Prof. Seralini's answer to ECII Report submitted to GEAC on Bt brinjal – January 2010

GM aubergine in India and conflictual scientific hypotheses for its evaluation
- An answer to GEAC report -
January 2010
by Prof. G.E. Seralini, University of Caen, Risk Pole CNRS, Institute of Biology, CRIIGEN and ENSSER.

Main points and references

This comment was published in Nature News, November 2009, after the paper "Bt brinjal put on ice".

Few aubergines varieties have been genetically modified to synthesize a mutated insecticide toxin (16-17 mg/kg) from a chimeric Cry1Ab-Cry1Ac transgenic sequence. There are also two antibiotic resistance marker genes. The process of authorization is debated in India, including in the Supreme Court. In order to study possible health risks, due to a new metabolism, or to the insecticide, it is necessary to evaluate the longest toxicological tests performed with blood analyses in mammals [1].

These are subchronic tests 90 day-long with goats, rabbits, and rats. A first debate is about the possible occurrence of unintended effects on a long term not visible in 90 days. The second debate is about the interpretation of statistical significant differences in the blood and physiological parameters within 90 days between animals eating the GM aubergine and those fed with its closest isogenic variety, used as control. Other various reference groups are added not demonstrated to eat aubergine-based diets of the same composition in the experiments. Among other signs, for instance in goats eating GM, bilirubin and prothrombin time as well as alkaline phosphatase are modified. The two first parameters were also changed in rabbits but also albumin, lactose dehydrogenase, alanine and aspartate aminotransferases... The liver weights were reached among other criteria in GM fed rats.

Similar cases were discussed for three GM maize producing other insecticides or that tolerate the main herbicide Roundup [2-5]. These clear significant differences in the raw data could be a reflect of natural variations for Mahyco company patenting the aubergine [1,5]. On the other hand another interpretation is that it cannot be excluded that these are first signs of hepatorenal toxicities varying according to the sex or nonlinearly to the dose of GM [6]. The only scientific issue is to repeat and prolong these toxicological tests before feeding with GM aubergine billions of people and animals.

• http://www.envfor.nic.in/divisions/csurv/geac/bt_brinjal.html Raw Mahyco's data (November 2008).
• Séralini GE, 5. Seralini GE, Cellier D, Spiroux de Vendomois J. New analysis of a rat feeding study with a genetically modified maize reveals signs of hepatorenal toxicity. Arch Environ Contam Toxicol. 2007; 52: 596-602.
• Doull J, Gaylor D, Greim HA, et al. Report of an expert panel on the reanalysis by Séralini et al. (2007) of a 90-day study conducted by Monsanto in support of the safety of a genetically modified corn variety (MON 863). Food Chem Toxicol. 2007; 45:2073-2085.
• Séralini GE, Spiroux de Vendômois J, Cellier D, et al. How subchronic and chronic health effects can be neglected for GMOs, pesticides or chemicals. Int J Biol Sci. 2009; 5: 438-43.
• Spiroux de Vendômois J, Roullier F, Cellier D, Séralini GE. A comparison of the effects of three GM corn varieties on mammalian health. Int J Biol Sci. 2009; 5: 706-721
• Maharashtra Hybrid Seeds Company Ltd.(Mahyco), Mumbai et al. Report of EC-II on Bt brinjal. Submitted to GEAC, october 2009.
• Séralini GE. Report for the Supreme Court (January 2009). http://www.criigen.org/brinjal

Conclusion of the summary:

The risk on human and mammalian health is too high for authorities to take the decision to commercialize this GM brinjal.

***

Developments

This is a detailed answer to the Mahyco EC-II report submitted to GEAC in October 2009.

