of each primer, and 1 lL DNA. PCRs were carried out
with an initial heating step for 3 min at 95?C and 35 amplifi-
cation cycles of 95?C for 30 s, 58?C (for EPEC and EAEC
primers) or 50?C (for EIEC and STEC primers) for 30 s, and 72?
C for 1 min followed by a final elongation step at 72?C for 7
min. PCR products were visualized by separation in a 2% aga-
rose gel and stained in ethidium bromide. Amplified 16S rRNA
genes from the following bacteria served as positive controls: E.
coli O157H45 (Laboratory of Food Biotechnology culture col-
lection) for EPEC, EAEC, STEC, and EIEC detection and
Shigella flexneri (American Type Culture Collection 29903) for
Primers used for quantitative real-time polymerase chain reaction (PCR) and multiplex PCR analyses of gut microbiota1
Target groupPrimer and sequence (5#–3#) Method used Reference
All bacteriaEub338 F, ACTCCTACGGGAGGCAGCAG
F_Lacto 05, AGCAGTAGGGAATCTTCCA
R_Lacto 04, CGCCACTGGTGTTCYTCCATATA
InvA 139, GTGAAATTATCGCCACGTTCGGGCAA
InvA 141, TCATCGCACCGTCAAAGGAACC
Eae 1, CTGAACGGCGATTACGCGAA
Eae 2, CCAGACGATACGATCCAG
Enteropathogenic E. coli
Enteroaggregative E. coli
Shiga toxin-producing E. coli
Enteroinvasive E. coli and Shigella spp.
1E. coli, Escherichia coli.
IRON AND GUT MICROBIOTA
by guest on June 6, 2011
Data analyses were done with SPSS software (version 16.0;
SPSS Inc, Chicago, IL) and Instat 3.0 GraphPad (Graph Pad
Sofware Inc, La Jolla, CA). Analyses were done by using a per
protocol approach. Nonnormally distributed variables were
1- or 2-factor analyses of variance were used, with post hoc t
tests. Because the lactobacilli numbers in the 2 groups were
significantly different at baseline, we analyzed these data with
analysis of covariance by using the baseline values as covariates.
Rate ratios of gastrointestinal illness episodes were calculated
with Poisson regression (95% CI) relative to the control group.
Linear regression was used to determine associations between
variables. Repeated-measures analysis by using general linear
models was used to compare changes in the binary variables
during the intervention. Comparisons of TGGE profiles were
performed using the Dice similarity coefficient (Dsc) analysis
based entirely on the results of band classification. Dsc values
were compared on the basis of the presence or absence of bands.
The Dice coefficient was defined as follows:
Dsc ¼ ½2j=ða þ bÞ?3100
where j is the number of common bands between samples A and
B, and a and b are the total number of bands in samples A and B,
respectively. The distance between 2 TGGE profiles was then
Distance ¼ 1002Dsc
This coefficient ranged from 0 (no common bands) to 1 (identical
band patterns) (19). Significance was set at P , 0.05.
At baseline, 74 children were assigned to the iron group, and
73 children were assigned to the control group. During the study,
there were 4 dropouts in the iron group (2 children dropped out
because of relocation out of the study area, one child developed
diabetes, and one child refused to consume the biscuits) and 4
dropouts in the control group (3 children dropped out because of
relocation out of the study area, and one child developed
splenomegaly); thus, 70 and 69 children completed the study.
There were no significant differences in baseline biomedical
characteristics between iron and control groups (Table 3). In
subjects who participated in the gut microbiota study (n = 30 in
each group), there were no significant differences in the varia-
bles shown in Table 3 between the 2 groups at baseline (data not
shown) with the exception of PF without inflammation, the
median (range) concentrations of which were 107 (22–270) and
61 (27–173) lg/L in the subgroups of the control and iron
At baseline in all children, 54% of children were infected with
helminths (of these, 93% of helminths were hookworm, 4% of
helminths were T. trichiura, and 3% of helminthes were A.
