Biotechnology, Commentaries, Health October 25, 2017

Goodbye to Golden Rice? GM Trait Leads to Drastic Yield Loss and “Metabolic Meltdown”

by Jonathan Latham

by Allison Wilson, PhD.

GMO Golden Rice is promoted as a potent tool to alleviate vitamin A deficiency. However, Indian researchers now report that the genes needed to produce it have unintended effects. When they introduced the engineered DNA, their high-yielding and agronomically superior Indian rice variety became pale and stunted, flowering was delayed and the roots grew abnormally. Yields were so reduced that it was unsuitable for cultivation (Bollinedi et al. 2017).

Rice is a good source of certain nutrients but it lacks Vitamin A. In low-income households in certain countries, a rice-based diet can therefore result in vitamin A and other nutrient deficiencies.

Parent (left) Golden Rice (right)
Parent (left) Golden Rice (right)

According to Syngenta and certain public sector biotechnologists, vitamin A nutrition can be improved by introducing transgenes that specify enzymes in the β-carotene biosynthesis pathway (Ye et al. 2000; Bollinedi et al. 2014). β-carotene is one of several carotenoid precursors to vitamin A.

A short history of Golden Rice

The first GMO Golden Rices had either two or three introduced genes (in addition to a transgene specifying antibiotic resistance as a selectable marker). Plants with two had a daffodil phytoene synthase (psy) transgene and a phytoene desaturase (crtI) transgene from the bacterium Erwinia uredovora. These were transferred together as a gene cassette. Those with three had in addition a daffodil lycopene β–cyclase (lcy) transgene, co-transferred in a separate cassette (Ye et al. 2000). Each independently generated event consisted of one or more gene cassettes integrated at a particular location in the plant genome.

The public sector group who developed these subsequently created a new set of events using just the psy and crtI transgenes. They also changed the vector, the selectable marker and used different rice varieties. These changes were intended to make Golden Rice “amenable to deregulation” (Hoa et al. 2003).

As critics pointed out, the rice from both sets of events had very low carotenoid levels, less than 1.6ug/g.

Simultaneously, Syngenta Corporation produced a third set of Golden Rice events (called GR1). For GR1, Syngenta replaced the viral promoter of the crtI transgene used by the public sector group with a promoter designed to give enzyme activity just in the rice grain. They also removed the selectable marker. Total carotenoid levels in GR1 rice grains were higher: 6ug/g (Al-Babili and Beyer 2005).

The hulled grains of these different iterations of Golden Rice were yellow due to the β-carotene. However, even the best GR1 events still had β-carotene levels too low to alleviate deficiencies (Bollinedi et al. 2014).

Engineering Golden Rice GR2

Syngenta then made the fourth and current version of Golden Rice, called GR2. They introduced three genes as a single cassette. As before, two specified enzymes in the β-carotene biosynthesis pathway — a bacterial crtI gene and, this time, a maize version of the psy gene (Paine et al. 2005). The third gene specified a (non-antibiotic) selectable marker gene.

In GR2, the crtI and psy genes were again placed under the control of an endosperm-specific promoter. The GR2 cassette was introduced into the genome of the American rice variety Kaybonnet.

Syngenta researchers selected 23 independent GR2 events with between 9 and 37ug/g total carotenoid in the endosperm (Paine et al. 2005).

Golden Rice for India?

Through its Humanitarian Board on Golden Rice (HumBo), Syngenta made six of these GR2 events available to public sector breeding programs (Chikkappa et al. 2011). The Indian Researchers chose GR2-R1, presumably the most promising event, to breed a Golden Rice variety suitable for India (Bollinedi et al. 2017).

However, when they introduced GR2-R1 into the high-yielding variety Swarna, the plants with the GR2-R1 event were “drastically altered” (Bollinedi et al. 2017). GR2-R1 Swarna rice had pale green leaves, various root defects, and extra side shoots (tillers). The plants also flowered later, were half the height, and half as fertile. Yield was one-third of non-GMO Swarna.

Most of these abnormalities, it transpired, had been present in the original GR2-R1 plants donated by Syngenta.

Molecular-Genetic origins of GR2-R1 defects

DNA analysis revealed a partial explanation for these defects. The GR2-R1 DNA had inserted into and disrupted a native rice gene called OsAux1. OsAux1 specifies a transporter for the important plant hormone auxin. The researchers suggested this disruption explains some of the root and shoot defects. 

A second contributor to the defects seems to be that, although the psy and crtI genes had been specifically engineered so that their enzyme products would be present only in grains, the researchers found evidence the enzymes were unexpectedly functioning in GR2-R1 leaves.

Chemical analysis of leaves, stems, and flowering parts showed GR2-R1 plants had altered levels of three other key plant hormones: abscisic acid (ABA), gibberellin (GA3), and cytokinin.

To explain this the researchers proposed that the presence of the genetically engineered PSY and CRTI enzymes in leaves depletes a compound (GGDP) needed to make other plant biochemicals, in particular hormones and chlorophylls. Lack of chlorophyll would explain the pale leaves, while altered hormone levels would explain the other growth defects and the yield loss seen in GR2-R1 plants.

The researchers did not report whether metabolic and hormonal disruptions also occurred in the rice endosperm, where PSY and CRTI were intended to function.

The End for Golden Rice?

Golden rice has for over 20 years stood as the exemplar of a “good GMO” and proponents have blamed its failure to reach the market on “over-regulation” of GMOs and on “anti-GMO” opposition (Lee and Krimsky 2016).

This latest research suggests a different narrative. It shows that problems intrinsic to GMO breeding are what have prevented researchers from developing Golden Rice suitable for commercialization (Schubert 2002; Wilson et al. 2006).

The second great significance of this research, is that it implies engineering sufficient levels of β-carotene is disruptive to the basic metabolism of the plants.