Background

Mahyco is explaining in its original file for Bt brinjal deposited to the Indian government that Bt brinjal was developed to produce a new insecticide toxin (16-17 mg/kg). This will not avoid all insecticides spread on the plant since this genetic engineering technique is supposed to produce a specific insecticide, toxic for some species of insects only. Thus, any claimed advantage of this should compare scientifically the toxicity for environment and health of this toxin in Bt brinjal in comparison to the toxicity of the insecticides really avoided (with scientific observations, and not claimed to be avoided) in similar length studies. For instance, the toxicity of most insecticides is known with the help of six month to two years long tests requested by the international regulation processes, which is unfortunately maximally three months for GMOs and associated insecticides including this Bt brinjal. In addition, since the technology could add additional metabolic changes, not visible by the study of the Bt brinjal chemical composition which is not enough detailed, the whole Bt brinjal should only be used in toxicological valid studies in comparison to its closest isogenic control in order to study the effect of the GM transformation and not irrelevant parameters.

Mahyco explains in Toxicity and Allergenicity studies vol. 1 p.32 on 142 that the toxin produced is a chimeric one containing Cry1Ab and Cry1Ac sequences, plus a change on Cry1Ac subsequence. Moreover, [neither] this toxin nor its cDNA have been fully sequenced after insertion in Bt brinjal.

CONSEQUENTLY, THIS TOXIN IS NOT EXACTLY CRY1Ac AND SHOULD BE REASSESSED PROPERLY, THIS IS CRUCIAL.

The first point submitted to GEAC in the EC-II report is thus wrong (it is not exactly the same Cry1Ac gene!).

The other marker genes described in the Background section, such as the antibiotic resistance gene, are unnecessary and should be removed instead of discussing risks or lack of risks without scientific proofs.

A decision of a committee is not, in any case, a proof of innocuity. The longest toxicological tests in each species should be instead analysed in details. This was performed and clearly reveals numerous side effects showed in the raw data from Mahyco, and not their abstracts. These have been detailed in my first report (January 2009) in comparison to the animals receiving in their diets the best isogenic controls. Other comparisons are irrelevant to study only and precisely the effect of the genetic transformation.

Moreover, any release in the environment will result in an inevitable dissemination that can also be called a genetic pollution, as shown in Europe. This will force the establishment of thresholds of contamination for labelling. No method is described for assays of Bt brinjal in Indian environment and labelling. The mixture of seeds in the seed containers, machines and factories will be superior to any cross-pollination and thus can pollute any kind of food or feed. The isolation distance of 300m will be useless in this regard. In United States, a sterile GM maize containing a vaccine for pigs has contaminated 500.000 tons of soya because some was remaining in the silo.

Details and comments

On pages 7/8. It is surprising that Mahyco committee details the insecticides normally sprayed on Bt brinjal, thus indicating its concern on pesticide toxicity and environmental impacts, without studying on a longer term than 90 days on mammals the new pesticide synthesized at high levels by the Bt brinjal.

One could also compare this spray of pesticides with its economy in organic agriculture.

On page 12. It is also indicated that the acute oral toxicity studies began for this file in 2003, one year after pollen flow studies! Is this a responsible behaviour from a big company constructing a toxin-producing plant? Moreover, these are really short term studies on a few days, the 3 monthlong studies (still really short) on rats only came in 2005, 3 years after field releases with inevitable contaminations! And then, nothing longer for toxicity studies!! We will have in 2010, 10 years after conception, a big hole in health assessment non compensated by any serious study more than 3 month-long with very large field releases! All the rest of the file is somewhat theoretical on health and environmental assessment on humans and mammals. Such as the genetic modification characterization if it is not obtained after re-sequencing in many Bt brinjal varieties to study the stability and real nature of the construct, together with the surrounding sequences that were interrupted.

On page 29. In consequence, it is obvious that there is a profound disagreement with Mahyco's conclusion stating that no other safety studies are required. Chronic toxicity studies 2 years-long on rat (studying the risks during the whole lifespan) are thus necessary before giving these kinds of possible toxic aubergines to mammals and humans during their entire life.

On page 33. There is the proof that to demonstrate genetic stability Mahyco did not sequence the insert and flanking regions in multiple Bt brinjal varieties, this is not in the list of the experiments performed. Thus these experiments are highly insufficient, for instance to see if different toxins are in fact produced by the GM plants.

