Golden Rice: Of Good Intentions, Insertional Mutants, Human Error, and the Need for Better Regulation of GMOs

After learning that the genetically engineered (GE) Golden Rice “event” being bred into varieties of rice favored by Indian farmers turned out to have a mutation—a mutation created as a result of the genetic engineering process itself (you can check out my first post on this subject here)—I was curious to learn more about the history of this genetic engineering project undertaken for humanitarian purposes.

The Golden Rice project is an attempt to use genetic engineering to alleviate vitamin A deficiency, a major cause of blindness, illness and death—mostly among young children—in developing countries where diets can consist primarily of rice and sometimes little else in the way of vegetables or other foods that contain beta-carotene, a compound that human beings can then convert into vitamin A. The project was initiated in the early 1990s and was widely publicized in 2000 when a photo of Ingo Potrykus, one of its “founding” scientists (along with Peter Beyer), appeared on the cover of Time magazine.

But it’s been 18 years since then and Golden Rice is still not available to the world’s poor who might benefit from it. Why is it taking so long to deliver this humanitarian product?

A very revealing paper by a member of the Golden Rice Humanitarian Board, Adrian Dubock, provides some answers. In it, Dubock asks “What can Golden Rice’s development history and trajectory teach us?”

SPOILER ALERT! My answer to that question—which is essentially diametrically opposed to his—is this: Check early in your GE project to identify any unintended insertional mutations in your prototype crops that were caused by the genetic engineering process, and then throw any GE “events” that have such a mutation out of your product development pipeline! (Dubock’s answer, on the other hand, might be summed up as: international regulation of GE crops, in the form of the Cartagena Protocol, impeded our progress every step of the way; more on whether regulation of GE foods is currently adequate at the end of this post.)

Are Other Developers of GE Crops Neglecting to Discard the Unintentional Mutants Created by the Genetic Engineering Process?

Golden Rice’s history also begs the questions: how many other developers of GE crops have neglected to screen their GE prototypes for insertional mutants? Are there insertional mutants among the GE crops currently on the market? (I’ve tried, unsuccessfully, to determine whether Monsanto’s NK603 GE corn was screened for this potential problem, for example.)

But I’m getting ahead of myself….

Timeline of the Development of Golden Rice

What follows is a timeline of the development of Golden Rice focused on events relevant to the problem of the potential for unintentionally creating GE mutants, a problem still faced by all developers of GE crops using the Agrobacterium-based and biolistics methods. This timeline is based on the information contained in “The present status of Golden Rice” by Adrian Dubock. (The information I mention from Dubock’s article can be found on the pages of it indicated in parentheses below; emphases are mine.)

2000: Ye et al. publish a paper describing the first “prototype” Golden Rice; the inventors of Golden Rice trade “commercial rights in the technology to Zeneca [soon to be Syngenta], in return for the companies [sic] support for the inventor’s humanitarian vision;” Zeneca is granted exclusive rights to Golden Rice technology “free of charge for humanitarian applications, but royalty bearing for commercial applications” (p.71).

The “Golden Rice Humanitarian Board is created “to advise the inventors” during its development and one of its “first agenda items was to hear the advice of a Zeneca biotechnology regulatory specialist about the molecular characteristics required of a genetically transformed GMO-crop to ensure that it would be able to be registered for use under the regulations to be derived from the very recently (2000) published, but not yet in force (2003) Cartagena Protocol on Biosafety” (p.73); the US biotech industry pays for television advertising “implying fields of golden rice were [already] growing in US fields…” and Gordon Conway, President of the Rockefeller Foundation, joins “anti-GMO activist Vandana Shiva in agreeing that ‘the public relations uses of golden rice have gone too far’” (pp.72-3).

2001: A license agreement is signed by inventor Potrykus and Dr. Ron Cantrell, director of the International Rice Research Institute (IRRI) in the Philippines, and samples of the Golden Rice prototype are hand-delivered to IRRI two days later (p.73).

2002: The Humanitarian Board asks the “network of public sector rice research institutions” to create more than 1000 additional transformation events “from which it was hoped an improved transformation event could be selected collaboratively, to be taken forward as the one lead event by all Golden Rice licensees” (p.73).

2003: New GE “events” of Golden Rice [GR1], the best of which accumulates 13 micrograms (μg) total carotenoids per gram (g) of rice, as opposed to the 1.6 µg/g in the prototype, were created and plans to conduct field and human feeding trials with some of them are made; an additional version of Golden Rice [GR2], with even higher levels of beta-carotene, looks promising in Syngenta’s labs (pp.73-4).