“What the Indian researchers show is that the Golden Rice transgenes given to them by Syngenta caused a metabolic meltdown,” says Jonathan Latham, Executive Director of the Bioscience Resource Project. “The classic criticisms of genetic engineering as a plant breeding tool have always been, first, that introduced DNA will disrupt native gene sequences and, second, that unpredictable disruption of normal metabolism may result from introducing new functions. Golden Rice exemplifies these flaws to perfection.”

This then is the fundamental challenge of GMO metabolic engineering. It seems that making the intended metabolic changes (in this case increasing β-carotene levels) is the easy part (Giuliano 2017). The real challenge is to not make unintended changes by disrupting the many intersecting biochemical pathways—and thereby disrupting the complex plant processes that depend on them (Schubert 2008).

With their BioBricks approach to biology, Syngenta and their public sector allies have shown negligible understanding of these complexities, leaving it once again to non-GMO breeders to successfully enhance nutrient levels in plants (Andersson et al. 2017).

For years the quintessential example used to support GMO plant breeding, Golden Rice may now become “Exhibit A” for those wishing to critique it.

Highly Recommended Further Reading: The place of Golden Rice in the broader politics of rice, including its prospects in the Philippines (or elsewhere) are well explained in an important new paper by Stone, G. D., & Glover, D. (2016).”Disembedding grain: Golden Rice, the Green Revolution, and heirloom seeds in the Philippines” This paper is full of insights and up-to-date information.

Stone, G. D., & Glover, D. (2017). Disembedding grain: Golden Rice, the Green Revolution, and heirloom seeds in the Philippines. Agriculture and Human Values, 34(1), 87-102.


Andersson, M. S., Saltzman, A., Virk, P. S., & Pfeiffer, W. H. (2017). Progress update: crop development of biofortified staple food crops under HarvestPlus. African Journal of Food, Agriculture, Nutrition and Development17(2), 11905-11935.

Al-Babili, S., & Beyer, P. (2005). Golden Rice–five years on the road–five years to go?. Trends in plant science, 10(12), 565-573.

Bollinedi, H., Gopala, K. S., Sundaram, R. M., Sudhakar, D., Prabhu, K. V., & Singh, N. K. (2014). Marker assisted biofortification of rice with pro-vitamin A using transgenic Golden Rice lines: progress and prospects. Indian Journal of Genetics, 74(4), 624-630.

Bollinedi, H., Prabhu, K. V., Singh, N. K., Mishra, S., Khurana, J. P., & Singh, A. K. (2017). Molecular and Functional Characterization of GR2-R1 Event Based Backcross Derived Lines of Golden Rice in the Genetic Background of a Mega Rice Variety Swarna. PloS one, 12(1), e0169600.

Chikkappa, G. K., Tyagi, N. K., Venkatesh, K., Ashish, M., Prabhu, K. V., Mohapatra, T., & Singh, A. K. (2011). Analysis of transgene (s)(psy+ crtI) inheritance and its stability over generations in the genetic background of indica rice cultivar Swarna. Journal of plant biochemistry and biotechnology, 20(1), 29-38.

Giuliano, G. (2017). Provitamin A biofortification of crop plants: a gold rush with many miners. Current Opinion in Biotechnology44, 169-180.

Hoa, T. T. C., Al-Babili, S., Schaub, P., Potrykus, I., & Beyer, P. (2003). Golden Indica and Japonica rice lines amenable to deregulation. Plant physiology, 133(1), 161-169.

Lee, H., & Krimsky, S. (2016). The Arrested Development of Golden Rice: The Scientific and Social Challenges of a Transgenic Biofortified Crop. International Journal of Social Science Studies, 4(11), 51-64.

Paine, J. A., Shipton, C. A., Chaggar, S., Howells, R. M., Kennedy, M. J., Vernon, G., … & Drake, R. (2005). Improving the nutritional value of Golden Rice through increased pro-vitamin A content. Nature biotechnology, 23(4), 482-487.

Schubert, D. (2002). A different perspective on GM food. Nature biotechnology, 20(10), 969-969.

Schubert, D. R. (2008). The problem with nutritionally enhanced plants. Journal of medicinal food, 11(4), 601-605.

Wilson, Allison K., Jonathan R. Latham, and Ricarda A. Steinbrecher. “Transformation-induced mutations in transgenic plants: analysis and biosafety implications.Biotechnology and Genetic Engineering Reviews 23.1 (2006): 209-238.

Ye, X., Al-Babili, S., Klöti, A., Zhang, J., Lucca, P., Beyer, P., & Potrykus, I. (2000). Engineering the provitamin A (β-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science, 287(5451), 303-305.

If this article was useful to you please consider sharing it with your networks.

Print Friendly, PDF & Email
Comments 66
  • I think this is very good article, that simplifies even while communicating the wrong direction of rice research. From this experience, and results, it is obvious that India particularly should abandon all efforts towards Golden Rice. In principle, GM Rice project is misfit to the problems of vitamin deficiency. Agro-ecology ensures wholesome diet, there is no doubt.

    • No, agro-ecology will never see beta-carotene produced in rice endosperm. Income distribution and subsequent improved diet will see more people have more beta-carotene in their diet but in the mean time I think you’d agree that biofortifying rice is a wonderful solution until income distribution is sorted out.

      • Actually solutions have already been implemented and have “slashed” vitamin A deficiencies in the Phillipines: “Recognizing major public health problems with a range of micronutrient deficiencies, the country has undertaken programs including food supplements, nutrition education, and a law in 2000 requiring food fortification of common ingredients with vitamin A, iron, and iodine (Phillipines Department of Health nd). As a result, the incidence of childhood VAD in the Philippines dropped from a peak of 40.1 % in 2003 to 15.2 % in 2008 (Food and Nutrition Research Institution nd). VAD figures from the 2013 national nutrition survey have not been released, but as other types of malnutrition have continued to fall, the VAD rate must be expected to have fallen too.”
        To learn more about this see: Stone, G. D., & Glover, D. (2017). Disembedding grain: Golden Rice, the Green Revolution, and heirloom seeds in the Philippines. Agriculture and Human Values, 34(1), 87-102.