lumbricoides). The geometric mean infection intensity of
hookworm was 108.8 eggs/g feces (95% CI: 81.2, 145.7 eggs/g
feces); no child exceeded the WHO cutoff values for light in-
fection. The hookworm prevalence and egg burden increased in
both groups during the intervention without significant differ-
ences between groups (Table 3). At the endpoint, there were no
T. trichiura or A. lumbricoides detected in any of the subjects in
Estimated daily mean (6SD) iron intakes before the in-
troduction of the iron fortificant were 12.2 6 3.8 mg Fe for
children ,10 y of age, 14.5 6 3.5 mg Fe for boys .10 y of age,
and 13.7 6 2.9 mg for girls .10 y of age. Assuming a 10%
dietary iron bioavailability, this translates into 138%, 94%, and
44% of the requirements for absorbed iron (20). During the trial,
the mean (6SD) number of biscuits consumed was 165 6 21
out of 200 biscuits offered (83% compliance), and compliance
between groups was comparable (P = 0.79). The iron content of
the fortified biscuits was 9.6 6 0.3 mg Fe. Thus, the total ad-
ditional iron consumed by children who received fortified bis-
cuits was 1585 6 217 mg electrolytic Fe, which corresponded to
8.8 6 1.2 mg fortificant Fe/d; therefore, the iron intervention
increased the overall dietary iron intake by ’60–70%.
The iron fortificant was very poorly absorbed; anemia and iron
status worsened during the 6-mo study period, and there was no
and ZPP) or storage iron (PF) (Table 3). Compliance was .94%
for monitoring of self-reported illness via the monthly ques-
tionnaire. There was no significant difference in gastrointestinal
illness between the groups: the rate ratios (95% CIs) of diarrhea,
constipation, and vomiting in the iron group relative to the
control were 1.0 (0.6, 1.4), 0.8 (0.5, 1.1), and 1.4 (0.8, 2.5),
Iron fortification modified the gut microbiota: in the TGGE
profiles of PCR amplicons (V2–V3 region) from the fecal
samples, iron fortification caused an increase in bacterial dis-
similarity compared with the control group (32.3 6 12.5%
compared with 15.0 6 7.5%; P , 0.0001) (Figure 1, A and B),
which showed a higher variation rate in bacterial diversity after
iron fortification. However, the mean number of bands in the
TGGE profiles that compared iron and control groups was not
significantly different at 6 mo (Figure 1). In the analyses using
Lactobacillus group-specific primers, there was no significant
change in the presence of Lactobacillus species with iron for-
tification. Although each volunteer in the iron group showed
a microbiota profile at 6 mo that strongly differed from that at
baseline (see Figure 2 under “Supplemental data” in the online
issue for examples), a specific band pattern related to iron for-
tification was not detected.
At baseline in all 60 children, there were significantly greater
mean (6SD) cell numbers (log cell number/g feces) of enter-
obacteria (7.86 6 0.87) than bifidobacteria (7.42 6 0.59) (P ,
0.02) or lactobacilli (6.49 6 1.14) (P , 0.001) (Figure 2).
During the intervention, there was no significant change in
numbers of total bacteria at 0 and 6 mo: (control: 11.29 6 0.67
and 11.28 6 0.81, respectively; iron: 11.62 6 0.94 and 11.78 6
0.59, respectively), Bacteroides (control, 10.01 6 0.67 and
10.23 6 0.70, respectively; iron, 10.13 6 0.70 and 10.31 6
0.85, respectively) or bifidobacteria (Figure 3). However, there
was an increase in enterobacteria in the iron group (P , 0.005)
(Figure 3) and a reduction in lactobacilli (P , 0.0001). The
enterobacteria population includes many of the enteric patho-
gens, and 26.6% of the children had positive samples for
ZIMMERMANN ET AL
by guest on June 6, 2011
Shigella spp. and enteroinvasive E. coli (these could not be
distinguished with IpaH1 and IpaH2 primers) and/or Salmonella
spp. at baseline but at low amounts (generally ?103bacteria/g
feces). Of these, Salmonella was the predominant bacterial
pathogen and was observed in 78.6% of positive samples,
whereas Shigella/enteroinvasive E. coli were detected in 21.4%
of positive samples. After 6 mo, more children were positive for
Salmonella in the iron group than in the control group (23.3%
compared with 16.6% of children, respectively), but this dif-
ference was not significant. There were no significant correla-
tions between baseline SF, TfR, or ZPP concentrations and
baseline numbers of enterobacteria, bifidobacteria, or lactoba-
cilli, nor were there significant correlations between changes
during the study in SF, TfR, or ZPP concentrations and changes
in the number of enterobacteria, bifidobacteria, or lactobacilli.
Systemic and gut inflammation
During the intervention, there was no difference between
groups in systemic inflammation determined by AGP or CRP
concentrations (Table 3). There was also no significant correla-
values) in systemic inflammation measures (AGP or CRP) and
change in the numbers of enterobacteria, lactobacilli or bifido-
bacteria during the intervention. In contrast, there was a signifi-
cant increase in the mean calprotectin concentration in fecal
samples from the iron group at 6 mo than in the control group
(P , 0.01) (Figure 3). There was no significant association
between changes in fecal calprotectin concentrations and changes
in numbers of bifidobacteria, lactobacilli, or hookworm during
the intervention, but there was a positive correlation between
change in fecal calprotectin concentrations and changes in
numbers of enterobacteria (r = 0.32, P , 0.05).