2004: Field trials of GR1 Golden Rice are conducted in the US; Syngenta donates GR1 and 13 GR2 “events,” at least the latter “of which were considered by [Syngenta Biotechnology, Inc.] scientists and regulatory specialists to be ready for and capable of complying with regulatory studies and standards” (p.76), to the Golden Rice Humanitarian Board and announces that it has “no commercial interest in the Golden Rice project” (p.75); the Humanitarian Board decides to concentrate efforts on 6 of the 13 GR2 individual transformation “events” donated by Syngenta (p.76).

2005: Paine et al. publish a paper describing the development of the GR2 events.

2006: IRRI requests, and Syngenta provides, molecular data on the 13 GR2 events (p.76).

2008: The field phase of a feeding trial with Chinese children is conducted; IRRI plants the first confined field trial of events of GR1 and GR2 Golden Rice in Los Banos, Philippines (p.77).

2009: A meeting of the Humanitarian Board is held in March to choose the top 2 Golden Rice events for product development; results of the feeding trial with human adults in the US are accepted for publication and presented at the meeting (p.77); agronomic data on 3 GR1 and 6 GR2 events, introgressed into 4 different indica rice varieties, were considered and Dr. S.R. Rao (Dept of Biotechnology, Government of India) asked “if there was any molecular data available to support the decision making. No such data was forthcoming (although IRRI had received it in 2006, it appeared to have been forgotten)” (p.78); nevertheless, based on the data the Board did have at hand, event GR2G was selected as the “Lead Transformation Event, with event GR2R as a back-up event if needed” (p.78); by December 1st scientists at IRRI had examined the molecular data on the GR2 events (provided by Syngenta back in 2006) and realized that the DNA inserted into GR2G, the lead event and the one used in the human bioconversion trials, contained an unexpected, unintended deletion of ~400bp; the newly found/examined molecular data also indicated that in both GR2G and GR2R the genetic engineering process had created mutations at the site in the rice genome where the foreign DNA had been inserted; but, “[d]espite the summary slide provided by IRRI, The Board had not reviewed, nor did most have the training, to ‘review all sequence data’ in any meaningful way, and it is unclear which other individuals had or the level of scrutiny afforded to it,” the decision was nevertheless made to drop the GR2G event and “bring forward” the GR2R event as the lead event (p.79).

2010: The grant to Peter Beyer from the Bill and Melinda Gates Foundation terminates; “Instead the Foundation intended to award a grant for development of Golden Rice itself to IRRI, for management of Golden Rice out of IRRI” (p.80).

2011: By October 3, GR2R seed had only been supplied to research institutes in two countries, India and the Philippines; the “inventors and the public sector Golden Rice licensees in other countries are very frustrated by this slow progress” (p.80).

2012: “Dr. Tang’s research with Chinese children, initially spoken of in 2003/2004 showed that” a bowl of cooked Golden Rice [GR2G?] “can provide ~60% of the Chinese Recommended Nutrient Intake of vitamin A for 6-8-y-old children;” “Greenpeace issued a press release condemning use of a GMO-crop, Golden Rice, with Chinese children as ‘guinea pigs of American researchers;’” [Tang et al. (2012) was later retracted (and China sacked officials in that country who were associated with the experiment) due to failure of the researchers to follow Tufts University rules regarding securing consent from research participants]; 5 multi-location field trials with GR2R are planted in the Philippines (p.80).

2013: In August one of the field trials was destroyed by anti-GMO demonstrators; data from the other 4 field trials indicated a “yield-drag” in Golden Rice GR2R compared to wild-type rice; in December, the “IRRI Network Coordinator appeared to recall the issue summarized in his December 2009 slide” indicating that the new lead event, GR2R, was an insertional mutant (pp. 80-1).

2014: IRRI posts information on its website about the poor yields with GR2R [but fails to mention the fact that it contains a mutation caused by the genetic engineering process] and that “a decision has been reached to move forward from work solely focused on GR2-R to also include other versions of Golden Rice, such as GR2-E and others” (p.81); Adrian Dubock publishes “The Present status of Golden Rice” in the Journal of Huazhong Agricultural University.

What can this history of Golden Rice teach us?

This is an astounding story. A group of humans with all the best of intentions, who apparently tried to take into account–from the very start of their work together–the regulatory requirements in the various countries for creating, transporting, and marketing a GMO food, and yet years of time and loads of effort and who knows how much money (the Gates Foundation?) were wasted on Golden Rice events that should have been culled from the “production pipeline” by 2005 (or 2006 at the latest). Sad!

Errors were made by multiple humans involved in this story. Why did Syngenta hand over mutant Golden Rice events to the Humanitarian Board in the first place? Why didn’t scientists at IRRI thoroughly examine the molecular data on those events right after they had specifically requested the information from Syngenta? Why show molecular data indicating that 2 of the 3 top Golden Rice events have their foreign DNA inserted into rice genes to a decision-making body incapable of understanding the ramifications of those mutations (that were caused by the genetic engineering process)?