  • Bunkum. It is yielding well in trails in Bangladesh and the Philippines. Release is in 2018 and trials look amazing.

    • Obviously, if the Bangladesh Star says the peer-reviewed literature is bunkum then it must be so.

      • Correct. The article posted us trying to infer doubt regarding the Golden Rice project. But on reality it is looking like being a huge success. I think we should all be very excited. If you talk tp the breeders at IRRI you will find that is the case

        • Stone and Glover have spoken to IRRI — and their excellent analysis in the academic article “Disembedding grain” linked to above indicates that there are still many problems and hurdles ahead for Golden Rice, even in the Philippines. They include yield issues also the many biosafety and nutritional questions that remain unanswered.

    • Please read this informative article about Monsanto’s “Let Nothing Go” Policy around GMOs, Monsanto, and pesticides:
      It may help explain why Brian Duggan and Ed Regis are so concerned about the positive image of Golden Rice.

  • In response to Mr. Duggan, I would like to point out a few important things.
    1. The Golden Rice event used in the Phillipines is not GR2-R1, as was studied by the Indian researchers and discussed in this article.
    2. The Phillipine Golden Rice is GR2-E, which is the GR2 event given to HumBo that has the lowest levels of beta-carotene (see the Paine et al. or Chikkappa et al. references above).
    3. The article Mr. Duggan refers to is a newspaper article, not a scientific paper. It does not provide any scientific verification of its claims. Until independent scientists can verify the yields and the levels of beta-carotene, and until the data is made public, we do not have any scientific reason to believe the positive claims about GR2-E.
    4. Important missing data for both GR2-R1 and GR2-E are data on the levels of hormones, nutrients (like Vitamin E and other grain nutrients), potential toxins and other metabolites in the rice grains. For example, it might be that beta-carotene levels are increased but tocopherol (vitamin E) levels are decreased (you can get some idea of this possibility by looking at the link for GGDP in the article — which shows the complex intersecting pathways involved).

    • Actually Allison,
      1) it is
      2)Paine et al used GR1
      3) not sure how familar you are with agronomy by yield is ‘the great indicator’. Fact is beta-carotene levels are at levels now which will overcome Vitamin A deficiency and grain yields are high enough so that farmers will grow Golden rice in preference to current white rice
      4) no indication that other vitamin levels are changed in Golden rice.
      I advise you before you publish articles like this you visit IRRI’s website on Giolden rice. Many of the faux pars you made in your article could have been avoided. Regards

      • I think that Dr Wilson knows that Paine is GR1. Your item 2 does not answer the point.

      • If you read Paine et al. you will see that they produced GR2 in that paper. Both GR2-R1 and GR2-E are described within it.

        If you read Stone and Glover 2016 referenced in comments above (who discuss IRRI and their work) and the references therein (including some from HumBo and others who work with GR and have inside knowledge) and if you read the scientific articles referenced in my article you will understand that the GR project has focused on getting decent yields (and failed so far although they are still trying to achieve decent yields) and on improving beta carotene content. To the best of my knowledge, there is no published data so far on other vitamin, metabolite or mineral levels in the grain — for GR2-E, GR2-R1 or any other GR2 events. I would like to be alerted to any published scientific papers that address this issue.

        • The paper I referenced a few comments below is a published scientific paper showing acceptable total carotenoid levels in the GR2E event in BRRI dhan 29 rice, and no problems with the plant, in Bangladesh, through 2015 and 2016 field tests:

        • Ed – what you have linked to appears to be a PowerPoint presentation describing a version of GR2-E created for Bangladesh.

          1.It is not a peer-reviewed or published scientific paper. Published scientific papers have materials and methods sections to clarify how experiments were carried out, so that independent analysis is possible and so that claims could potentially be tested by other researchers. It is not even clear from this PowerPoint, for example, whether the different experiments were carried out on plants with one copy (heterozygotes) or two copies (homozygotes) of the GR cassette.

          2. GR2-E is the same event being used in the Philippines. It is not the same as event GR2-R1, the event assessed by the Indian researchers and discussed in my article.

          3. GR2-E has only 11.4 ug/g total carotenoid (Paine et al. 2005). It has not been tested to determine whether the genetic engineering process or the presence of the GR2 cassette have altered the levels of other nutrients in the grain. There are also no published agronomic tests to indicate that it will perform well in the field, including under drought or flood conditions, or when challenged by pests or pathogens. Stone and Glover (2016) discuss many other important issues that are not being addressed by GR researchers.

          • Re point 1, I agree it was not published in a peer-reviewed journal. However, it was published in the sense of being made openly available to anyone on the Internet. I understand point 2 very clearly. I also agree with point 3. However, what we have in this document is data from three acknowledged rice scientists at two of the world’s top rice research institutions. Their data suggests that you were premature in pronouncing a death sentence on GR on the basis of a transformation event that has been known to be faulty since at least 2009, at which time the Humanitarian Board made GR2E the lead event. Everything reported in the presentation is positive and promising, even if it is incomplete.

          • Unfortunately without the materials and methods and all the other missing information we do not know if these experiments or demonstrations were carried out in such a way as to hide defects or to miss them altogether.

            Please see my answer below to Brian D. for all the other missing data we need to have before we can determine if GR2-E is truly “promising” or not.

  • Dr Duggan to you thanks. PhD in crop physiology and genetics with 23 years of R&D experience. Anyway I suggest you check with breeders at IRRI. They haven’t published any data from these trials due to attacks from NGO’s but I can assure you that if you do you will see that the newspaper article frpm Bangladesh is correct. Expression levels of beta-carotene in the rice is high and just as importantly the yields are as high or higher than rice hybrids or varieties in the markets where it has been tested. Again release is planned for 2018. Odd that you have no comment from IRRI regarding the progress of Golden Rice. Thanks.

