By David Dawe, Laurian Unnevehr
Golden Rice is genetically modified to provide beta-carotene in the rice
grain and it could potentially address widespread Vitamin A deficiency
in poor countries where rice is a staple. Political opponents have
viewed Golden Rice as representing the interests of multi-nationals and
as inherently unsafe for consumption.
Progress has been made towards adapting this crop to tropical-rice
growing environments, but it has not yet been introduced into farmer?s
fields. Efficacy and safety have not yet been fully tested. Substantial
work remains to target and deliver this intervention to Vitamin
A-deficient populations, and to overcome remaining resistance to this
technology. The political response to the on-going development of Golden
Rice is reviewed to draw lessons for biofortification efforts that
employ modern biotechnology. Within Asian countries, successful
development and delivery will require policy dialogue among
agriculturalists, health specialists, and advocates for the poor.

Introduction
Genetically modified rice that contains beta-carotene, widely known as
Golden Rice (GR), has not yet been introduced in any country. It was
developed to address Vitamin A deficiency (VAD) in low-income rice
consumers, but currently needs much more development and testing before
it can be introduced into farmers? fields. GR is the most famous
biofortification effort undertaken with modern biotechnology, due to the
initial publicity (e.g., the cover of Time magazine on July 31, 2000).
As such, it has been a lightening rod for the debate about the use of
GMOs in meeting nutritional needs. Thus, for this special issue on GM
foods and biofortification, a review of the lessons learned from the GR
case is crucial to understanding the political landscape for other
biofortification efforts. GR shows both the dramatic nutritional
benefits that can be achieved with use of modern biotechnology and the
considerable hurdles to eventual adoption and impact.

Below, this article presents the story of GR, including a review of the
controversies regarding its development and the literature estimating
ex-ante benefits, risks, and costs. The article closes with an
assessment of the current prospects for GR and lessons for other
biofortification efforts.

Impetus, Development, and Initial Reactions
The polished rice grain does not contain beta-carotene, a Vitamin A
precursor that the body converts into Vitamin A. In low-income
populations where rice is the primary staple, several micronutrient
deficiencies are chronic problems, including lack of Vitamin A. Such
deficiencies are particularly pronounced in small children, who need
greater nutrient density in food to meet their higher nutrient needs.
While the link between VAD and blindness captures public attention, VAD
also lowers immune response and increases the death rate from common
childhood diseases in developing countries, and as such VAD is often
considered primarily in terms of childhood-mortality effects.

VAD is widely recognized as a globally significant problem. The United
Nations Children?s Fund (UNICEF) (2004, p. 4) estimates that ?Vitamin A
deficiency is compromising the immune systems of approximately 40% of
the developing world?s under-fives and leading to the early deaths of an
estimated one million young children each year.? VAD is often a problem
where rice gruel is used as a weaning food. It is most important in the
poorest nations of the world, including most of South and Southeast Asia
where rice is the main staple, and that situation has not changed during
the past decade.

The idea of using rice as a vehicle to address micronutrient
deficiencies dates at least to the early 1980s. This idea emerged within
the Consultative Group on International Agricultural Research (CGIAR)
system and led to conventional breeding efforts to increase iron and
zinc in rice in the 1990s. Creating rice with beta-carotene content was
not possible until the advent of modern biotechnology techniques. The
Rockefeller Foundation (RF) funded the initial GR research through its
Rice Biotechnology Network, which was specifically established to
address the need for basic biotechnology research on this important food
crop that was likely to be ignored by the private sector in
industrialized countries. With support from the RF in the 1990s, Ingko
Potrykus at the Swiss Federal Institute of Technology and Peter Beyer at
the University of Freiburg, Germany, collaborated to introduce daffodil
genes into rice. The science was complex and cutting edge at that time,
as it was an early example of the use of pathway engineering. Their
success was hailed as a significant breakthrough in the application of
modern biotechnology, and the work appeared in Science (Ye et al.,
2000).

In conjunction with scientific publication, Potrykus appeared on the
cover of Time (July 31, 2000). Interestingly, the cover itself posed the
debate that has dogged this idea from the beginning: ?This rice could
save a million kids a year, but protesters believe such genetically
modified foods are bad for us and our planet.? The article appeared at
the height of the relatively new debate about the acceptance of GM
foods, largely triggered by trade conflicts between exporting countries
that had adopted the technology (the United States, Canada, and
Argentina) and the importing countries (largely the European Union [EU]
and other European nations).