We don’t know whether the insertional mutations in Golden Rice events GR2G and GR2R made the rice grains harvested from those plants unsafe for human consumption. But most mutations are deleterious, and determining whether a particular mutation would cause a problem(s) in rice that would make it unsafe for consumption would require further investigation (of each GE mutant food variety). The most expedient way to deal with insertional mutants caused by the genetic engineering process is to identify them early and eliminate them from commercial development.

Glenn et al., mostly employees of Monsanto, indicate that they now do cull GE insertional mutants in this way. But since they also indicate that they now eliminate GE events with other imprecise manifestations of the genetic engineering process from their commercial pipeline, when Monsanto certainly has not always done that, could there be any GE mutants already in our food supply? And do other developers of GE food crops follow the procedures now used by Glenn et al.?

My take on the Golden Rice story is that we need better, truly science-based regulation of GE foods, regulation that recognizes the imperfections in the technology and regulates its putative products accordingly. Regulators in every country around the world should deny “approval” of any GE food crop that contains a mutation caused by the genetic engineering process. They should do so because genetic engineering is imperfect, and humans–even those with the best of intentions–will continue to make errors, and combining this imperfect technology with error-prone humans could have negative effects on our food supply.



This entry was posted in Biotechnology and tagged , , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

9 Responses to Golden Rice: Of Good Intentions, Insertional Mutants, Human Error, and the Need for Better Regulation of GMOs

  1. Ed says:

    In a piece published in 2010 on the lessons learned from his work on Golden Rice, Ingo Potrykus argued that it was precisely over regulation that caused a 10-year delay in the product’s development. The Abstract says:

    Compared to a non-Genetically Engineered (GE) variety, the deployment of Golden Rice has suffered from a delay of at least ten years. The cause of this delay is exclusively GE-regulation. Considering the potential impact of Golden Rice on the reduction in vitamin A-malnutrition, this delay is responsible for an unjustifiable loss of millions of lives, mostly children and women. GE-regulation is also responsible for the fact that no public institution can deliver a public good GE-product and that thus we have a de facto monopoly in favour of a few potent industries. Considering the forgone benefits from prevented public good GE-products, GE-regulation is responsible for hundreds of millions of lives, all of them, of course, in developing countries. As there is no scientific justification for present GE-regulation, and as it has, so far, not prevented any harm, our society has the urgent responsibility to reconsider present regulation, which is based on an extreme interpretation of the precautionary principle, and change it to science-based regulation on the basis of traits instead of technology. GE-technology has an unprecedented safety record and is far more precise and predictable than any other ‘traditional’ and unregulated breeding technology. Not to change GE-regulation to a scientific basis is considered by the author ‘a crime against humanity’.

    The paper may be found at:

    Ed Regis

    • Belinda says:

      Well, first of all, Dr. Potrykus’ estimate that “Golden Rice should have reached the farmers’ fields in Asia by 2002” (stated in the paper you cite) is ridiculous. Developers were still working (in labs) on ways to deliver higher levels of beta-carotene (pro-vitamin A) in rice endosperm that year, and the Humanitarian Board was asking researchers to create thousands of new genetically engineered rice “events” from which they hoped to select the one event that would be eventually bred into farmer-preferred rice varieties; i.e. they didn’t even have GE rice plants to choose a top candidate from yet in 2002.

      Second, Dr. Potrykus states repeatedly in his paper that there was “no experience in the public domain with deregulation of a GMO-product” but that is not true; the GE papaya that saved the Hawaiian papaya industry was developed in the public sector.

      Third, while the AMA, WHO and others have asked developers of GE foods to avoid having genes that confer antibiotic resistance in their products, it is still (to my knowledge) not a requirement of regulatory authorities. In fact, many commercialized GE food crops contain a gene that confers resistance to kanamycin. And if the Humanitarian Board decided to remove the selectable marker gene from Golden Rice as a means to increasing public acceptance of it (or for any other non-regulatory reason)…blame is not with the regulatory agencies.

      Similarly, screening for “regulatory clean events” has not been a requirement of regulatory agencies in many countries (although I believe it always should have been); in fact, there are multiple commercialized GE crops out there that contain various “alterations of the [foreign gene] construct.” One of the most egregious examples I know of is NK603, the GE corn fed to the rats in Seralini’s study (which was so viciously attacked and subsequently retracted for merely being “inconclusive”). The developer of that GE food crop, Monsanto, identified not only an extra piece of DNA that had been inserted into that product unintentionally, but also two point mutations in one of the genes conferring resistance to glyphosate; Monsanto also found that RNA transcribed from the DNA inserted into NK603 contains not just the information contained in the inserted DNA but, unintentionally, some from the adjacent corn DNA as well. (See my post from 2/9/18 for more details.) And NK603, this GE crop/food that contains these various, multiple “alterations of the gene construct,” is currently in the same queue waiting for regulatory approval in the Philippines as Golden Rice, another indication that Monsanto/Bayer (at least) doesn’t believe that “regulatory clean events” are necessary for gaining regulatory approval.