Although each child in the iron group consumed ’1.5 g Fe
fortificant over the course of the study, there was no improve-
ment in iron status or anemia. Iron status deteriorated over the
course of the study in both groups, likely because of limited
food availability caused by the prolonged dry season in Co ˆte
d’Ivoire in 2007. Iron absorption was poor for the following
reasons: 1) elemental iron powders, such as electrolytic iron,
have low bioavailability because of their low solubility (1); 2)
Baseline and endpoint characteristics by treatment group1
Variable Study monthControlIron
Female sex [n (%)]
9.1 6 2.2
110.8 6 9.42
106.6 6 9.7
5.9 (1.7– 21.0)
123.4 (80.8, 188.6)7
385.1 (245.9, 602.9)
9.5 6 2.3
110.7 6 10.5
107.1 6 9.4
96.5 (63.5, 146.8)
380.3 (252.1, 573.9)
Anemia [n (%)]3
Plasma ferritin without inflammation (lg/L)
Serum transferrin receptor (mg/L)
Erythrocyte zinc protoporphyrin (lmol/mol heme)
Iron deficiency [n (%)]5
C-reactive protein (mg/L)
a1-Acid glycoprotein (g/L)
Systemic inflammation [n (%)]6
Hookworm [n (%)]
Hookworm infection intensity (EPG)
1EPG, eggs per gram of feces. All data are for positive individuals. There were no significant effects of iron treatment
on binary variables (repeated-measures analysis). For continuous variables, there were no significant time · treatment
interactions (2-factor ANOVA).
2Mean 6 SD (all such values).
3Hemoglobin concentrations ?80 and ?115 g/L for children ,12 y of age and ?80 and ?120 g/L for children ?12 y
4Median; range in parentheses (all such values).
5Plasma ferritin concentrations ,30 lg/L or zinc protoporphyrin concentrations .40 lmol/mol heme and transferrin
receptor concentrations .8.5 mg/L.
6a1-Acid glycoprotein concentrations .1.2 g/L or C-reactive protein concentrations .10 mg/L.
7Geometric mean; 95% CI in parentheses (all such values).
IRON AND GUT MICROBIOTA
by guest on June 6, 2011
fractional absorption of dietary iron is inversely related to body
iron stores, but only 15% of children were iron deficient at
baseline; 3) nearly 1 in 4 children had signs of systemic in-
flammation that may have reduced iron absorption and/or use
(39). Our findings are consistent with previous trials of elec-
trolytic iron in African populations that were not effective (40,
41). Thus, it is likely that nearly all of the iron fortificant passed
into the colon and was potentially available for the gut micro-
In this study, we combined TGGE and PCR to assess and
compare the changes in the dominant gut microbiota and other
specific populations during the intervention. This approach has
proven useful to show nutritional or pathologic modulation of the
gut microbiota (18, 19, 42). The baseline composition of the gut
microbiota in these Africanchildren was markedly different from
that reported in European populations (43–45). In humans, the
total enterobacteria population represents a subdominant pop-
ulation of ’106to 107bacteria/g feces (43–45). In this study, we
observed higher numbers of enterobacteria in African children
with estimated population amounts close to 108bacteria/g feces
at baseline. Thus, this population was already high and dominant
at baseline and was further increased by ’0.5 log with iron
fortification. After iron fortification, the children harbored, on
average, one hundred million more enterobacteria per gram of
feces, which was a substantial increase. At baseline, 1 in 4
children had fecal samples that contained potential pathogens
(mainly Salmonella), and there were greater numbers of enter-
obacteria than lactobacilli and bifidobacteria (Figure 2). This is
a striking reversal of the usual high ratios of lactobacilli and
bifidobacteria to enterobacteria observed in healthy white chil-
dren and adults (43–45). The microbiota profile in our children
was likely the result of chronic contamination of the local food
and/or water supply, and although it was not associated with
clear clinical disease, the high numbers of enterobacteria at
baseline may have made these children more susceptible to
colonization by enteropathogens (46) when colonic iron was
abundant. In healthy humans, high species diversity provides
ecologic stability (47, 48) so that after infancy, the composition
of the intestinal microbiota at the species level is remarkably
stable (49). This characteristic stability was perturbed by iron
fortification; TGGE analyses showed profound differences in
the dominant gut bacterial species at baseline compared with the
endpoint for children in the iron group. However, although the
iron group showed a microbiota profile at 6 mo that differed
from that at baseline, a specific band pattern related to iron
fortification was not detected, which suggests that the overall
effects of iron fortification on the gut microbiota balance were
host specific. Alternatively, our methods may have been of in-
sufficient resolution to identify consistent effects on the less
dominant members of the bacterial community.