Part of the publicity focused on the donation of intellectual property
(IP) rights for the GR technology so that it could be further developed
and adapted for introduction in the developing world. Apart from the
patent held by Potrykus, several enabling technologies were also needed
for further development. Potrykus formed a partnership with Zeneca
(which later became Syngenta Seeds AS after its merger with Novartis),
due to their history of work in carotenoids. Syngenta negotiated to put
together a package of rights to be donated for humanitarian use,
including patents held by Bayer AG, Monsanto Co., Orynova BV, and Zeneca
Mogen BV. The condition for use of these patents include a) that seeds
are developed for distribution to farmers in developing countries
earning less than $10,000 per year from farming and 2) that release only
takes place in countries with adequate biosafety regulations. The
donation of this package of IP rights for humanitarian purposes was
advertised as a model for the future transfer of this technology to
developing countries.

Negative reactions to GR were immediate and in many cases quite
emotional. The groups reacting included environmental advocacy groups
already engaged in arguing against GM crops in general, as well as
non-governmental organizations (NGOs) engaged in nutrition and
food-security issues in developing countries. In Southeast Asia, such
groups included Biodiversity and Community Rights Action (BIOTHAI) in
Thailand, the Cambodian Center for Study and Development in Agriculture
(CEDAC), the Development Research Communication and Services Centre
(DRCSC) in India, GRAIN-MASIPAG (Farmer-Scientist Partnership for
Development, Inc.) in the Philippines, and PAN-Indonesia and Policy
Research for Development Alternatives (UBINIG) in Bangladesh (BIOTHAI et
al., 2001). First-world opposition includes organizations opposed to GM
technology, such as Greenpeace, Friends of the Earth, and Food First, as
well as various groups in Europe (e.g., Institute for Science in Society
in the United Kingdom). Nutrition intervention groups do not seem to
have been as vocal in the debate.

Many of these reactions reworked long-standing concerns about Green
Revolution technologies and the commercialization of smallholder
agriculture and thus were not specific to GR. All of the opposing groups
agree that VAD is an important problem but objected to GR either as an
inappropriate or an ineffective solution. To summarize, the negative
reactions were based on these points:

Malnutrition is a result of poverty and interventions already exist to
address micronutrient deficiencies. Instead of developing GR, resources
should be focused on poverty alleviation, sustainable farming, and
proven strategies for nutrition intervention, such as supplementation
and diet diversification through backyard or community gardens.
GM foods are inherently unsafe to human health and the environment. GR
poses risks of these kinds and thus will not achieve its humanitarian
goals.
Rice is directly consumed by the poor, and thus the poor would be
?guinea pigs? for any human health impacts. Either GR will not provide
enough Vitamin A to do any good or will provide too much, resulting in
Vitamin A toxicity.
The IP arrangements are so convoluted that they do not preclude
commercial abuse and do not represent a useful replicable model. The
idea of ?donation? is an anathema to those who object to commercial
control of any agricultural IP.
GR is part of the continued use of ?Green Revolution? technologies that
are unsustainable and harmful to the poor.
It is not this case study?s purpose to debate these points but rather to
delineate issues that are under debate.

The virulence of the debate is surprising to someone who is agnostic on
the subject. On the one hand, scientists, multi-national seed companies,
and the CGIAR felt that they deserved credit for addressing a
humanitarian issue head-on and for donating technology for beneficial
use. Admittedly, multi-nationals were in need of positive publicity
following the negative reactions to first-generation GM crops in Europe,
and this was a strong motivation for their action on IP issues. But it
does seem that scientists involved were surprised to have their motives
questioned, as they genuinely believed in the positive humanitarian
potential of this technology. On the other hand, those opposed to GM
technology for ethical, environmental, or health concerns seem to have
felt that this represented a commercial conspiracy to win over the
public. They wanted to debunk this technology because it diverts
attention from potential negative impacts to potential positive ones,
thus changing the terms of the debate. They labeled it a ?Trojan horse?
for other biotechnology products in less-developed countries. For the
NGOs involved in poverty alleviation, it represents competition for
resources and influence. Thus, the debate has been quite hostile in that
each side accuses the other of acting in bad faith.

Subsequent Evolution and Current Status
The public attention to this potential new technology reinforced for its
advocates the need to address several issues in its development. It is
perhaps unfortunate that the first scientific breakthrough generated so
much attention when it remained fairly far removed from implementation.
The initial strains of GR utilizing daffodil genes did not contain very
much beta-carotene and might have had little impact on VAD in most Asian
diets. This point was noted almost immediately by astute advocates for
the opposition (e.g., Shiva, 2000). Later GR1 lines contain as much as 5
times more beta-carotene, although Dawe, Robertson, and Unnevehr (2002)
found that even this level may not have much impact in some populations
that are severely affected by VAD and for whom rice is not the only
staple.

