(Advances in agronomy 115) donald l sparks (eds ) advances in agronomy 115 academic press, elsevier (2012)

ADVANCES IN AGRONOMY
Advisory Board

PAUL M. BERTSCH

RONALD L. PHILLIPS

University of Kentucky

University of Minnesota

KATE M. SCOW

LARRY P. WILDING

University of California,
Davis

Texas A&M University

Emeritus Advisory Board Members

JOHN S. BOYER

KENNETH J. FREY

University of Delaware

Iowa State University

EUGENE J. KAMPRATH

MARTIN ALEXANDER

North Carolina State
University

Cornell University

Prepared in cooperation with the
American Society of Agronomy, Crop Science Society of America, and Soil
Science Society of America Book and Multimedia Publishing Committee
DAVID D. BALTENSPERGER, CHAIR
LISA K. AL-AMOODI

CRAIG A. ROBERTS

WARREN A. DICK

MARY C. SAVIN

HARI B. KRISHNAN

APRIL L. ULERY

SALLY D. LOGSDON


Academic Press is an imprint of Elsevier
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
225 Wyman Street, Waltham, MA 02451, USA

32 Jamestown Road, London, NW1 7BY, UK
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
First edition 2012
Copyright # 2012 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted
in any form or by any means electronic, mechanical, photocopying, recording or
otherwise without the prior written permission of the publisher
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;
email: Alternatively you can submit your request online by
visiting the Elsevier web site at and selecting
Obtaining permission to use Elsevier material
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or
property as a matter of products liability, negligence or otherwise, or from any use or
operation of any methods, products, instructions or ideas contained in the material
herein. Because of rapid advances in the medical sciences, in particular, independent
verification of diagnoses and drug dosages should be made
ISBN: 978-0-12-394276-0
ISSN: 0065-2113 (series)
For information on all Academic Press publications
visit our website at elsevierdirect.com
Printed and bound in USA
12 13 14 15
10 9 8 7 6 5 4 3 2 1


CONTRIBUTORS

Numbers in Parentheses indicate the pages on which the authors’ contributions begin.

Nanthi S. Bolan (215)
Centre for Environmental Risk Assessment and Remediation, University of South
Australia, Mawson Lakes, Australia; Cooperative Research Centre for Contamination Assessment and Remediation of the Environment, Adelaide, Australia
P. Dhakal (181)
University of Kentucky, Lexington, Kentucky, USA
Alison J. Eagle1 (79)
Institute for Land Use Innovation, University of Alberta, Edmonton, Alberta,
Canada
R. Evans (41)
Anglia Ruskin University, Cambridge, Cambridgeshire, United Kingdom
P. D. Falloon (41)
Met Office Hadley Centre, Exeter, Devon, United Kingdom
Robin D. Graham (1)
School of Biology, Flinders University of South Australia, Adelaide, Australia
P. M. Haygarth (41)
Centre for Sustainable Water Management, Lancaster Environment Centre,
Lancaster University, Lancaster, Lancashire, United Kingdom
A. M. Ismail (299)
International Rice Research Institute (IRRI), Metro Manila, Philippines
Won-Il Kim (215)
Chemical Safety Division, Department of Agro-Food Safety, National Academy of
Agricultural Science, Suwon-si, Gyeonggi-do, Republic of Korea
Marija Knez (1)
School of Biology, Flinders University of South Australia, Adelaide, Australia

1

Formerly with the Nicholas Institute for Environmental Policy Solutions, Duke University, Durham, North
Carolina, USA.

ix


x

Contributors

Anitha Kunhikrishnan (215)
Centre for Environmental Risk Assessment and Remediation, University of South
Australia, Mawson Lakes, Australia; Cooperative Research Centre for Contamination Assessment and Remediation of the Environment, Adelaide, Australia; Chemical Safety Division, Department of Agro-Food Safety, National Academy of
Agricultural Science, Suwon-si, Gyeonggi-do, Republic of Korea
R. V. Labios (299)
College of Agriculture, University of the Philippines, Los Ban˜os, Philippines
Seth Laurenson (215)
Land and Environment, AgResearch Ltd, Invermay, New Zealand
¨ller (215)
Karin Mu
Systems Modelling, The NZ Institute for Plant and Food Research Ltd., Hamilton,
New Zealand
D. J. Mackill (299)
International Rice Research Institute (IRRI), Metro Manila, Philippines; Mars,
Inc.; Department of Plant Sciences, University of California, Davis, USA
C. J. A. Macleod (41)
The James Hutton Institute, Craigiebuckler, Aberdeen, United Kingdom
C. J. Matocha (181)
University of Kentucky, Lexington, Kentucky, USA
Ravi Naidu (215)
Centre for Environmental Risk Assessment and Remediation, University of South
Australia, Mawson Lakes, Australia; Cooperative Research Centre for Contamination Assessment and Remediation of the Environment, Adelaide, Australia

Lydia P. Olander (79)
Nicholas Institute for Environmental Policy Solutions, Duke University, Durham,
North Carolina, USA
T. R. Paris (301)
International Rice Research Institute (IRRI), Metro Manila, Philippines
S. M. Pyzola (181)
University of Kentucky, Lexington, Kentucky, USA
U. S. Singh (299)
International Rice Research Institute (IRRI), New Delhi, India
Ross M. Welch (1)
Department of Crop and Soil Sciences, Cornell University, Ithaca, New York,
USA


PREFACE

Volume 115 contains six excellent reviews covering important global topics
including human health, climate change, nutrient and trace metal mobility
and bioavailability, and food production. Chapter 1 is a comprehensive
review on the role of zinc deficiency on nutritional iron deficiency in
humans. Chapter 2 deals with an assessment of climate impacts on hydrological mobilization of diffuse substances from agriculture. Chapter 3 provides an overview of greenhouse gas mitigation with agricultural land
management in the United States. Chapter 4 covers the role of abiotic
and coupled biotic/abiotic mineral controlled redox processes in nitrate
reduction. Chapter 5 provides a critical review of the role of wastewater
irrigation on the transformation and bioavailability of heavy metal(loids) in
soil. Chapter 6 addresses the development and adoption of submergencetolerant rice varieties.
I appreciate the fine reviews of the authors.
DONALD L. SPARKS
Newark, Delaware, USA

xi


C H A P T E R

O N E

How Much Nutritional Iron Deficiency
in Humans Globally Is due to an
Underlying Zinc Deficiency?
Robin D. Graham,* Marija Knez,* and Ross M. Welch†
Contents
1. Introduction
2. Agronomy of Micronutrients in Respect to the Green Revolution
1960–1980
2.1. Seed nutrient content
2.2. Iron deficiency in humans
2.3. Zinc deficiency and its impact on iron nutrition
2.4. Vitamin A deficiency and its significance
2.5. Food systems strategies
3. Iron and Zinc Interactions in Human Nutrition
3.1. Synergy or antagonism
3.2. Supplementation studies
3.3. Fortification studies show no antagonism
3.4. Zinc and anemia
3.5. The regulation of hemoglobin levels
3.6. Micronutrient deficiencies are occurring together
3.7. Iron and zinc transporters in enterocytes of the small intestine
3.8. Positive role of zinc in oxidative damage and protein synthesis
4. Whole Body Regulation of Iron and Zinc in Humans