Iron fortification resulted in an increase in the numbers of
enterobacteria and a decrease in the lactobacilli population. The
expansion of the enterobacteria was likely mainly due to the
increased growth of commensal and nonpathogenic E. coli, and
this may be important because abundances of closely related
species can predict the susceptibility to intestinal colonization
by pathogenic bacteria. In a recent study by Stecher et al (46),
the presence of high commensal E. coli densities in animals
correlated with higher amounts of Salmonella colonization.
Iron fortification favored the growth of enterobacteria over
lactobacilli, and this was likely due to their differing iron
requirements. Most enteric gram-negative pathogens, including
Salmonella spp., E. coli, Shigella spp. (8), take up iron-side-
rophore complexes via specific outer-membrane receptors. In
vitro, enteric bacteria display increased virulence in situations of
increased iron availability (50), and iron transporter [ferrous iron
transporter, protein B (FeoB)]–mediated ferrous iron acquisition
is required for bacterial virulence (51) and gastrointestinal tract
colonization (52). Thus, it is possible that more soluble forms of
iron, such as ferrous sulfate, could have a greater effect on en-
teropathogen growth than the very poorly soluble electrolytic
iron used in this study. Only a few bacteria do not require iron,
and Lactobacillus is the major enteric bacterial genera that do
not (53). Lactobacilli do not produce siderophores, and their
growth is similar in media with and without iron (54). Abundant
FIGURE 2. Enumeration (means 6 SEs) of the rrs gene copy number of
total bacteria and cell numbers of Bacteroides, enterobacteria, lactobacilli,
and bifidobacteria by real-time polymerase chain reaction performed with
fecal DNA from baseline fecal samples of Ivorian children (n = 59). Relative
cell numbers were calculated by normalization of the total rrs gene copy
number to the mean copy number per cell for each bacterial group (n = 4.63
for Bacteroides, 7.0 for enterobacteria, and 5.54 for lactobacilli). To compare
groups, one-factor ANOVA was used with post hoc t tests. Values without
a common letter differed significantly, P , 0.02.
FIGURE 1. A: Mean (6SE) numbers of polymerase chain reaction and
temporal temperature gradient electrophoresis (PCR-TGGE) bands in the
fecal samples of Ivorian children who received iron fortification or
a control for 6 mo. B: Mean (6SE) percentage distances of PCR-TGGE
bands in the fecal samples of Ivorian children who received iron fortification
or a control for 6 mo. Comparisons of TGGE profiles were performed by
using Dice similarity coefficient analysis. Values without a common letter
differed significantly, P , 0.0001.
ZIMMERMANN ET AL
by guest on June 6, 2011
lactobacilli and other commensal bacteria in the colon provide
an important barrier effect against colonization and invasion by
pathogens (55–57). Iron fortification reduced the lactobacilli
number and may have weakened this protective effect. The
higher ratio of enterobacteria to lactobacilli (10) may have
encouraged colonization by Salmonella in the iron group.
The increase in fecal calprotectin concentrations (Figure 3)
with iron fortification suggested that changes in the gut micro-
biota may have increased gut inflammation. Calprotectin is
a 36.5-kDa calcium-binding polypeptide observed in the cytosol
calprotectin concentrations reflect translocation and migration of
primarily neutrophils into the intestinal mucosa. It has higher
specificity than systemic inflammatory markers for gut in-
flammation and is more sensitive; this was evident in our data
where CRP and AGP concentrations did not increase in the iron
group, whereas calprotectin concentrations did increase. Other
potential factors that can increase fecal calprotectin concen-
trations are significant fecal blood (.100 mL) and chronic
nonsteroidal antiinflammatory drug use, but these were unlikely
in our population with a low intensity of helminth infection.
Gastroenteritis in children can elevate fecal calprotectin con-
centrations; with the use of the same assay as in our study, the
median (range) fecal calprotectin concentration in preschool Eu-
ropean children with acute gastroenteritis was 110 (0.3–244) lg/g
(59). A daily dose of 120 mg supplemental iron for 14 d did not
increase already elevated fecal calprotectin concentrations in
anemic adults with inflammatory bowel disease (60). The direct
correlation between increased fecal calprotectin concentrations
and increased enterobacteria numbers in our data suggest the
iron-associated increases in enterobacteria may have contributed
to the increase in gut inflammation.