      So it’s not the case that regulatory agencies require a “clean event;” and, again, if the Golden Rice Humanitarian Board decided that a “clean event” was necessary (for public acceptance or whatever), the regulatory requirements are not to blame.

      The main problem as I see it is that the “naïve public sector scientists” Dr. Potrykus mentions in his paper, who followed their “idealistic concept,” didn’t fully realize how inefficient and imprecise the current methods for genetically engineering crop plants actually are.

      Worse yet is that, despite Dr. Potrykus’ claim made in 2009 (in the paper you cite), that “all final Golden Rice varieties are based on such ‘regulatory clean’ events” (as I believe they should be), neither of the top two events of Golden Rice chosen for their product development pipeline in 2009 were such clean events; instead, they were both mutants created by the genetic engineering process itself, even though the developers of Golden Rice had possessed data indicating that they were both mutants since 2006. Clearly, not enough care was used in selecting those top GE Golden Rice events.

      In my opinion, more time in 2009 (and before) should have been spent carefully examining the molecular data on Golden Rice events, and less time spent writing papers disparaging the regulatory process for bringing a GE food crop to market. Regulation of GE crops may well have “prevented any harm” in the case of StarLinkTM corn, for example, since U.S. regulatory agencies were so concerned about that GE crop that they monitored levels of it in the U.S. corn crop for 7 years AFTER that product had been taken off the market. And, if peer-reviewed studies of other GE crops that indicate possible evidence of harm weren’t attacked by plant molecular biologists, but followed up on with additional scientific studies instead, we might now have an even more “science-based” process(es) for regulating new GE food crops.

      In sum, lack of taking appropriate care in utilizing what is currently an inefficient, expensive and imprecise (in its own ways) genetic engineering technology to develop food products for humans is not a reason to change regulation so as to be based on “traits instead of technology;” in fact, it is reason for strengthening regulation of GE food crops, not weakening it. For regulation of GE crops to be truly “science-based,” it is appropriate for developers to be required to screen for “regulatory clean events.”

  2. Ed REGIS says:

    Belinda says: “But it’s been 18 years since then [Time magazine Golden Rice cover story, 2000] and Golden Rice is still not available to the world’s poor who might benefit from it. Why is it taking so long to deliver this humanitarian product?”

    Part of the answer is: Because of the regulatory hurdles facing the development of any GM plant variety.

    In 2011, a private British consulting firm, Phillips McDougal, was hired to produce a study of the cost and time involved in the discovery, development, and authorization of new plant varieties created through biotechnology. The project was carried out by collecting data from six of the world’s major crop biotechnology companies, including BASF Corporation, Bayer CropScience, Dow AgroScience, and Syngenta.

    The total amount of time required for a developer to move from initial trait discovery to final authorization and release into the marketplace was an average of 13.1 years. “However, there was considerable variation in the responses between companies and between crop species,” according to the study report. “Overall variability in the responses received varied from a low value of seven years to the comparatively high value of 24 years.”

    Further, according to the study report: “Regulatory science, registration and regulatory affairs is the longest single phase in product development and is estimated to account for 25.8% and 36.7% of total cost and time involved respectively.”

    Given these figures, the 18 years between the invention of the Golden Rice prototype in 1999 and the first government approval, issued by Food Standards Australia New Zealand (FSANZ) in December 2017, is well within the normal range found by the study.

    • Belinda says:

      Have you informed Ingo Potrykus and Adrian Dubock of this study?

      • Ed REGIS says:

        No, because it only reinforces what they have maintained for years, namely that regulatory matters constitute a large share of the time and expense burden of developing any GM food. Nothing new in that for them.

      • Belinda says:

        The bottom line is that the development of Golden Rice would have been both faster and cheaper if its developers had utilized the molecular data at hand immediately and thrown all the putative “product” events that were mutants caused by the genetic engineering process away.

  3. Mary Darragh says:

    Great article. You clearly presented a complicated issue that was easy for this lay person to understand.

  4. JAN says:

    Really excellent article – very helpful explanation of the various roadblocks that have slowed up the deployment of GR. I wasn’t aware of the 2014 Dubock paper, good cite. Your point about weeding out problematic intentional mutants is well-made, particularly relevant for diploid species like rice,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s