4.1. Iron homeostasis
4.2. Hepcidin, an iron store regulator
4.3. Hepcidin regulates DMT1 and/or FPN expression and function
4.4. Zinc, an important regulator of iron absorption
4.5. The role of zinc in decreasing systemic intestinal inflammation
and iron deficiency
4.6. Anticipated mechanism of zinc action on iron deficiency

2
5
8
9
10
12
12
14
14
15
16
16
17
18
18
21
22
22
24
25
26
27

28

* School of Biology, Flinders University of South Australia, Adelaide, Australia
Department of Crop and Soil Sciences, Cornell University, Ithaca, New York, USA

{

Advances in Agronomy, Volume 115
ISSN 0065-2113, DOI: 10.1016/B978-0-12-394276-0.00001-9

#

2012 Elsevier Inc.
All rights reserved.

1


2

Robin D. Graham et al.

5. Healthy Food Systems
6. Conclusion
Acknowledgment
References

30
32
33

33

Abstract
This chapter recounts the impact of the green revolution (1960–1980) on
subsequent world food supplies and its consequences in terms of human
nutrition and health via its impact on the micronutrient status of staple foods
and of diets generally. Micronutrient deficiency disorders now occur in over half
of the total human population. This chapter then reviews the recent medical
literature on the molecular physiology of the human gut in relation to micronutrient absorption from food and the regulation of nutrient balance from diets
heavily based on cereals that are relatively poor in micronutrients. Weaving
these two literatures together leads to the conclusion that basing the green
revolution on low micronutrient-dense cereals to replace the lower yielding but
more nutrient-dense pulses and other dicotyledonous food crops is the probable cause of the epidemics of micronutrient deficiencies in the burgeoning
human population in the years since 1980. There are lessons in this for the
implementation of new efforts to increase food production in the face of even
further increases in population forecast to 2050, especially the new effort
starting in Africa, and for improving primary health care generally in resourcerich as well as resource-poor countries. We conclude that while complete
nutrient balance in our diets is the only satisfactory aim of a sustainable food
strategy, we focus attention on zinc deficiency and its alleviation as the most
extensive and urgent problem among several that arose as an unforeseen side
effect of the first green revolution.

1. Introduction
The first green revolution (begun in 1960) more than doubled cereal
production worldwide (Fig. 1), an achievement that, in the face of a rapidly
rising human population, turned aside the threat of mass starvation in
1960 and of continuing food shortages during the 1960s and 1970s to
reach a global surplus again by 1980. The emphasis by the international
consortium of agricultural scientists was naturally on increasing yield, both
by plant breeding and use of NPK fertilizers, and as it was known that across

varieties an inverse relation existed between yield of grain and protein
concentration in grain, the latter and other issues of nutritional quality
were largely set aside. No attention whatever was paid to micronutrient
density of the green revolution cereal varieties, a quality issue that was a low
priority among nutritionists at that time.


3

Cereal production
Pulse production
Population

200

200

Developing
nations

World

0

Developing
nations

0

Bangladesh

50

Pakistan

50

India

100

Developing
nations

100

Bangladesh

150

Pakistan

150

% Increase in population (from 1965 to 1999)

250

250

India

% Increase in production (from 1965 to 1999)

Iron and Zinc Deficiencies in Crops and Humans

Figure 1 Percent changes in cereal and pulse (grain legume) production and in population, 1965–1999 (Welch, 2002a,b).

Figure 1 shows the percentage increases of cereal and of pulse (grain
legume) production in developing countries between 1965 and 1999. Developing country population doubled during this period (represented by the
“100%” line). It is the great achievement of the green revolution that cereal
production much more than doubled due to rapid technological change.
However, pulse production per capita declined markedly; owing to the
urgency to produce more, the new technology was not applied to these
low-yielding secondary staples or to vegetables. These changes in production
altered the relative prices of these commodities—lower prices of cereals and
higher noncereal food prices—so it became even more difficult for the poor
to achieve mineral and vitamin adequacy in their diets. In the absence of
adequate knowledge among resource-poor populations of the importance for
health of micronutrient and vitamin intakes, diets have shifted toward increasing reliance on cereal staples (Graham et al., 2007), leading to micronutrient
malnutrition, poorer health, and much misery.
During the 1980s, a steady rise was noted in the extent of iron-deficiency
anemia in humans, especially among the resource-poor populations that
benefited most from the greater productivity of the green revolution
(Graham, 2008; Graham et al., 2007); however, a putative cause-and-effect
association between the rising extent of nutritional iron deficiency and the
low micronutrient density of the expanding green revolution cereal varieties,
vis-a`-vis the lower-yielding crops they displaced, was not canvassed until
much later. The anemia was treated by the medical community using diet

4

Robin D. Graham et al.

supplementation and food fortification strategies, with a major program called
for by the end of the 1980s decade. These programs were facilitated by the ease
of diagnosis of iron deficiency in a small sample of peripheral blood. During this
decade, three other micronutrient deficiencies affecting large numbers of
people, those of iodine, vitamin A, and selenium, were promoted and treatments developed (Ren et al., 2008). Deficiencies of iodine and selenium were
regional, associated with extreme low levels of the nutrients in the soil, and as
neither of them was known to affect crop production, these were treated
medically, as with anemia, by food fortification and supplementation in the
deficient regions. Vitamin A, however, was more generally associated with
population density, insufficiency of the food supply, and again like anemia,
associated with the production of the green revolution varieties of cereals; again
no attribution of cause and effect was made and health authorities deployed
supplementation and food fortification strategies. The new green revolution
varieties of wheat and rice were uniformly white-floured, containing very low
concentrations of yellow provitamin A carotenoids; however, yellow endosperm varieties were known and held in the germplasm banks of both crops.
A clinical deficiency of zinc in a human was reported in a remarkably
prescient paper in the 1960s (Prasad et al., 1963) and Prasad later published
results of a clinical trial in the 1980s (Prasad, 1991), but both efforts were
largely ignored. Only in the 1990s was a body of evidence accumulated that
attracted some recognition (Prasad, 2003), but as there was, unlike anemia, no
quick and simple diagnostic for zinc deficiency in humans, the problem
continued to be largely ignored. Not until Hotz and Brown (2004) edited
an important paper on the extent of zinc-deficient diets of the world, affecting
nearly half the global population, was zinc deficiency taken as a potentially
serious public health problem. Still little has been done about it even to the