Increasing iron intakes may increase diarrheal disease in
children; asystematic review (12) concluded that the provision of
diarrhea. Since that review, 2 large trials of iron supplementation
in children (12.5 mg Fe/d with 50 lg folic acid) in Nepal (53)
and Tanzania (13) reported cause-specific mortality and diarrhea
incidence as secondary outcomes. In Nepal, there was a non-
significant increased risk of death from diarrhea in the iron
group [relative risk (RR) (95% CI): 1.21 (0.66, 2.11)] but no
significant difference in diarrhea incidence [RR (95% CI): 0.94
(0.84, 1.05)] (61). In Pemba, there was also no significant dif-
ference in diarrhea incidence [RR (95% CI): 0.92 (0.68, 1.25)]
(13). However, recent controlled iron-supplementation trials
(12.5–15 mg Fe/d) in Peru (62) and Bangladesh (63) reported
a significant increase in diarrhea.
In contrast, there are no convincing data that iron fortification
increased risk of diarrhea, but to our knowledge, there have been
fortification studies in the review of Gera and Sachdev (12)
reported diarrheal outcomes; one study (64) showed a signifi-
no effect on self-reported diarrhea, butthe study was not powered
to detect this. Overall, the available data suggest that untargeted
oral iron supplementation is associated with a small increase in
risk of diarrhea in children. Our findings provide a potential
mechanism for this effect.
an iron intervention on the human gut microbiota in Africa, and
our findings need confirmation in other settings and populations.
This is important because diarrhea remains a major cause of
morbidity and mortality in African children, and iron supple-
mentation and fortification are the most common strategies used
in Africato combatanemia. It ispossiblethat higherdosesofiron
giveninsupplements[inchildren,2mgFe? kg21? d21;upto30mg
Fe/d (21)] have an even greater effect on the gut microbiota
than observed in the current study, in which the children re-
ceived ’8 mg Fe/d. If iron disturbs gut homeostasis and en-
courages growth of enteropathogens, particularly in vulnerable
periods such as infancy, it may be important to 1) consider the
use of a probiotic to maintain lactobacilli and bifidobacteria
FIGURE 3. Enumeration (means 6 SEs) of bifidobacteria (A), lactobacilli (B), and enterobacteria (C) by real-time polymerase chain reaction performed
with fecal DNA from Ivorian children in the control and iron-fortification groups as well as fecal calprotectin concentrations (D) at baseline (0) and after 6 mo
(6). Relative cell numbers were calculated by normalization of the total rrs gene copy number to the mean copy number per cell for each bacterial group (n =
4.63 for Bacteroides, 7.0 for enterobacteria, and 5.54 for lactobacilli). To compare groups, 2-factor ANOVA or, for lactobacilli, ANCOVA with zero time
values as covariates were used with post hoc t tests. Values without a common letter differed significantly, P , 0.01.
IRON AND GUT MICROBIOTA
by guest on June 6, 2011
populations during the provision of iron and/or 2) minimize the
amount of added iron by maximizing the bioavailability of both
the fortificant and the native iron in the diet. Strategies to do this
are available, including the use of chelated iron and/or absorp-
tion enhancers (40, 65).
We thank Andreas B Tschannen and Daniel E Sess for assistance in the
study, Mahamadou Bakayoko, Kouadio J Brou, Soste `ne Brou, Kouassi N
Lingue ´, Laurent K Lohourignon, Moussan N’Cho, Diabate ´ Saulio, Kigbafori
D Silue ´, Mahamadou Traore ´, and Evelyne N Yapi for technical assistance in
parasitology, and Samuela Rossi and Doreen Gille for assistance with the gut
The authors’ responsibilities were as follows—MBZ, FR, and HG: con-
ducted field work; CC, AD, and CL: performed gut microbiota analyses;
MBZ, FR, AD, and CC: analyzed data; MBZ and CC: wrote the first draft of
the manuscript; and all authors: helped design the research and edited and
approved the final manuscript. FR is affiliated with the Global Alliance for
Improved Nutrition, a not-for-profit organization that supports food-fortifi-
cation programs. MBZ, CC, EKN, CN, AD, JU, HG, CL, and RFH declared
no conflicts of interest.
1. Zimmermann MB, Hurrell RF. Nutritional iron deficiency. Lancet 2007;
2. Baltussen R, Knai C, Sharan M. Iron fortification and iron supplemen-
tation are cost-effective interventions to reduce iron deficiency in four
subregions of the world. J Nutr 2004;134:2678–84.