    • For recent updates from IRRI on how ready/unready GR is for release, see Stone and Glover 2016.

      This link from the IRRI website on 25 November 2017 says “The first high yielding varieties containing the GR2E Golden Rice trait are anticipated to be available to farmers towards the end of this decade.”

      It is interesting to see the constant new promises on GR “readiness for release” – first with each new iteration of GR – and now with one after the other of the GR2 events. An interesting article by Adrian Dubock (2014) of the GR Humanitarian Board describes how GR2-G was initially chosen as the most promising event, with GR2-R as the backup (p.78).

      Now, after so many prior failures and broken promises, GR proponents are hoping that GR2-E can somehow be more successful than its predecessors. So far I have found no peer-reviewed published papers allowing an independent assessment of GR2-E’s molecular, agronomic or nutritional qualities to address whether GR2-E can live up to these new claims.

  • Beta carotene expression levels in the Golden Rice strains used in the breeding program at IRRI are high as are the uptake levels in humans.

    • Dr Duggan! Do you have any special knowledge about this project, or just what is in the peer-reviewed literature? I ask because you seem remarkably confident about somethings for which there is very little published research? Second, the peer-reviewed paper you cite states that GR has up to 35ug of Beta-carotene per g. However, the authors dont say which GR they used but, since they fed 200g per person and that gave 1.5mg of Beta-carotene (to take their higher numbers) then that is about 7.5ug per g in the rice they fed the treatment group. Low enough that 6ug was deemed by Syngenta too low so that they then went ahead and made GR2. Am I missing some reason why you think that is high?

      • Hi Jonathon. Well I wouldn’t call it special knowledge, but simply what is 1) in the peer reviewed literature, and 2) what is on IRRI’s webpage. There is plenty of citations of peer-reviewed journal articles pertaining to Golden rice published there, including their data on how 1 cup of rice per day will provide the dietary requirements of a child suffering from Vitamin A deficiency in the developing work. I am just incredulous that the author of this peice never bothered to reference IRRI who is quite successfully developing this rice. And to your point, yes, anecdotal evidence, in the form of a newspaper article, is more credible as it is demonstrating what is happening in the real world. This piece is simply trying to sow the seeds of doubt. Well, the newspaper article proves that those fears are unfounded. Cool huh!

    • Brian Duggan – This paper you link to provides no information about whether GR (regardless of whatever GR version or event was actually used in the experiment) can help malnourished adults, much less malnourished children.

      From your link we see that “Golden Rice servings of 65–98 g (130–200 g cooked rice) containing 0.99–1.53 mg β-carotene were fed to 5 healthy adult volunteers (3 women and 2 men) with 10 g butter.”

      This paper only provides information about an unspecified GR version/event and its effects on healthy adult volunteers. It does not tell us whether β-Carotene derived from Golden Rice is effectively converted to vitamin A in malnourished humans, or those not eating it with 10g butter.

  • Golden rice is, and always has been, a high tech answer to a sociological problem. β-carotene has to be the second most abundant molecule on the planet. If it is missing from the diet then is going to be missing in the diet of the landless poor. Give them the land rather than GM rice. No problem the of Vit A deficiency

    • No, not necessarily Dave. It is not the ‘landless poor’ but rather the urbanised poor. It is a social issue, and one that we should absolutely address in terms of better income distribution etc but that can’t be solved in the short term. Many of the people who suffer from Vitamin A deficiency are urban poor who have very low income. A better diet would improve their lot in life, but all they can afford is rice. On such a diet they become Vitamin A deficient. Supplements degrade and expire and don’t often reach their target. The best short term solution is to bio-fortify the one thing they do eat with beta-carotene (so they can’t develop Vitamin A toxicity). Fantastic short term solution until we can solve income distribution.

      • Well, Dr. Duggan, I respectfully suggest that you are wrong. The “Fantastic short term solution ….” has been at least 20 years in the coming. In that time post revolutionary Cuba had completely reorganised the island’s agriculture, in the face of American embargo. It included intensive,”urban”, horticulture specifically designed to provide water-soluble vitamins in the diet, a move away from the cultivation of sugar and tobacco as cash crops, and a massive drop in the consumption of meat in the Cuban diet – all recognised as tennets of todays, sustainable, agroecology as distinct from the mass commodification of cereals which provide highly profitable, nutritionally poor, pap for the poor in most of the world, and highly profitable animal feed to produce meat for an increasingly obese population in “developed” countries.
        Golden rice is an irrelevance, a sticking plaster at best. It represents a justification for GM technology that appeals to a sense of altruism and reflects a world view in which, western/high-tech, science can be used to “solve” a problem that does not exist. The poor who exist on an exclusively rice diet are perfectly able to balance their diet given the education and opportunity to do so. It is the wrong answer to a social problem. If it ever does “come to market” then I predict it will cause the sort of social chaos represented by the introduction of bt cotton into India.

  • Glenn Davis Stone, who does have an inside track, writes, in his 2016 paper with Dominic Glover, Disembedding Grain (, that
    “Golden Rice……has failed to reach Philippine farmers because it has yet to meet standards of productivity.”

    • No the hold up moreover was that the first gene, GR1, did not express enough beta-carotene. This latest incarnation, GR2, produces enough beta-carotene such that a child only has to eat a cup of cooked Golden rice to receive half their daily requirement to not develop night blindness. The other complication is yield; if grain yield is not sufficient then farmers will not grow it and despite the success of the product in delivering beta-carotene the project will fail for economic reasons. Fortunately this has been overcome and the yields are as high or higher than current varieties and hybrids. I am sure you are looking forward to its launch next year as much as I am.

      • Brian you forgot the holdups due to the failures of events GR2-G and GR2-R, the first choices for GR2 release (Dubock 2014).