On page 36. It should be underlined that there is no need of gene transfer to other plants or other organisms that could arise, to create a genetic contamination by mixing of seeds and harvests, in real life.

On page 37. The presumed specificity of modified Cry1Ac toxin produced by Bt brinjal towards lepidopteran insects and not harmful to any other non target species including mammals and humans is a legend, not documented by any serious scientific experimental reference. The ecotoxicological experiments summarized are pretty short in regard to possible impacts on
biodiversity and not published nor peer reviewed.

On page 40. The soil is considered today as a blackbox concerning microflora in particular. The limited number of studies listed does not allow to conclude that there is no impact on soil with Bt brinjal. For instance, it is underlined that no Cry1Ac Bt toxin is found in the soil. This is almost impossible since roots exsudate this kind of toxin. Thus the specificity and sensitivity of the assays performed, not indicated, should be tripled checked before any conclusion. In particular, it is possible that the method performed only measures natural Bt toxin and not the artificially modified chimeric Cry1Ab-Cry1Ac toxin produced at high levels by the GM brinjal.

On page 43, toxicity and allergenicity, or digestive studies. These studies were not conducted with the real chimeric modified toxin produced and extracted from the plant, and are thus invalid.

On page 44, nptII studies. This antibiotic resistance marker is useless for food and feed and should be avoided in order to remove these limited and incomplete studies of risks associated to this gene and corresponding protein.

On page 45. The 90 day studies in rats do reveal significant findings that are very surprisingly not discussed (see my Report on January 2009)!!

On page 46. The compositional analysis do reveal some differences (see also my previous Report) and are insufficient to see pesticide residues for instance, and other toxic compounds.

On page 47-48. Studies on animals did reveal significant findings in the first Mahyco file that are not even discussed here. This is a very important problem for Mahyco's credibility and honesty. See previous Reports.

On page 55. EC II admits that the insecticide toxin issued from the inserted chimeric Bt gene is different, even a little, from the toxin tested in some of their studies. This gives power to our argument. Every serious scientist in the field knows that only one amino-acid difference in a protein can cause a genetic disease, a change in stability or toxicity, like for the prion in Bovine Spongiform Encephalopathy. Thus the corresponding experiments and reasoning of Mahyco are invalid.

Moreover, the proof of re-sequencing the gene after insertion in 3 GM brinjals cultivated from different sources in three different places (because wild mutations are always possible) is not given at all.

The fact that antibiotic resistance marker genes have been already used in other GM plants is not a proof of the necessity neither a proof of lack of toxicity (see references 2, 4, 5 on page 1 of this Report). There is not AT ALL history of safe use with traceability of this kind of antibiotic resistance -useless- marker genes in GM for two decades. This is not true. Where is the epidemiological scientific reference? Numerous people are deceasing all over the world from antibiotic resistance.

On page 58. The limits of quantification and detection of the ELISA test in food and feed are usually dependent on the matrix which is not precised. Moreover, it is given for the Cry1Ac toxin and not for the modified toxin purified and produced from the plant. These limits of assays are thus non valid, implying that there is no proof of lack of presence and / or lack of toxicity of the toxin or its metabolites in food and feed.

This applies for the following studies cited in the EC-II Report.

On the issue 8 on the necessity of long-term studies - The arguments raised by EC-II are not acceptable for the following reasons:

• The international guidelines are still in construction and do not even request 90 day tests on 3 mammals nor traceability; thus the point argued does not mean that it is scientifically serious. A number of products have been forbidden after being authorized because they were found toxic, even if their authorization process has followed to some extent some guidelines.
• The Cry1Ac toxin is not the one produced by Bt brinjal (see above).
• It has not been proved AT ALL safe on mammals on long-term.There are millions of chronic diseases in people from unknown origin that could be attributed to GM toxins because of lack of traceability in food / feed.
• We hope to have a lifespan more than 25 years!!
• The modified Cry1Ac extracted from Bt brinjal has not been studied, nor demonstrated as degraded in digestive fluids.
• Other metabolic changes could arise in Bt brinjal due to the genetic modification itself that could not be assessed with a toxin alone, and this must also be assessed on a chronic point of view with the whole food / feed.