Subsequent research has utilized cereal genes rather than daffodil genes
to generate much higher levels of beta-carotene in so-called GR2 lines
(Paine et al., 2005). In these lines, the enzymatic activity in the PSY
genes found within maize or rice is utilized to produce much higher
levels of beta-carotene in the rice grain. The new levels of
beta-carotene in GR2 lines are 20 times higher than the original line,
and these materials could provide all of the Vitamin A requirements for
children eating rice-based diets (Stein, Sachdev, & Qaim, 2006). This
improvement in beta-carotene content brought forth a few restatements of
the same general objections from those originally opposed to the
technology (e.g., see Greenpeace, 2005).

The initial framework to donate GR technology for humanitarian purposes
remains under the control of Potrykus and Beyer, who are advised by a
Golden Rice Humanitarian Board (Golden Rice Humanitarian Project
website). This Board does not make funding decisions. Much of the
current funding for development comes from USAID grants to the
International Rice Research Institute (IRRI), as well as country-mission
grants to National Agricultural Research Systems (NARS). Other funding
sources include the Bill & Melinda Gates Foundation, the Swiss
Development and Collaboration Agency, the Syngenta Foundation, and the
Rockefeller Foundation.1 Research collaborators include IRRI, as well as
NARS institutions in Bangladesh, Vietnam, the Philippines, India, China,
and Indonesia.

Field trials of the GR1 lines were conducted for the first time in 2004
at Louisiana State University (as US regulations allowed this step to
move forward more quickly than in any Asian country). The first trials
demonstrate that the crop is agronomically sound and may have higher
beta-carotene when grown under field conditions. Limited field trials
for the GR2 lines have also been carried out, but these GR1 and GR2
lines need to be crossed into appropriate indica varieties for use in
Asia.

In Asia, research samples of the initial GR arrived at IRRI in 2001.
Attempts were made to create new transgenic variants of IR64, BR29, and
other widely grown varieties in Asia with high levels of beta-carotene,
using the same genes for beta-carotene synthesis. Breeding work
(back-crossing) into leading Asian varieties was also undertaken with
the initial GR lines. With the availability of the new GR1 and GR2 lines
with higher beta-carotene content, the activities with the initial lines
were ended in 2003, and back-crossing work with the new lines began in
2004. As of yet, no field trials have been conducted in Asia, although
such trials of backcrossed GR1 and GR2 lines may be conducted in the
Philippines in 2007 and are possible in India in 2008 (G. Barry,
personal communication, June 20, 2007). These countries have relatively
well established biosafety guidelines, and have already approved other
GM crops for commercial purposes.

Beyond issues of agronomic viability, there are other development
efforts required to address issues of acceptance, safety, and impact.
Some taste tests have been carried out (Dubock, 2005), although not yet
in Asia. Bioavailability testing is currently ongoing at Tufts
University, using GR2 lines (Stein, Sachdev, & Qaim, 2007), and the next
phase will be a study in Asia (G. Toenniessen, personal communication,
August 1, 2006). Detailed work for biosafety risk assessment will
continue as the crop-development work advances. This risk assessment
work will be mostly carried out in Asia by NARS and will take several
years to complete.

Preliminary stability and retention studies are also underway in
Germany, the United States, and the Philippines in order to take account
of varying storage and cooking conditions in different socioeconomic and
cultural settings. For example, exposure to air, light, and moisture
during storage will vary across locations. As another example, rice is
parboiled in Bangladesh before eating. Conditions and food preparation
processes such as these could have large effects on the quantity of
beta-carotene in the cooked grain, so the results of these studies will
be critical for making a better assessment of the potential contribution
of GR to alleviating VAD. More work of this kind will need to be done as
more material adapted to local conditions is developed.

An Ex-Ante Analysis of Benefits, Costs, and Risks
As discussed above, the importance of VAD is widely recognized. Its
persistence is testimony to the limitations of current interventions
(discussed more fully in other papers in this special issue). GR has the
potential to reach important subpopulations that have not been targeted
by current interventions, most notably small children in parts of rural
Asia where rice is the predominant staple and weaning food. Several
different studies have now tried to assess the potential benefits of GR
using different economic methods and building their analyses on some
strong assumptions about nutritional benefits. Because GR is still so
far from actual production and consumption, little is known about
bioavailability, losses in storage or cooking, or many other factors
that would influence the actual delivery of Vitamin A. These studies are
beginning and will help define the deployment options for the product.