3. Hurrell RF, Ranum P, de Pee S, et al. Revised recommendations for iron
fortification of wheat flour and an evaluation of the expected impact of
current national wheat flour fortification programs. Food Nutr Bull 2010;
4. Flour Fortification Initiative. Global flour fortification database. Avail-
able from: http://www.sph.emory.edu/wheatflour/countrydata.php (cited
12 March 2010).
5. Oppenheimer SJ. Iron and its relation to immunity and infectious dis-
ease. J Nutr 2001;131:616s–35s.
6. Andrews SC, Robinson AK, Rodrı ´guez-Quin ˜ones F. Bacterial iron ho-
meostasis. FEMS Microbiol Rev 2003;27:215–37.
7. Bezkorovainy A, Kot E, Miller-Catchpole R, Halofis G, Furmanov S.
Iron metabolism in bifidobacteria. Int Dairy J 1996;6:905–19.
8. Naikare H, Palyada K, Panciera R, Marlow D, Stintzi A. Major role for
FeoB in Campylobacter jejuni ferrous iron acquisition, gut colonization,
and intracellular survival. Infect Immun 2006;74:5433–44.
9. Lee SH, Shinde P, Choi J, et al. Effects of dietary iron levels on growth
performance, hematological status, liver mineral concentration, fecal
microflora, and diarrhea incidence in weanling pigs. Biol Trace Elem
Res 2008;126(suppl 1):S57–68.
10. Reid CA, Hillman K. The effects of retrogradation and amylose/amy-
lopectin ratio of starches on carbohydrate fermentation and microbial
populations in the porcine colon. Br J Nutr 1999;68:503–10.
11. World Health Organization. World health report 2002. Available from:
http://www.who.int/whr/ (cited 12 March 2010).
12. Gera T, Sachdev HP. Effect of iron supplementation on incidence of
infectious illness in children: systematic review. BMJ 2002;325:1142–9.
13. Sazawal S, Black RE, Ramsan M, et al. Effects of routine prophylactic
supplementation with iron and folic acid on admission to hospital and
mortality in preschool children in a high malaria transmission setting:
community-based, randomised, placebo-controlled trial. Lancet 2006;
14. World Health Organization Secretariat. Conclusions and recom-
mendations of the WHO Consultation on prevention and control of iron
deficiency in infants and young children in malaria endemic areas. Food
Nutr Bull 2007;28:S621–7.
15. Rohner F, Zimmermann MB, Amon RJ, et al. In a randomized controlled
trial of iron fortification, anthelmintic treatment, and intermittent pre-
ventive treatment of malaria for anemia control in Ivorian children, only
anthelmintic treatment shows modest benefit. J Nutr 2010;140:635–41.
16. Asobayire FS, Adou P, Davidsson L, Cook JD, Hurrell RF. Prevalence of
iron deficiency with and without concurrent anemia in population groups
with high prevalences of malaria and other infections: a study in Cote
d’Ivoire. Am J Clin Nutr 2001;74:776–82.
17. Rohner F, Zimmermann MB, Wegmueller R, Tschannen AB, Hurrell
RF. Mild riboflavin deficiency is highly prevalent in school-age children
but does not increase risk for anaemia in Co ˆte d’Ivoire. Br J Nutr 2007;
18. Humblot C, Bruneau A, Sutren A, et al. Brussels sprouts, inulin and
fermented milk alter the faecal microbiota of human microbiota-associated
rats as shown by PCR-temporal temperature gradient gel electrophoresis
using universal, Lactobacillus and Bifidobacterium 16S rRNA gene
primers. Br J Nutr 2005;93:677–84.
19. Monira S, Alam NH, Suau A, et al. Time course of bacterial diversity in
stool samples of malnourished children with cholera receiving treat-
ment. J Pediatr Gastroenterol Nutr 2009;48:571–8.
20. World Health Organization and Food and Agriculture Organization.
Vitamin and mineral requirements in human nutrition. 2nd ed. Geneva,
Switzerland: WHO, 2004.
21. World Health Organization. The treatment of diarrhea: a manual for
physicians and other senior health workers. Geneva, Switzerland: WHO,
22. World Health Organization /UNICEF/UNU. Iron deficiency anaemia.
Assessment, prevention and control. A guide for programme managers,
Geneva, Switzerland: World Health Organization, 2001. (Publication
23. Zimmermann MB, Molinari L, Staubli-Asobayire F, et al. Serum
transferrin receptor and zinc protoporphyrin as indicators of iron status
in African children. Am J Clin Nutr 2005;81:615–23.
