Comments on the EFSA opinion to allow cultivation of BASF's starch potato Amflora

[img_assist|nid=108|title=|desc=|link=none|align=right|width=100|height=43]BASF applied for approval for the cultivation of the GM starch potato Amflora, as well as for its use as food and feed. Even though the application does not include any sufficient information to evaluate its environmental and food/feed safety, the EFSA gave a positive opinion.

Approval for BASF's starch potato Amflora (EH92-527-1) is sought under Directive 2001/18 for cultivation only, but an additional application has been filed under Regulation 1829/2003 for approval as food and feed.1

Potatoes produce two different types of starch. Amflora has been genetically modified in such a way that a gene is inhibited and therefore the chemical pathway to produce the starch Amylose is blocked. As a result the BASF potato starch is suppose to consist to 98% of the desired starch Amylopectin instead of 70-80% in a normal potato. For industrial production this has the advantage that the two starch types don't need to be separated.

In contrast to the usual GM crops with herbicide resistance or Bt toxin production where an additional protein is added, the BASF starch potato has been modified by blocking a normal metabolic pathway in the potato plant. Little experience exists about effects of the inhibiting of a normal plant process, both in the plant itself as well as how a drastically changed composition will impact on other organisms. Neither are there any experiences with GM potatoes, since none are grown world-wide. The only commercialized GM varieties - modified by Monsanto to kill Colorado beetle - have been withdrawn again from the market in the USA because food retailers demanded non-GM potatoes.

Some of the data in this application is more then 10 years old, and in general (considered to be) of poor quality. Annex G of the EFSA opinion already lists 22 pages of concerns of the Member States.

1) Molecular data

The molecular data contains numerous irregularities. Two partial copies of the transgenic DNA fragment, including its flanking regions, were inserted, so that the transgenic inserts consists of two connected copies in reverse orientation.

The most serious concern is that an open reading frame (ORF 4) is transcribed to the RNA level. The possible ORF4 protein shares a high degree of similarity to two different bacterial proteins. According to the Austrian Ministry of Health and Women the first 50 amino acids of this possible novel polypeptide is homologous to the bleomycin resistance protein. Bleomycin is used as chemotherapeutic for cancer treatment.

The transcription of such an ORF is the first of two steps to produce an new protein. However, the EFSA just states: “Extensive studies indicate that, although ORF4 transcript is detectable in the GM potato, there is no corresponding translation into a protein, confirming expectation from the molecular characterisation of ORF4.”

This is too simplified. Secondary metabolic pathways in plants are complex. The fact that no protein could be detected in the available studies, is not sufficient to guarantee that the RNA for this novel protein is not translated in other situations.

The data provided by the notifier acknowledges that the deletions and rearrangements during the genetic engineering may cause changes. Even the applicant seems to be unconvinced about the stability of the inserts and proposed to the EFSA to "monitor the stability of the inserts and the phenotypic expression during cultivation of the potato."

2) Compositional analysis

The are several significant differences noted in the compositional chemistry, in addition to the intended modification to starch. These include: vitamin C, saccharose, glucose and fructose, nitrates, digestible fibres, chlorogenic acid.

In contrast to other GM crops that are modified to produce an additional protein to turn them herbicide tolerant or insect resistant, the GM potato has an intentionally modified metabolic pathway in which the production of one starch is inhibited. This not only leads to the intended reduction of the starch amylose but also can lead to a accumulation of its precursors. This is confirmed by the notifier by the observation of increased levels of different sugars (fructose, glucose and sucrose) in the GM potato. Inhibiting one pathway and accumulation of precursors are likely to lead to further metabolic changes.

The compositional analysis reveals several unexpected changes. The EFSA states: “In addition to the intended alternations in starch composition of the GM potato some statistically significant differences between the GM potato and its control were observed each year, including a decrease in yield and dry matter and an increase in sucrose content [...] and vitamin C content [...] Other differences were also noted during single years, but not consistently throughout the three years, such as decreases in glycoalkaloid levels of solanine and chaconine in potato EH92-527-1 during two years.”

Unexpected effects are of outmost importance in potatoes as they have a very complex secondary chemistry, including the production of toxic compounds.

The EFSA, however, disregards such differences as not consistent and only assesses composition levels in view of the nutritional value of the potatoes. Unexplained and unexpected changes in the composition of a plant with modified metabolism, however have to lead to further investigation. Especially contradictory results need further attention, because they can indicate instability and/or so far unrecognised effects. The potatoes cannot be considered “substantially equivalent”.

3) Toxicity

In the original application, the use of GM potato pulp as animal feed was stated, but no feeding trials or toxicity tests were conducted at all. Since then the application for a feed approval was filed separately under Regulation 1829/2003. Two simple feeding studies have been conducted, but they are insufficient for an safety assessment. In addition, where one study showed significant differences between modified and unmodified potatoes, these were not followed up but simply disregarded. None of the two studies were conducted with the actual fresh potatoes.

In a 90-day feeding study, rats were fed with diets containing 5% freeze-dried potatoes. This is low compared to other studies that feed concentrations up to 30%. Significant differences were recorded in female rats for white blood cells and spleen weight. In male rats, the number of cysts in thyroids were increase. None of these findings were followed up. Even as the potatoes are freeze-dried, this is the only feeding study with the starch potato itself, even though next to no experience exists about GM crops with changed composition.

