Vitamin A is required for human growth. It plays an important role in the early embryonic development of all mammals, and in proper functioning of the immune system, the rod cells in the retina of the eye and mucous membranes throughout the body. Since mammals cannot manufacture Vitamin A on their own de novo, diet is the source of all human Vitamin A and pro-Vitamin A. A diversified diet with plenty of fruits and vegetables and animal products will amply provide enough Vitamin A — but for many people in the world, such a diet is beyond reach for all or part of every year. Some statistics:

• Roughly 400 million people in the world are at risk of Vitamin A deficiency.
• 100-200 million children are affected by Vitamin A deficiency.
• 1.0 to 2.5 million deaths per year of preschool children — up to 30% of total deaths in that age group — could potentially be averted by bringing Vitamin A deficiency under control worldwide.

Making rice an additional source of pro-Vitamin A readily available in the diet of poor and vulnerable populations is now a possibility, previously unattainable, using the technologies of agricultural biotechnology.

The Discovery of Vitamin A

Vitamin A, first on the list of Vitamins, was named in 1915 when it was found that a specific chemical substance in milk, butter and other foods was necessary for growth of humans and other animals. This essential micronutrient was shown to have several functions, and the earliest clinical signs of deficiency were identified as difficulty seeing in dim light and abnormal dryness of the eyeball, a condition known as xeropthalmia, or night blindness. But by the time rod cells are damaged to the point of causing night blindness, many other body functions have already been impaired, resulting in increased susceptibility to enteric and respiratory diseases, particularly among infants and young children.

The chemical designated "Vitamin A" was later shown to be retinol, a twenty-carbon alcohol. Human Vitamin A is provided through dietary sources in two ways. Roughly half is derived from pro-Vitamin A carotenoids of plant origin (e.g., beta-carotene, alpha-carotene, cryptoxanthin), which are converted to retinol by the intestines. The other half comes from ingestion of retinol itself, as part of animal products or as supplements. Beta-carotene, the most important of the carotenoids, is a forty-carbon photosynthetic pigment found in all green plants that can give rise (through central cleavage) to two molecules of retinol. However, due to digestive inefficiencies it takes six micrograms of beta-carotene in the diet to equal one retinol equivalent. Surplus retinol is stored in the liver of animals, and liver products (e.g., cod liver oil) are an excellent source of Vitamin A.

Vitamin A Deficiency and Children

In the early 1980s, researchers studying Vitamin A deficiency in Indonesia observed that young children diagnosed with mild night blindness were at a significantly higher risk of dying from other diseases in the next three to four months. Large-scale controlled experiments conducted in the region demonstrated that mortality rates were reduced by more than 30% by giving all children in a village a large dose of Vitamin A (15,000-60,000 micrograms of retinol equivalents, depending on age) every six months. This research was replicated and confirmed in other locations. Today, large-dose Vitamin A supplementation programs are supported by many governments and international agencies. Bangladesh, for example, has a "universal" Vitamin A supplementation program — that reaches roughly half of the target population. A lower dose of Vitamin A (e.g., 200 micrograms retinol equivalents) given daily produces even better results than bi-annual high-dose supplementation, with 40-50% reductions in mortality. However, daily administration is much more difficult and costly and is not widely practiced as a government program.

The success of the supplementation programs has clearly demonstrated that providing adequate Vitamin A decreases the incidence, duration and severity of childhood diseases, such as measles. They showed, too, that Vitamin A deficiency increases the risk in children of routine infections becoming severe infections that lead to death even before the signs of night blindness develop. And roughly half a million children still go blind each year due to Vitamin A deficiency. Supplementation programs are saving the lives and sight of thousands — but 100-250 million children remain severely affected by Vitamin A deficiency. More needs to be done for those not currently being reached.