present day, although two developments must be acknowledged: first, the
appearance of zinc deficiency as a priority in public health on the WHO
website in 2001, and second, zinc deficiency diagnosis in blood serum by ICP
atomic emission spectrometry is now deemed a valid diagnostic at a population
level but not for the individual; moreover, this analysis is still far from as easy
and inexpensive as is the simple test for anemia (de Benoist et al., 2007).
At the same time, soil scientists and agronomists were well aware that
zinc-deficient soils are widespread on Earth, about half of the major agriculturally productive soil types (Sillanpaa, 1982, 1990). In contrast, crops were
iron deficient on only 3% of soils (Table 1). Moreover, zinc is low in cereal
grains, now the basis of diets for the majority of people everywhere. More
zinc can be incorporated into cereal grains both by zinc fertilization of the
crop and by breeding new cereal varieties inherently richer in zinc (Graham
et al., 1992; Yilmaz et al., 1998), so the tools to solve zinc deficiency globally
have been available, but motivation is still lacking for an integrated “Food
Systems” approach that will provide a sustainable solution on a global scale.
This chapter reviews the medical literature on zinc deficiency, iron deficiency, and their interactions in the human gut, and presents a physiologically


5

Iron and Zinc Deficiencies in Crops and Humans

Table 1 Percentage of nutrient-deficient soils among 190 major soils worldwide
(Sillanpaa, 1982) and in parts of Bangladesh for comparison (Morris et al., 1997)
Macronutrients
Deficiency

World
Acute
Latent

Total
Bangladesh
Total

Micronutrients

N

P

K

B

Cu

Fe

Mn

Mo

Zn

71
14
85

55
18

73

36
19
55

10
21
31

4
10
14

0
3
3

1
9
10

3
12
15

25
24
49

100

22

2

69

3

1

24

15

85

A latent deficiency is one masked by an even more severe deficiency of another nutrient, often N or P,
such that the latent deficiency becomes limiting after the other, more acute deficiency is corrected.

based case that, potentially, a significant proportion of the iron-deficiency
anemia in humans is due to zinc deficiency. This is intended to strengthen
the case for a greater effort to eliminate zinc deficiency worldwide (and with
it some of the anemia) through an integrated Food Systems-based new
green revolution (Graham, 2008).
Because of the complex of homeostatic mechanisms in the body for
preventing excess iron accumulation that in turn prevents peroxidative cellular damage (Edison et al., 2008), this chapter also questions the wisdom of
some of the supplementation, biofortification, and process fortification
of iron, that is current practice, based on blood tests for hemoglobin

and ferritin alone, without showing improvements in health and physical
and mental work capacity. We therefore raise the question whether relatively
more of the global effort to relieve iron deficiency should be spent on
eliminating zinc deficiency and other overt, interacting micronutrient deficiencies, sustainably through an agriculturally based Food Systems strategy.
In this review, we deal first with the agronomy of the green revolution
effort and then we present a summary of a recent, extensive medical
literature on the molecular physiology of the human intestine and on its
implications for human nutrition. Finally, we bring these two facets
together to develop recommendations for radical change in the current
strategy to eliminate anemia and to propose a new Food Systems strategy.

2. Agronomy of Micronutrients in Respect
to the Green Revolution 1960–1980
In the time man has practised agriculture, crops produced on our soils
have become widely deficient in nitrogen and phosphorus and to a lesser
extent, in potassium and sulfur, nutrients that, until the turn of the twentieth


6

Robin D. Graham et al.

century, agriculture used to solve crop production problems on otherwise
fertile old-world soils. By then, European farmers were using new mineral
fertilizers such as Chilean saltpeter, superphosphate, and muriate of potash,
as well as sulfur, lime, and dolomite. These minerals brought production up
to general expectations, but to experienced eyes, anomalous results hinted
at limitations to production yet to be discovered. In the first half or so of
the twentieth century, a suite of new essential elements was proved essential
for all living things in smaller amount, known to agronomists as the trace

elements and later as “the micronutrients” (this term to human nutritionists
also includes the vitamins, nutrients not needed by plants). The use of
micronutrients contributed greatly to modern mechanized agriculture.
The essential micronutrients for growth of higher plants are iron, zinc,
manganese, boron, copper, cobalt, molybdenum, nickel, and chlorine, but
for animals and man, these and the additional elements, selenium, iodine,
chromium, tin, fluorine, lithium, silicon, arsenic, and vanadium, are required;
some of these additional elements may eventually be found necessary for plants
as well (Nielsen, 1997).
Once the macronutrient deficiencies of soils are treated, Sillanpaa (1990)
estimated that, of 190 major agricultural soils of the world, 49% are deficient
in zinc, 31% deficient in boron, 15% deficient in molybdenum, 14% deficient
in copper, 10% deficient in manganese, and 3% deficient in iron (Table 1).
These figures may be compared with corresponding figures for the human
population that depends on these same soils for most of its food production.
In the same broad terms, it appears that as much as a third of the human
population is deficient in iron (30% of people anemic, mostly iron-deficiency
anemia—WHO website, 2011), a third is deficient in zinc, and roughly
a seventh is deficient in each of iodine, selenium, and the plant-synthesized
organic micronutrient, b-carotene (a provitamin A dimer of vitamin A).
Obviously, multiple micronutrient deficiencies are common. Selenium and
iodine are not known to be required by plants (Lyons et al., 2009), and the
extent of boron deficiency in soils does not lead to the same high priority in
human nutrition as it does for crop growth. Iron deficiency in humans is
exceedingly complex yet it appears the iron in most foods is far more than
the requirement but its bioavailability from staple-plant foods is considered
poor (Hunt, 2003). Apparently, only zinc is directly linked in the food chain
such that deficiency is extensive in both humans and their food crops. The
comparison of crop and human micronutrient deficiencies and the nature of
zinc deficiency in humans raises the question whether zinc deficiency should

be the highest priority among micronutrients for agriculture to address
because to increase the zinc available to crops and to the food chain is
achievable with current technology, and there are flow-on benefits to iron
and vitamin A status in humans. An agricultural solution to zinc deficiency in
humans is all the more compelling because mild to moderate zinc deficiency
in humans is still difficult to diagnose (Fischer-Walker et al., 2007), so the use


Iron and Zinc Deficiencies in Crops and Humans

7

of zinc with all macronutrient fertilizers wherever justified by production
gains is an obvious primary agricultural strategy.
Our emphasis on zinc is based on our analysis of the agronomy of the
green revolution 1960–1980. Its features were a focus on the cereals
(mainly wheat, rice, and maize) utilizing new, high-yielding varieties,
coupled with the use of NPK fertilizers in large amounts to match the
yields of the new varieties. For rice and wheat, the most extensive of the
cereals, the new varieties had no provitamin A or related carotenoids
(whereas maize has both white and yellow types). In general, besides
their large yield advantage, these cereals had, as cereals generally do,
more tolerance to extremes of stress such as heat, cold, drought, flooding,
and pests and diseases, than do the crops they replaced, especially the
pulses (grain legumes). The impact of the green revolution in this respect
is well shown in the data of the UN Food and Agriculture Organization
(FAO) in Fig. 1 where the availability of pulses per head was decreased by
population growth as land was given over to the high yielding and more
reliable cereals. Features of the green revolution that induced or aggravated a low density of zinc in the grains of the cereals used, and subsequently in human populations dependent on them, are:









Low soil–zinc status: 49% of global soils zinc-deficient (Sillanpaa, 1990)
Use of P fertilizers that tend to decrease zinc uptake by plants (Webb and
Loneragan, 1990)
Use of N fertilizers that tend to reduce zinc retranslocation from leaves to
seeds in low-zinc soil (Chaudhry and Loneragan, 1970)
Owing to soil degradation and population growth, agricultural expansion
to higher-pH, lower-rainfall soils characteristic of cereal production
where zinc deficiency is common
Loss of diet diversity toward more refined cereal-based diets lower in
nutrients especially zinc, provitamin A carotenoids, iron, and calcium
Low levels in rice and wheat of provitamin A carotenoids that are synergistic with iron in enhancing zinc absorption from cereal diets (see later).

Some ecologists have argued that the sustainable population of Earth is
about 2 billion humans (Pimentel et al., 1999; Rees, 1996), but the effect of
the mass production of antibiotics during World War II and improved
sanitation is said to have decreased death rates so much that the population
exploded postwar to a current population in excess of 6.7 billion, with
a projected 9 billion before the numbers stabilize and hopefully begin
to decline by 2050 (United Nations Population Division, 2004). It is not
too far-fetched to claim that up to 4.7 billion (i.e., 6.7–2 billion) people are
living today on synthetic urea applied to cereal crops, urea produced
in fertilizer factories from oil/gas, and electricity in the highly energydemanding Haber process (Coates, 1939). This huge production of synthetic
fertilizer, unique to the late decades of the twentieth century, has placed

8

Robin D. Graham et al.

equally huge demands on the world’s agricultural soils to supply matching
amounts of the other essential nutrients.
The contribution to production of food for such a large population
made by the use of micronutrients added to NPK fertilizers is undoubtedly
significant but as yet far from the optimal that must be reached to achieve
sustainability because increases in productivity on land already in cultivation
are needed to relieve the global warming effect of clearing of more forested
land. Because micronutrients are needed in such small amounts, the economics of their use is generally highly favorable, as in one case of the authors
where increases in wheat production were valued at $287/ha for each
93 cents worth of copper fertilizer invested in the crop (Graham et al.,
1987). While the economics of micronutrient use is compelling in most
cases, the challenge is to get both the diagnosis and the delivery right
because adding the wrong micronutrient can seriously decrease yields. Principles for use of micronutrient fertilizers were developed in the third
trimester of the past century (Graham, 2008), although further development
is certainly warranted.

2.1. Seed nutrient content
An important strategy is to increase the micronutrient content of the seeds (or
other edible product), a significant factor in production as well as in nutritional quality for human consumption (Welch, 1986). Indeed, high nutrient
content is one reason for the advantage of certified seed, usually grown on
the best soils, over farmers’ seed. Plant breeders can select for higher micronutrient content of seeds but greater enhancement of most micronutrients can
be achieved by fertilizers, either soil-applied or sprayed on the reproductive
organs including flowers, seedpods, or ears, one to three times during seed
development. Nutrient concentrations can be increased greatly, from less

than double for zinc in rice to 100 times in the case of selenium in wheat
(Lyons et al., 2004). However, while spectacular increases are possible, we
caution against aiming for increases greater than what brings the deficient
nutrient up to a relative abundance that roughly matches that of the other
nutrients in the system, because replacing one imbalance (zinc too low) with
another (zinc too high) will induce a deficiency of another micronutrient and
so represents no progress toward healthy food.
An increment in seed content of critical micronutrients can materially
increase the vigor, stress tolerance, disease resistance, and grain yield of the
subsequent crop produced from those seeds on soils deficient in the target
nutrient. In Bangladesh in comparison to farmers’ seed, yields in responsive
soils over 4 years averaged 24% higher in wheat-growing soils by using seeds
previously enhanced in micronutrients by foliar sprays on the mother
plants (Johnson et al., 2005). Studies of the genetics of seed-nutrient loading
traits indicate a number of genes involved, so the genetic approach, though


Iron and Zinc Deficiencies in Crops and Humans

9

it has potential, is not easy (Lonergan et al., 2009). However, iron salts are
relatively poor fertilizers even when foliar applied so the breeding strategy is
a more viable option to enhance iron levels, if needed. In contrast, zinc
in seeds is easily enhanced, as, for example, the results of Genc et al. (2000)
where on severely deficient soil, 1.5kg/ha of zinc as zinc sulfate increased
seed zinc concentration threefold.

2.2. Iron deficiency in humans
The human population is astonishingly iron deficient despite the planet, its

rocks and soils, being especially rich in iron. The World Health Organization
(1995, 2005, 2011) on its website estimated in 2005 the global incidence of
iron deficiency to be between 4 and 5 billion people, and the current website
identifies 2 billion severely deficient, that is, anemic. Apparently, more than
half the total problem is dietary in origin. Iron deficiency is most severe and
widespread among growing children and premenopausal women, as adult
males until old age are reasonably resistant to anemia despite poor diets in
resource-poor countries (Markle et al., 2007). Most iron-deficient women
and children are debilitated to some degree in both physical and mental work
capacity. In severe cases, this results in morbidity, complications in childbirth,
and mortality for both mothers and children (www.who.int/nutrition/
topics/ida/). Iron deficiency, even when mild, can increase the food required
by 5–10% for the same amount of physical work done (Zhu and Haas, 1997);
a similar increment in yield of 5–10% by modern plant breeding may take up
to 10 years to achieve.
Iron deficiency is an epidemic that exists in spite of few problems in crop
plants. For example, iron deficiency in humans is severe in the acidic
lateritic soil areas of the Asian wet tropics where iron deficiency in crops
is rare, and if anything, it is iron toxicity that is better known, especially in
rice (Phattiyakul et al., 2009).
For humans in resource-poor populations heavily dependent on cereals
for their sustenance, at least 10 times their needs of iron are ingested daily
from those cereal products (other than white rice), but the bioavailability of
that iron is reportedly low (Fairweather-Tait and Hurrell, 1996). The reason
for the low bioavailability of cereal–iron, largely in the form of soluble
monoferric phytate, is thought to be the precipitation by dietary calcium of
complex phytates and other insoluble forms in the small intestine, making it
unavailable. Absorbed and utilized iron, measured by isotopic methods, can
be as little as 1% of ingested iron (Donangelo et al., 2003). In the HarvestPlus
Challenge Program (www.harvestplus.org), that aims to increase the nutritive

value of common staple foods to eliminate iron-deficiency anemia in the
world by biofortification, increasing iron in cereals by selecting iron-dense
genotypes is the main strategy. The effectiveness of this strategy is yet to
be fully established. Due to simpler genetics, it may prove more effective to


10

Robin D. Graham et al.

breed for increased bioavailability-promoting substances (e.g., prebiotics) to
enhance the absorbability of such nonheme iron in staples than to increase the
iron itself in staple food grains (Graham et al., 2007).