On page 59, issue 9. The significant effects are found by Mahyco, but presented only in raw data (ref 1, p.1 of this file) and not in their abstracts! There are between parameters of animals and their proper controls fed with the closest isogenic line. These differences have been already detailed in my Report (January 2009).

These are statistically significant according to Mahyco raw data.

• The normal physiological range is not a validated well defined concept; it thus should correspond to the proper controls in their experiments.
• The biochemical data are very often disrupted before histopathological changes.
• The transient changes within 90 days could well be signs of a progressive settling of a chronic pathology like a cancer, their transient recovery by organ reactions / transient defenses do not mean that there is no sign of pathology AT ALL !
• Then there is a lie: of course significant changes listed in my first Report are in raw Mahyco's data, the Committee assessing this dossier is urged please to check carefully.

On page 60. The impact on organic farming will be great because of inevitable contaminations if the Bt brinjal is covered by one or several patents, and without careful traceability and insurances in India.

On page 63, point 6. The Bt cotton has not been approved for extensive human and animal consumption. A toxin can be folded and glycosylated or modified differently in a different plant.

On page 64, points 14-16. It is completely wrong: there are significant differences in the raw data in comparison to closest controls, and this gives a holistic view of hepatorenal toxicity that necessitates chronic toxicity studies, by contrast to what is said in point 15. Point 16 is wrong and has already been discussed: there is no history of use with this modified insecticide, different from Cry1Ac, and even Cry1Ac has not been authorized for consumption in mammals. Thus it was never proven as safe (where is the experimental reference on human cells for instance?)

Table answering to my points, p. 69-70... See detailed answers in Annexes of this document. Each time there is a clear significant difference with the closest isogenic control for a parameter in the raw data, Mahyco either concludes that it is "in the range of biological variations », « variations are expected in dynamic biological systems" or "variations were considered as incidental and not related to Bt brinjal treatment". All that means nothing is not scientifically and medically serious at all and hides the truth. There is no point to make such experiments to conclude like that every time! What do they expect in mammals for pesticide-like effects within three months? That the rats explode or develop a burning cancer? Without measuring specific markers like cytochrome P450, steroids or other hormones, it is very hard to be sure within three months, and any sign of toxicity in comparison to controls should be taken very seriously, and not as a joke: "variations in dynamic system"! The tests should be prolonged up to two years for chronic entire life studies for rodents in particular, because the health of billions of people and animals is in the game for their entire life!

The fact that an observation was not seen in the other sex (end p. 72) is a reason to find it not biologically relevant for Mahyco!! This is completely wrong! (Seralini et al., 2009, cited in page 1).

On p. 81. Of course the pollen flow depends on the size of the field; this has been concluded by all environmental studies in Europe.

In fact, all Mahyco arguments for the underlined points are empirical and not acceptable. The author maintains his remarks and a contradictory independent expertise is highly necessary in front of the raw data. The laxity of Mahyco's interpretations will be put in the light.

CONCLUSIONS / RECOMMENDATIONS

Because of the above arguments the EC-II report and the commercialization of the Bt brinjal are unacceptable in their present forms, This Bt brinjal release in the environment includes major risks. It is not serious to give to billions of people and animals for their entire life a food / feed that has not been tested more than 3 months with blood analyses. We do not know the long term consequences of the genetic modification itself nor the effects of the modified insecticide toxin produced at very high levels. Moreover there were clear signs of hepatorenal toxicities, among other effects, shown within 90 days by significant differences in Mahyco's toxicological subchronic tests in mammals: goats, rats and rabbits. These are not clear proofs because the tests are too short, but preoccupying enough to forbid Bt brinjal release at this stage.

More...