        Now you are asking us to believe that somehow GR2-E, the latest GR2 being promoted, will do what none of the other GR2 events have been shown to do:
        1. Exhibit high yield and agronomic quality, with no unintended defects.
        2. Show no unintended alterations to other nutrients or anti-nutrients.
        3. Produce sufficient levels of beta-carotene to address vitamin A deficiency, when grown in farmer’s fields.
        4. Retain high enough levels of beta-carotene to address vitamin A deficiency, even after weeks or months in storage, at normal storage and transport temperatures.
        5. Demonstrate that the beta-carotene in GR2E can be efficiently absorbed and converted to Vitamin A in malnourished people.

        Given the failure of previous GR genetic engineering attempts to achieve anything like this, we cannot believe such improbable claims without peer-reviewed published data and verification from independent researchers.

        • IRRI applied to the Philippine Department of Agriculture for a biosafety permit to use GR2E as human food, on Feb. 28, 2017. This application was accompanied by more than a thousand pages of supporting materials, including many peer-reviewd publications in the application itself, together with mountains of data. You can access these documents at:

          In view of the above, it is silly to speak of “The End of Golden Rice,” when al you have evidence for is the end of GR2R Swarna.

          • Thanks Ed, some of these documents do have useful information about GR2E in them. The “compositional assessment” indicates significant unintended differences in “minor fatty acids” between GR2E and the near isogenic control, in each of the two years that were tested. The rainy season showed more overall unintended differences (10 sig. difference/69) while the dry season showed 4 significant differences/69). Also, the fact that the GR2E cassette was backcrossed into several different genetic backgrounds and the background that gave the lowest levels of beta-carotene after backcrossing was the one chosen for further analysis (7.31 ug/g in background PSB Rc82) does lend support to my theory that high levels of expression of the transgenes cause either metabolic, yield, or other problems for the grains/plants. I did not see any yield or agronomic data in these documents. The insertion site data was interesting as it indicated that new ORFs were generated and raised the possibility they might be expressed.

            Like Ted Greiner, I would like to say Goodbye to Golden Rice because I think it is a slow, extremely costly, unscientific, and highly unlikely way to address malnutrition and public health. See Greiner’s article and other comments below for inexpensive and known-to-work community based solutions.

        • Re Point 5 above, the beta carotene in GR is no different from the beta carotene in other sources. And in fact, peer-reviewed scientific studies have shown that beta carotene can be converted to vitamin A even in malnourished children. In a 2007 study of 116 Filipino schoolchildren, some of whose diets were so poor that there was a high prevalence of subclinical vitamin A deficiency among them, a team of researchers found that carotene-rich plant foods improved both their blood serum vitamin A levels as well as their stores of liver vitamin A even when their diets included extremely low amounts of dietary fat.
          “Regardless of the amount of fat ingested by the Filipino children per meal or the total amount of fat they ingested per day, their mean serum provitamin A carotenoid concentration, total body vitamin A pool sizes, and liver vitamin A concentrations were increased similarly,” the authors reported. “Thus the dietary fat requirement for optimal bioavailability and effectiveness of plant carotenoids is minimal.”

          J D Ribaya-Mercado, et al.: “Carotene-rich plant foods ingested with minimal dietary fat enhance the total-body vitamin A pool size in Filipino schoolchildren as assessed by stable-isotope-dilution methodology.” Am J Clin Nutr April 2007 85: 4 1041-1049.

          • ED- The article you link to is not about golden rice nutrition, but it does support the excellent points made in some of the other comments below, – that providing people with the ability to grow (or purchase) a variety of carotene rich vegetables (eg. carrots, bok choy, squash and cabbage, as used in the study), when coupled with an adequate diet, including some amount of oil, could “enhance serum carotenoids and the total-body vitamin A pool size and can restore low liver vitamin A concentrations to normal concentrations.”

            Another very interesting article that addresses some of the wider issues of vitamin A deficiency is:
            Greiner, Ted. “Development. Malnutrition, Vitamin A.” World Nutrition 6.11-12 (2015): 792-810.
            It is well worth a read because it details the considered opinions of many researchers who have worked on the ground for decades trying to eliminate malnutrition (including Vitamin A deficiency) and the problems associated with it. These researchers question the rationale, science, and “success” of expensive top-down approaches (which would include genetically engineered crops) and instead point out the real success of community food based programs, including homesteading and school garden programs, to ensure diverse healthful diets and end malnutrition. This is a very useful article for learning about the big-money politics driving top-down initiatives like VAC (vitamin A capsules) and Golden Rice.

  • Why not just grow carrots instead? Because carrots cannot be patented?

  • In reading this interesting article and the back and forth discussion that followed, no one mentioned the fact that vitamin A, along with E, D, and K, require dietary fat for proper digestion and absorption into the body. So providing a crop with higher beta-carotene content will not solve the vitamin A deficiency unless dietary fat is also available.

  • My understanding is that it would be a far better solution to help people to grow a varied crop of Vitamin A rich food plants that are also supplemented with animal protein with high levels of Vitamin A, chickens and their eggs and goats and pigs.
    It seems that a global effort in training people in permaculture would be a far better long term solution and putting aid money into this kind of education.
    All other solutions such as food aid and GM crops appear to be pointless short term solutions that really lead to nothing in the end.

  • Regarding the Bangladesh experience, everyone involved in this discussion should read:

  • Ed – as I mention above, this is a PowerPoint presentation you link to and not a scientific publication, so it does not have anything useful to contribute to the discussion.

    • I really don’t see why it doesn’t. Scientific communication occurs in a variety of ways and contexts outside of publication in peer-reviewed journals. It is common for scientists to announce results at conferences, in which they make liberal use of Power Point presentations, well before writing up those results for submission to a journal. Audience members take the results so presented seriously enough, and indeed find that they do advance the discussion. Such procedures are absolutely routine in the sciences and so I do not see why the data presented in the Power Point I referenced should be treated any differently and be summarily dismissed as you have done above.