2.3. Zinc deficiency and its impact on iron nutrition
Older human nutrition texts identify iron-deficiency anemia as one symptom of zinc deficiency (Prasad et al., 1963). While subsequent studies in
humans that gave supporting results have been deemed of poor design
(Prasad, 1991), this does not disprove the proposition, and studies with
animal models including monkeys have, under more controlled conditions,
supported the hypothesis of zinc deficiency as one cause of iron-deficiency
anemia (Golub, 1984). Recent studies indicate that improved dietary zinc
facilitates the absorption of nonheme iron (see later sections). If this were a
widespread phenomenon, it could explain some of the current extent of
anemia and nutritional iron deficiency, and the failure of the gut to absorb
enough of the iron ingested to meet metabolic needs. Additionally, vitamin
A deficiency, also widespread in humans, can aggravate both iron and zinc
deficiencies, and conversely, correcting any one of these deficiencies can
make more of the other two nutrients available from an otherwise similar
diet (Thurlow et al., 2005; Fig. 2). Carotenoid pigments have been deliberately bred out of wheat and other staples during the twentieth century in
response to consumer demand for white flour (whiteness may be perceived

as evidence of its purity/cleanliness), and iron and zinc concentrations in

Vitamin
A

Absorption
RBT Transport
Utilization

Absorption
Utilization

Zinc

Iron

Absorption

Figure 2 Synergy of iron, zinc, and vitamin A in the human gut: an increase of any one
may enhance absorption and/or utilization of the others when all are low in the diet
(Graham et al., 2000).


11

Iron and Zinc Deficiencies in Crops and Humans

green revolution cereals appear to have decreased even further over time as
yields have been increased by breeding (Graham et al., 2007). Intestinal
infection by Helicobacter pylori and other gut pathogens is also linked to zinc

and iron deficiencies in developing countries (DuBois and Kearney, 2005).
Deficiencies of iodine and selenium induce poor utilization of absorbed iron
that aggravates iron deficiency in humans (Welch, 1986). Finally, vitamin
B12 deficiency can cause anemia (iron-resistant or pernicious anemia), and
although there are no extensive maps of cobalt-deficient soils (vitamin B12
contains cobalt), the extent of vitamin B12 deficiency is increasing as more
extensive testing is conducted (Stabler and Allen, 2004). The collective extent
of deficiencies of zinc, iodine, selenium, vitamin A, and vitamin B12 is more
than sufficient to explain some of the nutritional anemia quantified by World
Health Organisation. More importantly, newly published mechanisms of the
regulation of iron uptake by dietary zinc in humans (Sections 3 and 4) detail
the mechanisms by which zinc deficiency could indeed be the cause of up to
half of the global burden of iron-deficiency anemia.
The agricultural perspective on zinc is much clearer than is the human
nutritional perspective. Zinc fertilizers are remarkably effective, yet half
of the world’s soils are intrinsically deficient, as well as the lithosphere
generally where zinc abundance is barely one thousandth that of iron
(Chesworth, 1991). Zinc deficiency occurs in all the world’s major cropping areas, climates, and soil types. Copper, iron, molybdenum, chlorine,
and manganese have more than one oxidation state and so are easily
manipulated by redox transitions in biological systems in the soil to release
soluble ions of these elements even in the presence of an unfavorable pH.
On the other hand, zinc, nickel, cobalt, and boron rely on coordination
chemistry for changes in solubility, movement through soil and the biosphere, and so these elements tend to function biologically in stable systems
such as structural molecules like DNA, structural proteins, and enzymes,
both metabolic and regulatory. Zinc has been identified to bind with 925
proteins in humans and over 500 proteins in plants (Table 2), 10 times
more than does iron in the same organisms (the opposite of their relative
abundances in the lithosphere/soil). It is not surprising, therefore, that the
occurrence of zinc deficiency is widespread, and in both plants and humans
causes a wide range of symptoms, depending on allelic variation in genes

Table 2 Metal-containing and metal-binding proteins in two species identified by
proteomic techniques
Genome

Total proteins Zn

Homo sapiens
25,319
Arabidopsis thaliana (plant) 27,243
From Gladyshev et al. (2004).

Cu Mg Fe Ca Ni Co Mo

925 31 74 86 59 0
536 19 51 81 14 1

4
4

6
6


12

Robin D. Graham et al.

controlling each of the known zinc-containing/binding proteins. As such,
zinc participates in almost all processes and pathways in living organisms.
It can be deemed the most important metabolic promoter among the

known essential nutrients. Because zinc interacts with such a large number
of proteins, symptoms of zinc deficiency in humans may be many, varied,
and somewhat indiscriminate, and consequently many disease states are not
associated with its deficiency when they should be, and in these respects, it is
not surprising that zinc deficiency is quite difficult to diagnose in humans
and animals. Zinc deficiency is the ultimate “hidden hunger.”
More importantly, in humans, zinc is described as a “type II” element, that
is, its concentration does not markedly decline in the blood stream as severity
of deficiency increases, in contrast to iron, a “type I” nutrient, the concentration of which does decline in the blood markedly as deficiency increases
in severity. When zinc supply is low, the body sacrifices bone zinc stores and
skeletal muscle mass, releasing zinc to the circulation in order to maintain vital
internal organs, whose zinc concentrations also do not fall greatly (Golden,
1995). Thus, unless an individual child has been monitored for height/weight
over many months, there is neither good nor easy diagnosis of zinc deficiency
in an individual (Hess et al., 2007). Until the release of the map of zincdeficient human diets, zinc deficiency was low on the WHO list of important
nutritional problems and this may be a reason zinc deficiency has not been
identified as a potential cause for some of the nutritional anemia reported.

2.4. Vitamin A deficiency and its significance
Vitamin A is widely deficient in humans (Abed and Combs, 2001). Vitamin A
is not a nutrient for plants as they can biosynthesize the carotenes that
the human body converts into vitamin A. Important here is that its deficiency
can cause anemia, and solving the problem of vitamin A deficiency is
important to eliminating anemia (Bloem et al., 1989; Suharno and Muhilal,
1996). As carotenes are not nutrients for plants, there is no fertilizer strategy,
and new foods must be added to vitamin A-deficient food systems or existing
staples enriched with provitamin A carotenes by plant breeding. These
strategies combined with a zinc strategy thereby address not only the vitamin
A deficiency problem in humans but may also address more effectively the
iron deficiency in humans than any iron fertilizer is likely to do. We advocate

introducing carotene-rich secondary staples and increasing zinc in diets by
fertilizer use and by plant breeding of major staples where appropriate.