24. Katz N, Chaves A, Pellegrino J. A simple device for quantitative stool
thick-smear technique in Schistosomiasis mansoni. Rev Inst Med Trop
Sao Paulo 1972;14:397–400.
25. Ogier JC, Son O, Gruss A, Tailliez P, Delacroix-Buchet A. Identification
of the bacterial microflora in dairy products by temporal temperature
gradient gel electrophoresis. Appl Environ Microbiol 2002;68:3691–
26. Heilig HG, Zoetendal EG, Vaughan EE, Marteau P, Akkermans AD, de
Vos WM. Molecular diversity of Lactobacillus spp. and other lactic acid
bacteria in the human intestine as determined by specific amplification
of 16S ribosomal DNA. Appl Environ Microbiol 2002;68:114–23.
27. Guo X, Xia X, Tang R, Zhou J, Zhao H, Wang K. Development of
a real-time PCR method for Firmicutes and Bacteroidetes in faeces and
its application to quantify intestinal population of obese and lean pigs.
Lett Appl Microbiol 2008;47:367–73.
29. Ramirez-Farias C, Slezak K, Fuller Z, Duncan A, Holtrop G, Louis P.
Effect of inulin on the human gut microbiota: stimulation of Bifido-
bacterium adolescentis and Faecalibacterium prausnitzii. Br J Nutr
30. Furet JP, Firmesse O, Gourmelon M, et al. Comparative assessment of
human and farm animal faecal microbiota using real-time quantitative
PCR. FEMS Microbiol Ecol 2009;68:351–62.
31. Frank DN, St Amand AL, Feldman RA, Boedeker EC, Harpaz N, Pace
NR. Molecular-phylogenetic characterization of microbial community
imbalances in human inflammatory bowel diseases. Proc Natl Acad Sci
32. Fukushima H, Tsunomori Y, Seki R. Duplex real-time SYBR green PCR
assays for detection of 17 species of food- or waterborne pathogens in
stools. J Clin Microbiol 2003;41:5134–46.
34. Meile L, Ludwig W, Rueger U, et al. Bifidobacterium lactis sp. nov.,
a moderately oxygen tolerant species isolated from fermented milk. Syst
Appl Microbiol 1997;20:57–64.
35. Klappenbach JA, Saxman PR, Cole JR, Schmidt TM. rrnDB: the Ri-
bosomal RNA Operon Copy Number Database. Nucleic Acids Res
36. Lee ZM, Bussema C, Schmidt TM. rrnDB: documenting the number of
rRNA and tRNA genes in bacteria and archaea. Nucleic Acids Res 2009;
37. Cleusix V, Lacroix C, Dasen G, Leo M, Le Blay G. Comparative study
of a new quantitative real-time PCR targeting the xylulose-5-phospha-
te/fructose-6-phosphate phosphoketolase bifidobacterial gene (xfp) in
faecal samples with two fluorescence in situ hybridization methods. J
Appl Microbiol 2010;108:181–93.
38. Aranda KR, Fagundes-Neto U, Scaletsky IC. Evaluation of multiplex
PCRs for diagnosis of infection with diarrheagenic Escherichia coli and
Shigella spp. J Clin Microbiol 2004;42:5849–53.
39. Nemeth E, Valore EV, Territo M, Schiller G, Lichtenstein A, Ganz T.
Hepcidin, a putative mediator of anemia of inflammation, is a type II
acute-phase protein. Blood 2003;101:2461–3.
ZIMMERMANN ET AL
by guest on June 6, 2011
40. van Stuijvenberg ME, Smuts CM, Wolmarans P, Lombard CJ, Dhansay Download full-text
MA. The efficacy offerrous bisglycinate and electrolytic iron asfortificants
in bread in iron-deficient school children. Br J Nutr 2006;95:532–8.
41. Andang’o PE, Osendarp SJ, Ayah R, et al. Efficacy of iron-fortified
whole maize flour on iron status of schoolchildren in Kenya: a rando-
mised controlled trial. Lancet 2007;369:1799–806.
42. Sokol H, Lepage P, Seksik P, Dore ´ J, Marteau P. Molecular comparison
of dominant microbiota associated with injured versus healthy mucosa
in ulcerative colitis. Gut 2007;56:152–4.
43. Lay C, Sutren M, Rochet V, Saunier K, Dore ´ J, Rigottier-Gois L. Design
and validation of 16S rRNA probes to enumerate members of the
Clostridium leptum subgroup in human faecal microbiota. Environ
44. Singhal A, Macfarlane G, Macfarlane S, et al. Dietary nucleotides and
fecal microbiota in formula-fed infants: a randomized controlled trial.