      • Ed- We certainly can ask questions that arise from the powerpoint you link to.
        For example why do slides 27 and 28 only have n=3 for wild-type control and null segregants – won’t this make it difficult to find a statistical difference if there actually is one? Three plants measured is not very many.
        Also slide 18 seems to show that the GR2-E plants measured had a very wide range of total carotenoid levels — will this be the case for farmers — their GR2-E crop plants could have any of a wide range of carotenoid levels? What is determining whether a GR2-E has very very low levels of carotenoid or higher levels?
        On slide 17 It looks like there are a wide range of yields for the GR2-E plants tested. And does the yellow box indicate the yield range of control plants or does the red arrowhead indicate the yield of the control plants? In either case it looks like for the most part yields of GR2-E plants are less than the control plants, in that slide.
        On slide 28, the GR2-E plants have substantially less amylose as % of starch than the controls — why is this? This difference is statistically significant, according to the powerpoint.
        And slide 23’s “Gel image” seems to be mislabeled for various lanes. Either that or there are some unexplained bands present and missing that need explaining.

        So you see — just looking at the power point raises numerous questions about data and possible errors — and without the written explanations and the materials and methods that are present in a scientific paper or report — we can only speculate…..That is why I keep insisting on published publicly available data — so people can examine the data and its quality for themselves.

  • Hi Dr. Wilson, I’m trying to understand your explanation about the engineering development of golden rice here. In the short history section you mention about the first Golden Rice, third and fourth set, what’s the most significant differences from the first to the fourth set beside the changes from daffodil to maize that’s being used? Thanks before for your respond.

  • Thank you for your question. Al-Babili and Beyer (2005) describe the various sets of Golden Rice events. Below I summarize the main differences:

    Public sector 1. The first so-called public sector GR was produced by Ye et al. (2000). These lines were created using Agrobacterium-mediated transformation of a Japonica rice variety; the CaMV viral promoter was used to regulate the bacterial crtI gene and the antibiotic-resistance marker gene. The psy gene from daffodil was regulated by the Gt1 promoter. These lines had up to 1.6 ug/g carotenoid in the endosperm.

    Public sector 2. When the second public sector GR were lines developed (using the pCaCar vector) the differences were that Hoa et al. (2003) transformed Indica and Japonica rice lines. Also they screened for simple insertions because the original Ye et al. (2000) line with the highest carotenoid levels had “multiple integrations and recombination events.” These lines used the same bacterial crtI and daffodil psy genes and promoters as before. However, they used the sugar-based selectable marker gene pmi instead of an antibiotic resistance gene. The three best lines resulting from this effort had 1.2ug/g carotenoid (Japonica) and 0.4 and 0.8 ug/g carotenoid (Indica).

    Syngenta GR1. Japonica cultivar Cocodrie was transformed with Agrobacterium. No CaMV viral promoters were used. Instead the same crtI and psy genes used before by public sector scientists were both regulated by the Gt1 endosperm specific promoter. No selectable marker remained in the GR1 plants. GR1 events had carotenoid levels up to 6.0 ug/g.

    Syngenta GR2. Japonica rice cultivar ‘Kaybonnet’ was transformed using Agrobacterium. They used same crtI bacterial gene regulated by Gt1 promoter as used to make Syngenta GR1. However, they used a maize psy gene (instead of daffodil) that was regulated by the Gt1 promoter. A pmi selectable marker gene regulated by a maize Ubi promoter was used and this was present in the final GR2 lines. Plants with up to 37ug/g total carotenoids were produced.

    • Our disagreements aside, I must say that this and your “Short history of Golden Rice” are excellent characterizations of the four different Golden Rices.

      • Allison, Having done some additional work on this I notice what seems to be an error (or at least an inconsistency) in the term you employ for the carotene desaturase gene used in three of the GR versions. You several times use ctrI for this (which is not correct, so far as I can tell), but in one case (Public sector 2, just above), you give the term correctly, as crtI. (Paine, 2005; Al-Babili and Beyer 2005, 566, Fig. 1, 567, Fig. 2.) Or am I missing something here? (Easy to confuse the two.)

        • Ed, thank you very much for alerting me to the ctrl typo/error. You are right, it should be crt1. If possible, I will have the mistake corrected throughout.

          • Not crt1, but crtI, i.e., the last character should be an upper case letter I, as in India. These sans serif fonts make this so confusing!

  • Golden Rice release nears in Bangladesh

  • Sorry, I’m going to have to be Captain Obvious here; If the answer to vitamin A deficiency is in our food, and knowing the poorest can’t afford meat, therefore it must come in a vegan food soiurce, then here it is. Any seed, almost, can be soaked and sprouted and greened up to be a microgreen and guess what is in the live green? Beta Carotene, misnamed and called legally for some reason Vitamin A (which it really is NOT, vitamin A truly is only in animal products and is a fat soluble nutrient). Beta Carotene is a Provitamin and must be eaten best in vegetables like yellow and orange squash and carrots, cooked with butter and the person’s gut system has to be able to digest it to make the conversion to true vitamin A. if any of these factors are lacking, the conversion is inefficient and not likely to happen with any great usefulness. Carrots are among the easiest plants to grow, and Please explain to me why it wpould be so hard for poor people to grow these. Also, greens being easier to eat raw than squash, [please explain to me why people who have seeds at all like rice, or beans could not soak them, sprout them and green them up?

    • Well, one Obvious answer is that many poor people suffering from VAD live in large cities, do not have feee access to farmland, and need large amounts of leafy green leaves to eat every single day, despite seasonal weather fluctuations and consequent interruptions in the supply of fresh produce, not to mention all the other impediments that even large commercial farming operations face all the time. Further, many of these poor people affected by VAD are hardly in the best of health, they are sick and not full of the energy required to engage in labor-intensive, backbreaking activities such as growing, harvesting, and preparing their own food.