2.5. Food systems strategies
Nutritional anemia (iron deficiency) is promoted, among other things, by
deficiencies of a number of other nutrients, especially zinc, iodine, selenium,
vitamins A, B12, C, and folate, and is reduced by synergistic interactions


Iron and Zinc Deficiencies in Crops and Humans

13

among these nutrients when their supply is increased in the range from
deficiency to adequacy.
Among various agricultural strategies, the Consultative Group on
International Agricultural Research (CGIAR) Global Challenge Program,
HarvestPlus, utilizes plant breeding to improve diets in target countries,
especially for resource-poor populations, using staple food crops as a
vehicle for delivering more micronutrients (principally iron, zinc, provitamin A carotenoids). The challenge is to minimize the number of genes
involved to accomplish this end (Graham et al., 1999). Another agricultural approach to help meet the challenge is supplemental use of fertilizers
where they have a comparative advantage, especially on soils inherently
low in these nutrients. So far, we have seen little prospect of breeding
for high selenium or iodine content (Welch, 1986), so fertilizer strategies
are appropriate for these (Cao et al., 1994, Welch, 1986) and for zinc as
already discussed.
To combine effectively the HarvestPlus strategy with the resources of
the fertilizer industry, we need to work within individual food systems
that collectively support the bulk of the populations at risk of micronutrient
deficiencies. Clearly, a fertilizer strategy will not sustainably solve iron deficiency or vitamin A deficiency in a target population. These can be solved

by breeding more iron-dense and provitamin A-dense staples, a primary
HarvestPlus strategy, but also by use of more zinc, iodine, and selenium
fertilizers where the soils of the food system are deficient in them (Graham
et al., 2007). Vitamin A must be addressed by breeding or by introduction
into the food system of an additional food crop naturally rich in provitamin
A carotenoids, such as orange-fleshed sweet potato. Often, where a food
system is struggling to meet basic expectations for calories to avoid starvation, an additional food requires that land be allocated for it and to achieve
this in turn means productivity needs to be increased on existing land. Thus,
emphasis on macronutrients must be considered an integral component of
any holistic approach to developing micronutrient-adequate food systems.
Besides selenium and iodine already mentioned, other minerals and vitamins are likely to be limiting for humans in particular food systems and may
require additional fertilizers (calcium, magnesium, copper, cobalt, boron)
and vitamins (from vegetables, cassava, potatoes, sweet potatoes, a little fish
or meat products); and a stable, economic food system must be capable
of including the preferred crops and providing at the same time sufficient
calories, and be both economic and socially acceptable. Integrating all this
requires successful deployment of expertise in several disciplines and
includes agronomic, fertilizer, plant breeding, sociological, and nutritional
expertise. Delivering on this complex agenda will be challenging, but once
a successful food system is established, it will be readily extended to all
comparable communities on similar soils and to new areas once their soil
and crop characteristics are defined.


14

Robin D. Graham et al.

3. Iron and Zinc Interactions in Human
Nutrition

3.1. Synergy or antagonism
Iron and zinc deficiencies in humans occur as a consequence of inadequate
dietary intake or, where intake is adequate, of low or impaired intestinal
absorption. Factors that decrease absorption include dietary inhibitors, such as
phytate or certain types of fiber, drugs or other chemicals, and interactions
between essential nutrients (Whittaker, 1998). The interaction between iron
and zinc has drawn particular attention. Meat is the best food source of
bioavailable iron and zinc, so in developing-country vegetarian populations,
iron and zinc deficiencies usually coexist. However, if additional iron and
zinc are to be provided together, it is important to evaluate whether, and if so,
how they interact biologically.
In the past, because of their chemically similar absorption and transport
mechanisms, zinc and iron were thought to compete for the same absorptive pathway since both are commonly absorbed as divalent cations
(Solomons, 1998). There are studies which demonstrated inhibitory effects
of zinc on iron absorption and vice versa. However, most of these studies
used high doses of soluble forms of iron and zinc that are not likely to be
found in food. Further, they were commonly given in a water solution or
administered in a fasting state, which further amplifies competitive (antagonistic) interactions. An additional limitation is the fact that most of these
studies used only serum or plasma zinc concentrations as a measure of zinc
absorption. Measurements of circulating concentrations do not necessarily
indicate true zinc uptake or status, and plasma zinc concentrations are
hormonally regulated (Lopez deRomana et al., 2005).
The probability of antagonistic interactions appears to be much lower
when zinc and iron intake are closer to “physiological” concentrations
(Lonnerdal, 2000). Further, a number of studies showed no negative effect
of iron fortification of food on zinc absorption and vice versa. Recently,
several studies provided evidence suggestive of positive interactions between
iron and zinc in absorption (Chang et al., 2010; Hininger-Favier et al., 2007;
Smith et al., 1999). All these findings support an hypothesis of possible iron
and zinc synergism, or at least no antagonism, when small or complexed

sources of these minerals are used together. This section summarizes findings
in order to shed some light on ideas about the relative significance of iron and
zinc synergy (as opposed to antagonism) in normal human nutrition.
An important condition for expression of synergy between nutrients, in this
instance, is that individual subjects be moving from deficiency to adequacy, or
perhaps more rarely, in the reverse direction. The review mainly includes the
studies that look at iron and zinc interactions when these nutrients are supplied


Iron and Zinc Deficiencies in Crops and Humans

15

in modest amounts (closer to normal consumption levels than those often used
in clinical trials), and/or chemically bound or complexed as in food.

3.2. Supplementation studies
Solomons (1986) proposed that chemically similar ions compete for the
same absorption sites in a common absorptive pathway; by his proposal,
a high concentration of zinc or iron could theoretically inhibit the absorption of the other. Our review of the literature suggests that his view is
supported only by studies using high doses of soluble ionic forms of iron and
zinc given together in unbound forms, that is, without binding ligands or
food. The summary review of Fischer-Walker et al. (2005) provided much
support for noncompetitive absorption of iron and zinc added together.
Findings from randomized placebo-controlled trials of supplementation of
iron and zinc separately, or in combination, in children under 5 years of age
and in women of child-bearing age, including pregnant women taking quite
high doses (Baqui et al., 2003), were included in the review. Opposing the
scenario of Solomons (1986), all trials showed no adverse effect of zinc on
hemoglobin or serum ferritin. One small trial even showed a positive effect

of zinc on hemoglobin and another positive effect on plasma ferritin.
Moreover, none of the trials showed a negative effect of zinc supplements
on iron status indicators and the studies looking at whether iron supplementation affects zinc absorption showed no adverse effect of iron on serum
zinc status. An additional benefit of zinc-with-iron supplements for small
children was lower rates of diarrhea (Chang et al., 2010; Smith et al., 1999;
Solomons, 1986), the last recommending joint supplementation of children
in Bangladesh for its benefits in reduced diarrhea and hospitalization. Further
studies have reported synergy between iron and zinc with quite high doses
(Harvey et al., 2007; Penny et al., 2004; Smith et al., 1999). Serum–zinc may
be taken as a valid indicator of zinc status averaged across all the individuals
in these trials, as it is on a population basis (de Benoist et al., 2007; Hotz and
Brown, 2004).
Contrary results were mostly confined to studies of short duration (Berger
et al., 2006) or studies on babies (rat pups) less than 6 months old whose
absorptive systems have not yet matured (Kelleher and Lonnerdal, 2006).
Recently, Dekker and Villamor (2010) performed a systematic review of
randomized trials that examined the effect of food-based zinc supplementation on hemoglobin concentrations in healthy children aged 0–15 years.
Their quantitative analysis showed no adverse effect of zinc on hemoglobin
concentrations and no evidence for effect modification by age, zinc dosage,
duration of treatment, type of control, and baseline hemoglobin status. The
authors concluded that there could be additional benefits of zinc supplementation among children with severe anemia or zinc deficiency. All these