In the second study2 2x16 cows were fed on GM pulp for periods of 8 weeks to measure weight gain as an indicator for the nutritional value. The study was not set up to assess any toxicological parameters. For example blood or urine were not not studied, nor where any physiological factors recorded to assess animal health. This study has been criticised in detail by the Finnish Plant Production Inspection Centre. They conclude that:

"The animal feeding trial does not answer, unambitiously, the question on the potential impact of feed GM-starch by-products on feed intake, digestibility or weight gains. Nor is it possible to draw any conclusions about the effects of the product on milk production. In the rumen the microbes metabolise the side products to volatile fatty acids. Whilst one might not expect to see any effects on the activity of rumen microbes caused by the different potato pulps this question has not been addressed."

The starch potato has been modified with the goal of a change metabolism and a changed composition and besides the intended change other changes in composition have been recorded, but the application only provides two animals studies, only one of which actually uses the GM crop (and even then only freeze dried) and the second uses a production by-product. One of the study only measures weight gain in a study set up with fundamental flaws that could mask possible effects. In the other study significant differences in physiological parameters were found and not followed up. There is no digestion study, even though the complex digestive system of cows who are supposed to feed on the GM potato are known. Also, the information provided is not sufficient to exclude allergic risks.

Nevertheless the EFSA disregards questions about animal and human health and just looks "from a nutritional point of view" and declares that there were "no outstanding safety issues." In addition, the EFSA does not consider any post-market monitor of the GM potato as food or feed necessary.

4) Antibiotic resistance

The GM starch potato contains the antibiotic marker nptII as selection marker. It provides resistance not only for kanamycin, but also for neomycin, paromomycin, ribostamycin, butirosin, gentamicin B and geneticin (G418).

According to Directive 2001/18/EC Article 4, Member States and the Commission shall ensure that antibiotic markers shall be phased out for GMOs regulated under Part C by 31 December 2004. Nevertheless, the EFSA gave a positive opinion about the use of this antibiotic resistance marker in the GM potato. In contrast to EFSA's opinion that kanamycin resistance was already widely spread in bacteria anyway, Other national and international authorities come to different conclusions about the importance of kanamycin.

Kanamycin is listed in the WHO Essential Medicines Library as a drug against multi-drug resistant tuberculosis.3 This kind of tuberculosis is a growing problem worldwide and the potential need to use kanamycin should therefore be taken seriously.

The EFSA however simply states that “any additional contribution from potential transfer to soil organism is considered to be insignificant.”

5) Environmental effects

The list of insects, bacteria and fungi that interact with potatoes in Europe is long. However, in the original application there was no environmental risk assessment of the GM potatoes; no information was given on the interactions of the GM potato with the biotic and abiotic environment. Instead it is simply stated that “a detailed investigation of effects on the micro-flora before placing on the market or in a monitoring program would not be proportional to the possible risk from the putative detection of a minor difference in micro-flora composition.” However, any change in microbial ecology could affect soil fertility and be of the utmost importance.

The EFSA requested further information on the impact of the GM potato on plant-associated organisms (e.g. invertebrates). According to the EFSA opinions the applicant then provided data from field studies suggesting that there is no greater susceptibility nor greater resistance to pests and diseases, nor is there a change in sensitivity to a number of potato associated viruses.

"5.2.4 Interactions with the GM plant with non-target organisms
From the field studies carried out in Sweden, Germany and The Netherlands, the applicant provided data on the impact of the modified crops on plant-associated organisms. The result of field studies suggest neither greater susceptibility nor greater resistance to pests (e.g. aphids, leafhoppers, potato cyst nematodes (sp Globodera)) and diseases (e.g. late blight (Phytophthora infestans), potato early blight (Alternia solani), Erwinia rots) than non-GM potato. There was no evidence of changes in sensitivity to the plant associated viruses PVY, PLRV, PMTV, and TRV. In view of this and of the equivalent composition of the GM potato plant, it is considered that no adverse effects on plant-associated organisms would be expected from cultivation of the potato EH92-527-1."

These studies were obviously performed to study the performance of the new GM potatoes. They only recorded the susceptibility of the GM potato to pests and diseases, but they give no information about the effects of the potato on these organisms, and they don't give any information at all about possible effects of the GM potato on other organisms that are not considered pest, like beneficiary insects, earthworms or other wild animals. There are no studies about possible impacts on the biodiversity on or around the field.

The starch has by intent a different starch composition then the non-modified potato, and it has higher sugar levels - possible a result from the changed metabolism of the plant. There are also indications that other compositions (e.g. vitamin C levels) are different. These changes could make the GM plants more or less attractive to animals as food source.

This different starch composition could also result in a different decomposition. EFSA states in its opinions that "GM tubers have a different starch composition, and therefore may be decomposed by a changed microbial community [...]" These effects are not studied at all, but need to be studied as part of an e.r.a.

EFSA acknowledges, that "some different interactions with plant associated organisms may occur in Southern regions" but conclude that "there is no indication that plant-associated organisms would be adversely affected." Such conclusion cannot be drawn, on the basis of so little information.

There also is no environmental risk assessment of the intended use of GM potato juice (a by-product from starch production) as fertilizer.

Edited version of comments submitted to EFSA's 'Open consultation on Starch potato EH92-527-1', December 2006.