      • Ed – As mentioned before, there are many good papers from experts in the field that address effective ways to eliminate any real cases of Vitamin A deficiency. Read: by the late Cornell Nutrition Professor, Michael Latham. He discusses how sustainable methods would include making available: red palm and other vitamin A-rich plant oils, combating disease such as measles and parasite infestation, and supporting breast feeding and plant-based diets. Vitamin A rice is at best a distraction, at worse, a health hazard. This is still a real possibility as the altered biochemistry of GR2 plants has never been carefully examined and so far not even a superficial metabolomics study has been carried out and published. It is interesting to note how rapidly the beta-carotene in GR2-E is broken down and how this alters the level of apocarotenoid products are in GR2-E, again with unknown health implications. And are you suggesting that rice is less back-breaking to grow than a variety of plants? Or that supplying beta-carotenoid containing vegetables or suitable oils to city people is harder than supplying them with as yet untested unproven GR2-E?

        • This is the paper that shows the rapid degradation of beta-carotene in GR2-E: Schaub, Patrick, et al. “Nonenzymatic β-Carotene Degradation in Provitamin A-Biofortified Crop Plants.” Journal of agricultural and food chemistry 65.31 (2017): 6588-6598.
          This paper is not free and open access but here is a link to the paper’s table showing rapid beta-carotene degradation and the levels of apocarotenoids in GR2-E in the Kaybonnet genetic background:

        • Allison, I am sure there are many different ways of eliminating VAD, including eradicating poverty, supplying vitamin A capsules and lots of red meats, eggs, and green, leafy vegetables for free to the millions of people who might benefit from these and multiple other methods. Equally, making Golden Rice available is also one of those methods. You have not shown it is a health hazard. To say it has unknown risks is to say nothing. If they are unknown, then how are they risks? Perhaps there are also unknown health benefits. That possibility has the same epistemological status as repeated unsupported claims of unknown risks. Death is one of the known risks of VAD, and blindness is another.

          • Ed – We know that engineered alterations to important biochemical pathway can have potentially harmful unintended effects We also know the genetic engineering techniques themselves cause potentially harmful unintended effects For this reason it is even more important to test for unintended metabolic effects in GR2-E, yet this has not been done. Those concerned with the quality of the GR product and its safety advocate for adequate testing.
            Unlike GR, the methods I mention above are known to eliminate Vitamin A deficiency and its attendant problems. And yet, unlike the others, GR has taken over 20 years and many millions of dollars to develop and it still is not ready or even likely to work. Furthermore, those who work in the field do not advocate GR as a solution:
            For those who want to know more about the efficacy and safety of GR2-E, in her submission to the Australian Ministerial Forum on Food Regulation, Madeline Love does an excellent job of detailing the failure of developers and regulators to adequately test GR2-E before approval In it she details the extremely low levels of beta-carotene and the relatively high levels of apocarotenoids in GR2-E. She notes, even if a normally yielding line can be developed, the data so far indicate that due to low levels of beta-carotene and its rapid degradation, GR2-E would be very unlikely to have any significant impact on alleviating Vitamin A deficiency. The potential biosafety problems with high levels of apocarotenoid degradation products are also discussed in this submission. So, with the data we have so far, GR2-E looks to be all risk and no benefit.
            Schubert, David R. “The problem with nutritionally enhanced plants.” Journal of medicinal food 11.4 (2008): 601-605.
            Wilson, Allison K., Jonathan R. Latham, and Ricarda A. Steinbrecher. “Transformation-induced mutations in transgenic plants: analysis and biosafety implications.” Biotechnology and Genetic Engineering Reviews 23.1 (2006): 209-238.

  • In addition to having been approved by New Zealand, Golden Rice has now been approved also by Canada. Allison,it seems that your pronouncing a death sentence on Golden Rice is a bit premature.

  • ED- These are in essence fake approvals as they are in countries that do not intend to grow or eat GR2-E, see It appears these countries have approved GR2-E to please the GMO industry and to put pressure on countries like the Philippines to approve GR2-E for cultivation, even though a final GR2-E product is not yet available and no GR2-E line has undergone adequate testing for efficacy and safety. What they show is not that GR2-E is ready to be grown or is safe, but rather that the regulatory agencies of these countries are acting irresponsibly and politically — approving a GR2-E rice without appropriate safety or efficacy testing. And these governments are approving GR2-E based on data that is not from actual lines that would be grown, as breeding for acceptable yield is acceptable backgrounds is still ongoing. And from the available data, beta-carotene levels vary greatly in GR2-E rice, depending on environment and background, so all tests should be done on the final lines put up for approval, not on proxy lines.

    Also, regarding efficacy and safety of GR2, there is an important paper showing the degradation of beta-carotene in Kaybonnet Vitamin A Rice GR2-E. It shows that 1 day after harvest beta-carotene levels are 31.1 ug/g. By 36 days (less than 2 months) after harvest the levels have dropped to 13.7 ug/g (less than half the starting amount). By 72 days, the levels are only 4.1 ug/g. This paper raises serious questions about the nutrition of GR2, that were not addressed in the regulatory documents.
    Unfortunately this paper is not free and open access but here is a link to the table with the data cited: Schaub, Patrick, et al. “Nonenzymatic β-Carotene Degradation in Provitamin A-Biofortified Crop Plants.” Journal of agricultural and food chemistry 65.31 (2017): 6588-6598.

  • Allison, another fake approval for you, this time from the FDA:

    “Goodbye to Golden Rice” indeed!


      Ed; FDA “approval” (as in FDA merely accepts IRRI’s assertion of safety) yesterday (see above link) of Golden Rice contains this rather crucial caveat: “the concentration of Beta-
      carotene in GR2E rice is too low to warrant a nutrient content claim”. Just because IRRI/Syngenta/The Golden Rice crew make the claim of nutritional content or benefit does not mean it is actually true. FDA is acknowledging it is not true in any meaningful sense.