Not surprisingly, it is the children of the poorest families, who spend 50-80 percent of their income on food and who depend heavily on low-cost, high-energy starchy staples such as rice, who are the most vulnerable to Vitamin A deficiency. Infants being weaned on rice gruel are particularly vulnerable, since they eat little else. Educational programs promoting home gardening can help, but this is not a year-round solution since Vitamin A deficiency tends to be seasonal, peaking in the late dry season, when fruits and vegetables are not available, stored rice is the principal food, and body stores of retinol have been exhausted.

The Area of Greatest Need: Rice

The greatest vitamin A deficiency occurs in South and Southeast Asia, where 70% of the children under five are affected. Here, rice is the staple food, accounting for 80 percent of caloric intake in some countries. Rice grain is the world's most important source of human food. It is a good provider of calories and protein, but rice scientists have long recognized its micronutrient deficiencies. Milled white rice contains essentially no pro-Vitamin A, and unmilled brown rice contains a very small amount (roughly 0.1 micrograms beta-carotene equivalents/gram - less than one percent of the daily requirement). Even when there are government programs to fortify rice, they do not reach the rural poor, who grow their own rice and mill it by hand pounding at home, or in small mills scattered throughout rural areas.

Like all green plants, rice has all the genes necessary to produce carotenoids. It does produce them in its green tissues, where they are essential pigments in the photosynthetic process and in small amounts in rice husks (which are removed, by polishing, from stored rice to prevent rancidity developing). In many other plants, such as yellow maize and yellow sorghum, carotenoids are also produced in nonphotosynthetic tissues such as endosperm — the starchy portion of the grain left after milling. It is the yellow color that indicates that the endosperm contains carotenoids. Rice breeders have for decades been on the lookout for a rice variety with yellow endosperm to cross with other strains as a potential new source of dietary pro-Vitamin A. But no rice containing carotenoids in the endosperm was ever found, and consequently traditional breeding of rice for increased pro-Vitamin A content was never possible.

Seeds of Hope

In the early 1980s, advances in plant molecular biology offered the promise of achieving genetic improvements in crops that could not be accomplished with conventional plant breeding. For the most part, however, such advances in crop biotechnology were not being applied to rice or other food crops of primary importance in developing countries. To help make sure the benefits of this powerful new technology would be available to poor farmers and consumers, the Rockefeller Foundation, beginning in 1985, committed roughly half of its agricultural funding to an international program on rice biotechnology. The primary objective of this program was to build rice biotechnology capacity in Asia, and an important part of it was funding the training of Asian scientists at advanced Western laboratories, where they invented techniques and worked on traits important for genetic improvement of rice - skills and knowledge which they then brought back home. About two dozen high-priority traits were targeted by the program, selected because they 1) would benefit poor farmers and consumers, and 2) were not readily achievable through conventional breeding. Beta-carotene production in rice endosperm was one of these targeted traits.

The Missing Phytoene. Foundation funding for research on this trait initially focused on two things: understanding genetic control of the carotenoid biosynthetic pathway in yellow maize, and biochemical analysis of rice endosperm. In a laboratory at Iowa State University, the dominant Y1 gene of maize — known to cause carotenoid production in the endosperm of yellow maize — was cloned. The Y1 gene was shown to code for phytoene synthase — an enzyme that combines two molecules of twenty-carbon geranylgeranyl diphosphate to form phytoene, the first forty-carbon compound on the carotenoid pathway. In another laboratory, analysis of rice endosperm showed that it contained geranylgeranyl diphosphate — but not phytoene.

In the early 1990s, two European scientists who were also interested in using rice biotechnology to benefit poor people in developing countries approached the Foundation. They too had identified yellow endosperm as a target trait. Dr. Ingo Potrykus of the Swiss Federal Institute of Technology in Zurich was a specialist in plant genetic transformation, and his lab was one of the first to genetically engineer rice. Dr. Peter Beyer of the University of Freiburg in Germany specialized in the biochemistry and genetics of the carotenoid biosynthetic pathway, using daffodil plants as a model system. These two scientists proposed to genetically engineer rice with daffodil genes to produce nutritionally significant levels of beta-carotene in the rice endosperm. At a Foundation-sponsored workshop, other scientists agreed that this task was difficult but achievable, and the effort was funded.