16

Robin D. Graham et al.

findings clearly oppose the existence of a negative interaction between iron
and zinc delivered at low doses or with food.

3.3. Fortification studies show no antagonism
Iron deficiency is a common nutritional problem in infants and children and
to address it, weaning cereals are routinely fortified with iron. However,
the undesirable side effect of fortifying foods with iron, observed in some
studies especially in infants, is the possibility of inadequate absorption of
zinc to sustain their rapid growth (Ziegler et al., 1989; Lofti et al., 1995).
Fortification with reduced iron in a weaning food for 9-month-old infants,
both normal and anemic, over a wide range of iron:zinc ratios had no
adverse effects on zinc absorption unless given without food (FairweatherTait et al., 1995; Friel et al., 1998; Lopez deRomana et al., 2005) or using
zinc oxide in lieu of sulfate (Herman et al., 2002). These results extend
the earlier results of Davidsson et al. (1994) who used chelated iron
(FeNaEDTA) to prevent adverse effects of quite high iron fortification on
zinc absorption.

3.4. Zinc and anemia
Although zinc deficiency and iron-deficiency anemia were causally linked
(Prasad et al., 1963) in the case of a single individual, relevant literature on a
possible causal relationship between them and between the corresponding
elemental concentrations in blood has accumulated only more recently,
involving studies of the interaction between zinc and iron in dual or
multinutrient intervention studies and physiological and molecular studies
of the absorption sites in the human gut. Iron and zinc have a similar
distribution in the food supply, and the same food components affect the
absorption of both minerals, so nutritional causes of iron deficiency and zinc
deficiency are without doubt related. Additionally, over the years, a number
of data sets have clearly demonstrated a positive correlation between anemia
and signs of the risk of zinc deficiency in adult males, children, and pregnant
women (Ece et al., 1997; Ma et al., 2004). The correlations were stronger
in anemic than nonanemic populations. A study by Gibson et al. (2008) with
pregnant women in Sidama, Ethiopia (75% of the subjects were iron and

zinc deficient) showed plasma zinc to be the strongest predictor of hemoglobin concentrations (compared to plasma ferritin, gravida, status of vitamins B12 and A, and folate and C-reactive protein). The study of Smith
et al. (1999) also showed significant responses in serum hemoglobin to either
vitamin A or zinc treatment, or both together, and zinc concentrations had
direct effects on hemoglobin levels, more so in older children; in contrast,
the nil-zinc control group declined in serum hemoglobin levels over the
same 6-month period.


Iron and Zinc Deficiencies in Crops and Humans

17

A large number of studies show that anemic children are often zincdeficient, and zinc is shown to be a strong predictor of hemoglobin
concentrations. Moreover, iron supplementation, by itself, is not always
effective in treatment of anemia.

3.5. The regulation of hemoglobin levels
Iron deficiency has been reported to be the most common cause of
low hemoglobin concentrations. Consequently, provision of iron supplementation is the main focus of programs that aim to treat anemia. Increasingly, however, studies are showing the incomplete improvement of
hemoglobin after iron supplementation, especially in anemic children.
Allen et al. (2000) showed that, after 1 year of supervised iron supplementation, the children’s hemoglobin concentrations were not significantly
higher than those of nonsupplemented children, a result that could not be
attributed to short duration, noncompliance, or lack of iron absorption.
Many iron-supplemented children remained anemic (30% at 6 months
and 31% at 12 months), as was the case in other studies (Palupi et al.,
1997). In a meta-analysis of the efficacy of such iron supplementation
trials in developing countries, Beaton and McCabe (1999) concluded
that “there is a suggestion in the data that ‘something other than iron
may be operating to limit hemoglobin response and anemia control.”
Could this factor be zinc?

Zinc deficiency was implicated quite early. In 1976, Jameson proposed
that some refractory anemias of pregnancy are due to zinc deficiency. Low
serum zinc concentrations were found in the majority of 33 pregnant
women whose anemia did not respond to iron, vitamin B12, or folate.
In addition, 13 of 20 pregnant women selected for very low serum zinc
levels had hemoglobin levels indicative of anemia (<110g/L) (Jameson,
1976). Studies by Kolsteren et al. (1999) with 216 nonpregnant anemic
women 15–45 years old in Bangladesh and Alarcon et al. (2004) with
Peruvian children, both showed a positive effect of zinc or zinc plus vitamin
A delivery, with iron, on hemoglobin responses, with the added benefit of
less diarrhea. Zinc may increase vitamin A concentrations through promoting the production of retinol-binding protein, and in this way can redress
iron deficiency (Rahman et al., 2002). Nishiyama et al. (1996a,b, 1998) in
studies with 52 anemic women showed parallel zinc and iron deficiencies.
Marginal zinc deficiency possibly contributes to the manifestation of anemia, as the combined administration of ferrous citrate and zinc was the most
successful in increasing the concentration of iron, red blood cells, hemoglobin, and albumin levels. In cases of anemia in women endurance runners,
disabled patients, pregnant women, and in premature infants, combined
iron and zinc interventions helped in faster recovery from anemia
(Nishiyama, 1999; Nishiyama et al., 1996a).


18

Robin D. Graham et al.

