New rat feeding study is further proof that GM soy isn’t substantially equivalent to non-GM soy. Report: Claire Robinson
A diet containing GM soybean oil damaged the liver and kidney of rats in a new 90-day feeding study conducted by Iranian scientists. The study provides further proof that GM soy is not substantially equivalent to non-GM soy, meaning that regulatory authorisations given on the assumption of equivalence are invalid.
The study, by Horyie Taheri of Tabriz University of Medical Sciences and colleagues, was conducted on 18 male rats in three different groups (6 rats per group). One group was fed a diet containing 10% GM soybean oil for 90 days, while the other two groups served as control groups, receiving either non-GM soybean oil or a standard lab diet, respectively.
The scientists carried out biochemical analysis of the blood and at the end of the experiment, microscopic tissue analysis (histopathology) of the liver and kidneys.
The scientists found that GM soybean oil caused several histological abnormalities in the liver, including congestion, necrosis, and bile duct hyperplasia (increased cell production, which may indicate a pre-cancerous state).
Similarly, congestion, haemorrhage, and glomerulosclerosis (scarring of small blood vessels) were found in the kidney analysis. Moreover, GM soybean oil significantly increased gamma-glutamyl transferase (a possible sign of liver disease or damage) and insulin (often associated with type 2 diabetes) levels compared to a standard diet.
Furthermore, urea and triglycerides were significantly higher in GM-fed rats compared to rats fed with standard or non-GM diet. These are respectively indications of kidney failure and a failure to break down fats, which can lead to hardening of the arteries and other diseases.
The scientists conclude: “A 90-day treatment with transgenic soy-based oil caused significant organ changes in the liver and kidneys of rats. Further studies are needed to evaluate the long-term effects to better elucidate these impacts.”
Limitations of the study
The study has certain limitations, as follows:
1. As Taheri and colleagues state, 90 days is not long enough to reveal long-term effects – for which a 2-year or longer study would be required. However, the fact that they saw adverse effects of GM soybean oil even over this medium-term period (roughly equivalent to 7-8 years in human terms) is concerning and suggests that over time, more serious disease could develop. They also suggest that measuring inflammatory markers would have been a useful addition to the study, to explain insulin resistance and other findings.
2. The error bars, in this case based on what’s known as the standard error of the mean, on some of the biochemical findings are wide. In simple terms, this indicates uncertainty in how those findings can be interpreted. This is likely due to the relatively small sample size (6 rats per group), which has caused the study to be statistically underpowered. To be fair to the authors, however, studies with as small, or smaller, sample sizes that conclude that there was no adverse effect from feeding the GM diet are promoted as showing GMO safety.
3. The authors didn't analyse the oils for contaminants, which could have affected the results. However, feeding studies with GMOs reporting that the GM diet tested was safe also typically fail to do this.
Regarding points 2 and 3 above, there is no reason to hold a study finding harm from the GMO to different standards than a study finding safety. Moreover, we’re not aware of any other studies testing the safety of extracted GM soy oil. So currently, Taheri et al’s research seems to represent the best information we have.
4. The authors state that “having direct access to transgenic plants and their oil would be better”. Indeed, they don’t say where they obtained the GM and non-GM soybean oil or where the soybeans were grown. It’s likely that they simply bought it on the open market, as it’s generally easy to access GM and non-GM soy oil, with GM oil having to be labelled as such in Iran. If this assumption is correct, the study shows that GM soy oil bought on the open market caused adverse effects that non-GM soy oil did not. That’s a serious enough take-home message.
There are also some drafting errors in the paper which are addressed in an appendix below.
Regulatory authorisations based on a lie
GM soybean oil is routinely used by restaurants and other eateries in the UK and the EU (as an investigation of their dustbin areas often reveals), and it's likely that the same is true of Iran. A considerable body of evidence shows that GM soy has adverse effects on animals that eat it – though we don’t know if that’s a result of the herbicide residues present in GM glyphosate-tolerant soy, or of compositional changes caused by the GM process, or a combination of both. What is clear is that GM soy isn’t substantially equivalent, in terms of biological effects, to non-GM soy. That means its regulatory authorisations worldwide, which are based on the assumption of an equivalent safety profile with that of non-GM soy, are based on a lie and should be revoked.
The new study (open access):
Taheri H, Mesgari-Abbasi M, Khordadmehr M, Rahimi Mamaghani A, Abbasalizad-Farhangi M. Effect of genetically modified soybean oil consumption on biochemical and histological changes of liver and kidney in rats. Int J Drug Res Clin. 2024; 2: e11. doi: 10.34172/ijdrc.2024.e11. https://ijdrug.com/Article/ijdrc-3048
Appendix
The authors, editors, and peer-reviewers of the manuscript should have picked up on some drafting errors prior to publication.
First, the figures in Table 2 on one measurement seem to be at odds with the explanatory text. In the Results section, the authors state, “The GM soybean oil group showed a significant increase in serum insulin and GGT [gamma-glutamyl transferase] levels compared to the standard diet group (P = 0.047).” But Table 2 shows mean insulin levels in the standard diet group of animals as 15.5 μIU/ml; in the non-GM oil group as 42.1 μIU/ml, and the GM oil group as 17.8 μIU/ml. So, the non-GM oil group levels are recorded as being far higher than the GM oil group levels. It seems that either there has been a mis-tabulation of these results, with the values of 42.1 μIU/ml and 17.8 μIU/ml needing to be transposed, or the authors have misinterpreted their results.
There are other minor errors. For example, during their discussion of other studies, the authors state that “Key et al conducted a study demonstrating negative effects on kidney and liver function and tissue in rats fed with GM maize for 91 days”, referencing two papers – one by Key et al and the other by Manuela Malatesta et al.
The first paper, by Key et al, is actually a review. These authors didn't do any animal feeding experiments, nor do they cite any paper in their review that appears to fit Taheri et al’s description. Malatesta, on the other hand, did conduct a study in mice fed a GM diet and found that the GM diet caused signs of ageing in the liver. But her study was long-term, not over 91 days; it looked at soy, not maize; and was in mice, not rats.
Similarly, Taheri and colleagues correctly note that a previous study found that GM “MON 863 Bt corn caused significant kidney and liver inflammation and damage in rats”, but they incorrectly attribute the study’s authorship. This re-analysis of a Monsanto study that had misleadingly claimed to show no adverse effects from the GM maize was authored by Prof GE Séralini and colleagues – not the US-based campaigner Jeffrey Smith, as the authors incorrectly write. Smith simply wrote an article about the re-analysis.
However, these drafting errors do not invalidate the experimental findings, which should be evaluated on their own merits.