      • Jonathan: Re: “Just because IRRI/Syngenta/The Golden Rice crew make the claim of nutritional content or benefit does not mean it is actually true.”
        They do not merely “make the claim,” they have provided scientific evidence in peer-reviewed journals that the claims they make as to total carotenoid levels in the rice are true. To dismiss their claims in this manner is irresponsible and offensive. It basically calls them frauds or incompetents without citing any evidence for the assertion. In science journalism, or in science proper, one does not have the option of selectively accepting or rejecting claims based on whether they happen to align with one’s personal biases or preconceived narrative lines that one wants to present. That would undercut the integrity of science journalism and indeed that of science itself.

        Secondly, re: “this rather crucial caveat: “the concentration of Beta-
        carotene in GR2E rice is too low to warrant a nutrient content claim”.

        Here is the response to that caveat provided to me by Adrian Dubock: “the FDA comments are regarding the US population, their average consumption of rice (45g daily), and with reference to the Recommended Daily Allowance of ‎ beta-carotene. The RDA includes enough beta-carotene to maintain 3 months liver stores. Such are not required to combat VAD. RDA is 2.5X standard deviations larger than the Estimated Daily Intake. And nutritionists consider that 30% to 40% of the latter consumed daily is sufficient to prevent the nutritionally acquired immune deficiency syndrome which is VAD, and causes 23 -34% of all under 5 years mortality globally as well as being the leading cause of childhood blindness.”

        This analysis refers to the FDA’s prior summary document mentioned in my reply to Allison dated 5/28/2018, and which can be found at:

        • Ed- For our reply to your statements, please read our new article at: Our new article discusses the low beta-carotene levels found in Golden Rice GR2E. This is the Golden Rice GR2E variety bred for the Philippines, whose compositional data (from 2 field trials carried out in the Philippines) was submitted to FDA and other regulators by IRRI. Our article also references papers that discuss the lack of evidence for Golden Rice effectiveness and safety. These are from researchers who point out that long-term low-dose toxicity feeding trials on whole grains are still missing for Golden Rice, as are trials of its effectiveness in target populations. Furthermore, Testbiotech, in its analysis of the FSANZ approval, points out that basic -OMICs studies have never been carried out to look for unintended effects caused by the genetic modification procedures or the addition of 3 new enzymes to Golden Rice: Then and Bauer-Panskus (2018) Testbiotech on Risk Assessment for Golden Rice by FSANZ February 2018. Available at:
          Testbiotech also points out that toxicity testing should be done on the Golden Rice varieties that would be grown that have the highest possible levels of beta-carotene, as beta-carotene related compounds can be toxic. Also higher levels of production could potentially disrupt other nutritional pathways more severely.
          Approved or not, the most basic scientific tests have not been carried out that could show that Golden Rice is either either safe or effective.

    • HI Ed, also it is interesting to note that your link is to an ACSH article boosting Golden Rice and blaming Greenpeace for its failures. These are standard GMO-industry talking points– and in fact, ACSH is known to be an industry front group — see the fascinating expose by Mother Jones: and also One funder of ACSH has been Syngenta — developer of the GR2 lines, including GR2-R1 and GR2-E.

  • Allison, I do not doubt that what you say is true, but it is an ad hominem argument that does not address the issues I raised, that we now have a third approval of Golden Rice GR2E, all three of are from technologically advanced and scientifically literate countries. And in view of these facts, it is premature to keep waving goodbye to GR. It is still here, like it or not. Jonathan’s comment is at least on point, but it makes me wonder why we can trust the sentence he quotes, since both of you are apparently quite skeptical of every other aspect of these approvals. I will have more to say concerning Jonathan’s comment when I am back in the office Monday and have access to documents. Cheers.

  • Ed, contrary to what the ACSH article you link to says, the FDA did not approve GR2-E. Neither did it assert the safety of GR2-E. If you read the actual FDA letter it explicitly says (and I quote)”IRRI also concludes that genetically engineered GR2E rice does not raise issues requiring premarket review or approval by FDA.” The letter also says, and again I quote “Based on the safety and nutritional assessment IRRI has conducted, it is our understanding the IRRI concludes that human and animal food from GR2E rice is not materially different in composition, safety, or other relevant parameters from rice-derived food currently on the market except for the intended β-carotene change in GR2E rice.” The FDA explicitly did not approve GR2-E or say it, the FDA, found GR2-E to be safe.
    The statement Jonathan quoted is from paragraph three of the FDA letter: “Although The concentration of carotene in GR2E rice is too low to warrant a nutrient content claim, the β-carotene in GR2E rice results in grain that is yellow-golden in color.” From the context, it is not clear whether the statement about low levels of β-carotene in GR2E rice is FDA quoting IRRI or their own determination. If you can find out more information, it would be useful to know what data FDA (or IRRI) used to make that determination. Again the FDA letter is here:

  • Allison: You are right that the letter does not say that GR2E is safe (or unsafe). But the fact the USFDA has considered all the information provided to them by IRRI and has no further questions for IRRI, and its further claim that “Although GR2E rice is not intended for human or animal food uses in the United States, when present, it would be a producer’s or distributer’s responsibility to ensure that labeling of human and animal foods marketed in the United States, meets applicable legal requirements,” this is equivalent to allowing GR2E to be used as human food in the US so long as it is properly labeled and meets other applicable laws. This is as much as to say that the food is safe to eat.
    Further evidence that the FDA regards GR2E as safe is provided the summary document, available at
    the penultimate paragraph of which says: “FDA evaluated IRRI’s submission to determine whether GR2E rice raises any safety or regulatory issues with respect to its uses in human or animal food. Based on the information provided by the company and other information available to the agency, FDA did not identify any safety or regulatory issues under the FD&C Act that would require further evaluation at this time.” Here the FDA explicitly says that it did not identify any safety issues. What this means is self-explanatory.

Leave a comment