Costs of development will include basic research, adaptation to local
conditions, biosafety testing, and costs of consumer and producer
education, as well as any specific marketing regulations and future
maintenance breeding. In 2002, Dawe et al. made very crude estimates of
GR costs for Asia, which now appear to have underestimated the costs of
development and promotion. Stein?s (2006) estimates of the costs for
bringing GR to market in India are $21-28 million total for the next 30
years (discounted to the present), or $0.7-0.9 million annually. This
includes costs of development within India of $4.1-8.7 million, $2.2-2.5
million for regulatory review, and $15.6-30.7 million for promotion and
marketing. These estimates show that significant investments must still
be made to bring GR to farmers? fields in Asia, above and beyond
international research and development (R&D) to support understanding of
bioavailability and biosafety.

Every ex-ante study has shown benefits from GR, and these are usually
substantial and cost-effective. Dawe et al. (2002) found that the
initial GR strain would deliver very modest amounts of Vitamin A in the
diets of VAD children in one area of the Philippines. They also
estimated that the initial GR was very cost-effective compared with
other interventions, such as wheat fortification or supplementation.
Zimmermann and Qaim (2004) estimated the benefits in terms of saved
disability-adjusted life years (DALYs), and found potential reductions
in annual health-related costs of $16-88 million in the Philippines.
Anderson, Jackson, and Nielsen (2004) used their results to estimate
benefits of better health for unskilled workers in a general equilibrium
framework and found that health benefits potentially dwarf any
agricultural productivity benefits from GM rice, maize, and oilseed
crops in Asia.

The most recent study is by Stein (2006) for India, and it finds that
the newer GR would reduce the burden of VAD in India by 5-54%, depending
upon assumptions about adoption and who consumes it. The
cost-per-DALY-saved would be $3.40-35.47 for GR, which compares
favorably with alternative interventions. However, these
costs-per-DALY-saved for GR are significantly higher than equivalent
costs for biofortification of rice or wheat with either iron or zinc.
The latter biofortifications are easier to achieve and to promote, as
they involve less complex breeding applications and fewer
consumer-acceptance issues. The Stein (2006) study confirms that major
benefits are possible from GR, but also that it may be a more
challenging biofortification application than other potential
biofortification interventions.

What are the risks for supporters of GR? One risk that seems minimal at
this stage is that NGOs will be able to derail field testing on a large
scale. Several countries in Asia (including the Philippines and India)
have already approved GM crops for commercial purposes, and there are
procedures for such approval in many countries. If the data support the
effectiveness and safety of the new crops, it seems politically likely
that field testing will proceed. However, exporting countries like
Thailand and Vietnam are cautious about GM content that might reduce
export prospects. GR at least provides a visible marker (golden color)
that would facilitate market segmentation.

However, NGOs may have more influence on adoption by farmers and
consumers than on field testing. Many large NGOs are likely to support
GR if it is safe and effective, e.g., the influential NGOs in Bangladesh
such as Grameen, Bangladesh Rural Advancement Committee (BRAC), and
Proshika. These NGOs are unlikely to take a major lead in promoting GR,
but they will probably not oppose it and may lend some support to its
dissemination if there is strong evidence it will help the poor. But
many other NGOs advocate organic farming, and it seems unlikely that any
amount of evidence will convince them to support GR: their objections
are due to ethical or ideological considerations, not scientific
skepticism. Their influence is not to be ignored, and if GR is to be
adopted, educational campaigns targeted to farmers and the general
public will be of crucial importance.

There are several other risks that could be important. First, after
substantial investment, GR may not be widely adopted and will have
little semblance of the impact envisioned. Farmers who wish to sell it
in markets (most rice in Asia is traded in markets, not consumed at
home) may not want to take the risks of adopting a new variety (e.g.,
lower yield, susceptibility to pests and diseases) unless they are
compensated with higher prices or yields. However, such higher prices
would work against its incorporation into the diets of the poor,
possibly causing it to wind up as a niche product for rich consumers.
One possibility to counter these incentives would be to bundle the
increased beta-carotene content with other new desirable traits that
farmers find helpful. Alternatively, GR could be grown by poor farmers
for their own consumption, although again they may be discouraged by the
risks noted above. Furthermore, this strategy would limit the potential
impact of GR because the poorest of the poor typically buy much of their
rice on markets. Yet another possibility would be for governments to
subsidize the production and/or consumption of GR through public
distribution systems to encourage adoption by farmers and consumption by
poor consumers. However, it should be noted that targeted government
subsidies in agriculture and food are difficult to deliver without
substantial leakage of financial resources.