Am J Clin Nutr 2008;87:1785–92.
45. Nakamura N, Gaskins HR, Collier CT, et al. Molecular ecological
analysis of fecal bacterial populations from term infants fed formula
supplemented with selected blends of prebiotics. Appl Environ Micro-
46. Stecher B, Chaffron S, Ka ¨ppeli R, et al. Like will to like: abundances of
closely related species can predict susceptibility to intestinal colonization
by pathogenic and commensal bacteria. PLoS Pathog 2010;6:e1000711.
47. Flint HJ, Duncan SH, Scott KP, Louis P. Interactions and competition
within the microbial community of the human colon: links between diet
and health. Environ Microbiol 2007;9:1101–11.
48. Stecher B, Hardt WD. The role of microbiota in infectious disease.
Trends Microbiol 2008;16:107–14.
49. Zoetendal EG, Akkermans ADL, de Vos WM. Temperature gradient gel
electrophoresis analysis of 16S rRNA from human fecal samples reveals
stable and host-specific communities of active bacteria. Appl Environ
50. Bullen J, Griffiths E, Rogers H, Ward G. Sepsis: the critical role of iron.
Microbes Infect 2000;2:409–15.
51. Boyer E, Bergevin I, Malo D, Gros P, Cellier MF. Acquisition of Mn(II)
in addition to Fe(II) is required for full virulence of Salmonella enterica
serovar Typhimurium. Infect Immun 2002;70:6032–42.
52. Tsolis RM, Baumler AJ, Heffron F, Stojiljkovic I. Contribution of TonB-
and Feo-mediated iron uptake to growth of Salmonella typhimurium in
the mouse. Infect Immun 1996;64:4549–56.
53. Archibald F. Lactobacillus platarum, an organism not requiring iron.
FEMS Microbiol Lett 1983;19:29–32.
54. Pandey A, Bringel F, Meyer JM. Iron requirement and search for side-
rophores in lactic acid bacteria. Appl Microbiol Biotechnol 1993;40:
55. Coconnier MH, Lievin V, Bernet-Camard MF, Hudault S, Servin AL.
Antibacterial effect of the adhering human Lactobacillus acidophilus
strain LB. Antimicrob Agents Chemother 1997;41:1046–52.
56. Lievin V, Peiffer I, Hudault S, et al. Bifidobacterium strains from resi-
dent infant human gastrointestinal microflora exert antimicrobial activ-
ity. Gut 2000;47:646–52.
57. Weinberg ED. The Lactobacilli anomaly: total iron abstinence. Perspect
Biol Med 1997;40:1–6.
58. Gisbert JP, McNicholl AG. Questions and answers on the role of faecal
calprotectin as a biological marker in inflammatory bowel disease. Dig
Liver Dis 2009;41:56–66.
59. Berni Canani R, Rapacciuolo L, Romano MT, et al. Diagnostic value of
faecal calprotectin in paediatric gastroenterology clinical practice. Dig
Liver Dis 2004;36:467–70.
60. Erichsen K, Ulvik RJ, Nysaeter G, et al. Oral ferrous fumarate or in-
travenous iron sucrose for patients with inflammatory bowel disease.
Scand J Gastroenterol 2005;40:1058–65.
61. Tielsch JM, Khatry SK, Stoltzfus RJ, et al. Effect of routine prophylactic
supplementation with iron and folic acid on preschool child mortality in
southern Nepal: community-based, cluster-randomised, placebo-con-
trolled trial. Lancet 2006;367:144–52.
62. Richard SA, Zavaleta N, Caulfield LE, Black RE, Witzig RS, Shankar
AH. Zinc and iron supplementation and malaria, diarrhea, and re-
spiratory infections in children in the Peruvian Amazon. Am J Trop Med
63. Chang S, El Arifeen S, Bari S, et al. Supplementing iron and zinc:
double blind, randomized evaluation of separate or combined delivery.
Eur J Clin Nutr 2010;64:153–60.
64. Brunser O, Espinoza J, Araya M, Pacheco I, Cruchet S. Chronic iron
intake and diarrhoeal disease in infants: a field study in a less-developed
country. Eur J Clin Nutr 1993;47:317–26.
65. Troesch B, Egli I, Zeder C, Hurrell RF, de Pee S, Zimmermann MB.
Optimization of a phytase-containing micronutrient powder with low
amounts of highly bioavailable iron for in-home fortification of com-
plementary foods. Am J Clin Nutr 2009;89:539–44.
IRON AND GUT MICROBIOTA
by guest on June 6, 2011