3.6. Micronutrient deficiencies are occurring together
Deficiencies of iron and zinc remain a global problem, especially among
women and children in developing countries. Current intervention programs
address mostly iron, iodine, and vitamin A deficiencies, mostly as single
nutrient interventions, with fewer programs operating for other limiting
essential trace elements (Gibson, 2003). Whether there is a common underlying cause of these micronutrient deficiencies or whether one micronutrient

deficiency leads to another deficiency cannot be answered from such studies,
but it is clear micronutrient deficiencies are occurring together in many
regions of the world. A diet rich in phytate and low in animal proteins, as is
common in most developing countries, predisposes to insufficient intake and
absorption of both iron and zinc (Kennedy et al., 2003). Dijkhuizen et al.
(2001) showed that deficiencies of vitamin A, iron, and zinc occur concurrently in lactating mothers and their infants in rural villages in West Java,
Indonesia. In addition, Anderson et al. (2008) demonstrated a high prevalence
of coexisting micronutrient deficiencies in Cambodian children, with zinc
(73%) and iron (71%) as the most prevalent deficiencies.
If micronutrient deficiencies are occurring together, it is essential to treat
them together, rather than separately. The positive effect of doing so was
reported in a number of studies. Shoham and Youdim (2002) investigated the
effect of 4-week iron and/or zinc treatments on neurotransmission in the
hippocampal region in rats. Iron or zinc alone was not effective whereas
together they caused a significant increase in ferritin-containing mossy fiber
cells (cells important for memory and learning). This is the classical response to
the addition of two limiting essential nutrients acting together on a physiological or developmental pathway. Ramakrishnan et al. (2004) undertook metaanalyses of such randomized controlled interventions to assess the effects of
single vitamin A, iron, and multi-micronutrient (iron, zinc, vitamin A, vitamin B, and folic acid) interventions on the growth of toddler children. In their
summary of around 40 different studies, they clearly found greater benefits
from multimicronutrient interventions that they explained by the high prevalence of concurrent micronutrient deficiencies and the positive synergistic
effects between these nutrients at the level of absorption and/or metabolism
(e.g., vitamin A and iron, vitamin A and zinc, iron and zinc, all three)
(Ramakrishnan et al., 2004, Fig. 2). The results also suggest that competitive
interactions between iron and zinc are not a problem when zinc is included in
a multivitamin–mineral food-based supplement (Ramakrishnan et al., 2004).

3.7. Iron and zinc transporters in enterocytes of the
small intestine
Early studies on body iron balance revealed that humans have a limited
capacity to excrete iron so that the iron content of the body is tightly

regulated through control of absorption by the intestine (Donovan et al.,


Iron and Zinc Deficiencies in Crops and Humans

19

2006). Development of cloning technology has helped identify proteins that
are involved in iron movement into and across the human enterocytes.
Because most dietary iron is in the ferric (Fe3þ) form, it must be reduced to
ferrous ion (Fe2þ) via the ferric reductase, Dcytb (1) (Fig. 3) in order to be
transported by DMT1 (2) across the brush border membrane.
Once within the enterocyte, iron may be stored within ferritin (3) or
transported across the basolateral membrane and into circulation via ferroportin (FPN), also known as IREG1 (5). Basolateral transport of iron also
requires the iron oxidase, hephaestin (4) which oxidizes Fe2þ to Fe3þ prior
to its entry into the blood. In the past, DMT1 has been proposed as the site
for iron–zinc antagonism (Fleming et al., 1998; Gunshin et al., 1997), but
more recent studies show that DMT1 is an unlikely site for absorptive iron–
zinc interaction (Kordas and Stoltzfus, 2004). DMT1 was implicated in
intestinal iron absorption when it was identified as the gene mutated in
the microcytic anemia mouse and the phenotypically similar Belgrade rats
(Fleming et al., 1998). In these two animal strains, orthologous mutations in
the DMT1 gene resulted in severely decreased absorption of dietary iron
and low iron uptake by erythroid cells.
An earlier view of the main role of DMT1 was iron homeostasis. In that
view, the iron status of enterocytes strongly affects DMT1 expression and so
regulates the amount of iron transported into the mucosa (Tallkvist et al.,
2000). Although DMT1 is known to be an iron transporter, it was originally
thought that it also transported other divalent cations, including zinc.
However, in one of their experiments Lopez de Romana et al. (2003)

found no relation between serum ferritin and zinc absorption. Uptake of
metals by DMT1 is dependent on a cell membrane potential and redox
status, but when Kþ solution was used to depolarize the cells, changes in
iron uptake only were recorded, without changes in zinc absorption. This
clearly demonstrated that zinc does not depend on DMT1 to enter intestinal

Figure 3 A summary of the main pathway by which iron crosses the duodenal
enterocyte. DMT1, divalent metal transporter 1; DCYTB, cytochrome B oxidase;
FPN, ferroportin; Heph, hephaestin.


20

Robin D. Graham et al.

cells as thought earlier (Fleming et al., 1998; Gunshin et al., 1997) and is
unlikely to compete with iron for absorption (Sacher et al., 2004). Subsequently, a family of human intestinal transporters (ZIP) has been identified
as zinc transporters, indicating separate mechanisms for iron and zinc absorption. Some ZIP transporters (Zip14 in particular) may have iron-transport
activity (Liuzzi et al., 2006); however, Zip14-mediated iron uptake does not
seem to be essential in maintaining intracellular iron status (Lichten and
Cousins, 2009).
In an experiment with Caco-2 cell lines, Iyengar et al. (2009) examined
the mechanism of interaction of iron and zinc using kinetic analysis and
showed remarkable differences in Km, Vmax, and uptake of iron and zinc,
which negates the possibility of direct competition for a single transporter.
They also showed that zinc pretreatment modulates iron uptake, highlighting the importance of cellular zinc as a determinant of iron uptake. Western
blot analysis showed that zinc increases DMT1 expression which probably
explains increased iron uptake upon zinc pretreatment.
The earlier study of Kelleher and Lonnerdal (2006), where they investigated the effect of zinc supplementation on iron absorption in suckling
rats, showed that, although Zn supplementation had negative effects on iron

absorption during early infancy, this effect was completely reversed in late
infancy. This view reconciles some of the conflicting data in the literature.
They postulated that the difference is caused by DMT1 and FPN localization. During early infancy, DMT1 and FPN were located intracellularly.
This may be considered “immature localization,” but the possibility exists
that this reflects homeostatic control in response to high neonatal iron
stores, by internalizing DMT1 and FPN in the enterocyte (Trinder et al.,
2000) to prevent further uptake of iron. However, during late infancy, both
DMT1 and FPN were appropriately localized to the apical and basolateral
membranes, respectively. These age-dependent effects are consistent with
the earlier reported results of Smith et al. (1999). Additionally, liver hepcidin
expression was lower in zinc supplemented pups. These data indicate that
decreased iron absorption during early infancy is actually a consequence of
increased iron retention in the small intestine, facilitated through reduced
basolateral iron efflux and enterocytic iron trapping. Interestingly, by the time
of weaning, this effect is resolved, potentially as a result of the “maturation” of
iron absorptive mechanisms (Leong et al., 2003).
The idea, that DMT1 is inversely regulated through changes in enterocyte iron levels, suggests that during early-mid infancy, when enterocyte
iron is elevated, DMT1 expression should be decreased. However, the
Leong et al. study showed that DMT1 expression did not change according
to intestinal iron concentrations but rather in accordance to changes of
intestinal zinc concentrations. The fact that DMT1 is not inversely related
to intestinal iron concentration, but is positively associated with intestinal zinc
concentration, suggests that zinc plays a direct role in positive regulation of