Second, GR may cause unforeseen health risks, particularly if it is the
first GMO to be widely consumed by children. This speaks to the
importance of extensive testing to ensure that GR has limited side
effects and, after storage and cooking, has enough bioavailable
beta-carotene to substantially reduce VAD. The lengthy approval process
still underway for commercialization of Bt rice in China shows the
concerns that governments have over GM food crops (as opposed to Bt
maize intended for animal feed or Bt cotton). These issues are discussed
more fully in the China and Philippines case studies elsewhere in this
special issue. If Bt rice is approved in China, this will most likely
smooth the path for approval of GR?certainly in China and possibly in
other countries as well.

Third, GR may be adopted and have a positive impact, but one that is
difficult to perceive or measure, so that little ?credit? is given to
the innovation. To remedy this situation, it will be important to
undertake education campaigns to inform any skeptical farmers,
consumers, or NGOs about the nutritional benefits of GR (assuming a
successful variety is developed) and its benefits for farmers. Such
campaigns will be important for ensuring widespread adoption as well as
giving due credit once it is adopted.

Many members of the educated general public in Asia are convinced that
most new technologies hurt farmers, especially if the corporate sector
is involved in the technology. To some extent, this is a legacy of the
Green Revolution, which is still viewed with skepticism by many even
though it increased productivity, has been adopted widely by farmers,
and was a major force in averting the widespread famines forecast by
many observers. As a specific example, many people assume that because
of the corporate sector?s involvement in the origins of GR, farmers will
need to purchase this seed every year. While this is true for many crops
in developed economies, no company at present has plans to commercialize
GR in developing countries. In addition, the technology to create GR was
donated by its inventors and private companies holding intellectual
property licenses so that any organization or farmer can freely
distribute or replant seed. Thus, farmers will be under no compulsion to
buy new seeds every year, but this fact will need to be clearly and
creatively communicated to the public. The public may also need to be
convinced that a reasonable share of benefits from adoption of GR goes
to farmers. Economists typically assume that adoption by farmers is
prima facie evidence that it provides them benefits, but this line of
reasoning is not necessarily convincing to others.

Future Potential and Lessons for Other Biofortification Efforts
GR technology still needs considerable research investment to be viable
in farmers? fields and to meet a rigorous standard for consumer safety.
Moving past regulatory hurdles will not be easy, and thus, this crop is
unlikely to play a role in meeting micronutrient needs before the next
decade.

From the political standpoint, there are still no advocates for this
technology within Asian countries. Ministries of Agriculture and NARS
are production-oriented by training and mandate; thus, they have little
interest in a project that diverts attention from production goals and
requires innovative cooperation with nutritional science. The IP
arrangements have not given participating NARS a sense of ownership or
control of this technology. Ministries of Health may find that
biofortification poses a threat to their traditional programs, although
it can potentially save lives and government expenditures in the long
run. NGOs have yet to embrace this technology.

To move forward, it seems clear that GR must be agronomically viable at
a minimum. To be acceptable to consumers and accomplish its nutritional
goals will require that countries make some strategic decisions about
implementation, adoption, and promotion. Such decisions include
desirable beta-carotene levels, target populations, desirable agronomic
characteristics, and methods for distribution and promotion. These
choices would be best informed if the health and agricultural
policy-makers can agree on the need for and potential benefits from this
technology and if NGOs who work with the poor embrace it. As this
approach to biofortification is relatively challenging, future
investments in research need to increasingly be driven by policy
dialogue.

The GR story provides guidance for other biofortification efforts.
First, any biofortification of a staple crop using GM technology will
likely encounter greater political resistance, as well as more
challenges in safety assessments and delivery, than non-GM approaches.
Second, any biofortification effort will need support and guidance from
NARS and NGOs within countries with nutritionally deficient populations
in order to be designed and targeted appropriately. These lessons for
future strategy are explored further in the final article in this
special issue.

Endnotes
1 Rockefeller has shifted almost all of its agricultural development
funding to Africa and currently has only some funding that is partly for
GR in Vietnam, the Philippines, and China. These are legacy grants from
earlier investments in the Rice Biotechnology Network.

www.checkbiotech.org