Daffodils Bearing Phytoene. Potrykus and Beyer initially confirmed that rice endosperm was capable of synthesizing geranylgeranyl diphosphate but not phytoene. Hence, to produce beta-carotene in the rice endosperm would require adding four enzymatic steps. The first task was to introduce the daffodil gene for phytoene synthase under the control of a rice promoter that would assure expression of the gene only in the rice endosperm. (A promoter is the regulatory portion of a gene that controls where and when the gene is expressed.) This was done, and the results were encouraging. The engineered rice plants produced phytoene in the endosperm at levels that would be nutritionally significant if converted to beta-carotene. Potrykus and Beyer then worked to introduce daffodil genes for the three remaining enzymatic steps required to convert phytoene to beta-carotene. One of these daffodil genes turned out to be unusually complex and difficult to work with, so they also tried a bacterial gene coding for an enzyme that could catalyze two steps in the pathway, including the step that was causing the problem. While initially intending to introduce the genes independently and combine them by crossing, they also tried adding them together in one sophisticated transformation experiment. The later approach provided the breakthrough they needed. Resulting rice plants containing two daffodil genes and the one bacterial gene carried out all four steps in the pathway and produced beta-carotene in the endosperm. The plants were normal, except that after milling, their grain was a beautiful golden yellow. Some of these plants produced amounts of beta-carotene, which at a daily intake of about 300 grams of rice could make a significant contribution toward meeting Vitamin A requirements.

Dissemination: IP Constraints and Testing

From the beginning of their research on Golden Rice, Dr. Potrykus and Dr. Beyer had intended to share it at no cost with public-sector rice-breeding programs for use by poor farmers in developing countries. However, as with nearly all academic research in crop biotechnology today, Golden Rice was produced using techniques that are patented in some countries and materials obtained under legal agreements that restrict further dissemination. As the inventors sought permission to share Golden Rice, a number of intellectual property (IP) constraints surfaced that appeared difficult to resolve. An additional concern was testing. Both the inventors and the Foundation wanted thorough testing for biosafety and nutrition, but such testing is expensive. They eventually concluded that the best and quickest way to overcome the IP constraints and test the product was to enter into a partnership with a company, Zeneca, that already had strategic and research interest in both rice and nutritional enhancement of food and consequently access to a large IP portfolio relevant to modifying the carotenoid biosynthetic pathway in plants, plus extensive experience in biosafety testing of crops and foods. This partnership arrangement required patenting their results.

Under the partnership agreement, Zeneca will help the inventors further modify their research product to produce a commercially viable variety of Golden Rice for dissemination to breeding programs, and will also facilitate biosafety testing and nutrition studies. It is known that the genes that have been added to the rice endosperm affect a small portion of the complex isoprenoid pathway in plants, and it will be important to analyze their effect on other metabolites to determine if there are health or environmental hazards — or, possibly, additional benefits. The allergenicity of the new gene products also needs to be assessed. It is possible that daffodils are responsible for what has been called "daffodil pickers' rash," and while it is unlikely, it needs to be confirmed that neither the daffodil genes nor the bacterial gene have introduced a new allergen to rice.

Drs. Potrykus and Beyer have the rights under this agreement to share Golden Rice with public-sector rice breeding programs to generate new Golden Rice varieties for use by resource-poor farmers in developing countries, defined as farmers generating less than US$10,000/yr. income from Golden Rice. This is known as the Humanitarian Project. Zeneca has retained all commercial rights in all countries and will donate support to the inventors in the Humanitarian Project. This agreement seemed to be the best all-around way to speed dissemination to public-sector rice-breeding institutions, and thus to poor farmers in developing countries, while assuring that all appropriate health and nutritional studies are conducted. Field testing for potential environmental impacts will still need to be done at the national and local level with all regulatory decisions being the responsibility of the relevant national authorities.