A recent study published in the open-access journal, African Journal of Biotechnology, entitled "Addressing the issue of horizontal gene transfer from a diet containing genetically modified components into rat tissues" has been making the rounds on social media and throughout the anti-GMO community.
Before we break down the study, it's important to note that it follows up on a 2013 study by Spisak, et al, in PloS One (1), that caused some controversy, because it's results suggested that whole plant genes crosse from the digestive tract into the human blood stream. These findings were immediately jumped upon by the typical anti- GMO media (Natural News, Collective Evolution, etc etc here, here, here, here) where the story incited fear that trans-genes in plants could be floating around in our blood stream. The actual study wasn't specific to GMOs; actually, GMOs are only briefly mentioned. The findings were controversial because the idea that whole genes can pass across the intestinal tract is generally considered physiologically impossible - DNA is digested into short fragments and its constituent nucleotides before absorption. The authors of the Spisak paper even mention that the mechanism of this absorption is unknown. Unfortunately, the Spisak paper didn't include a negative control, so we're not sure if this finding is due to contamination, rather than actually having whole plant DNA in human blood. Making the study even weirder, they found high levels of plant and tomato DNA in all of their samples and only looked at DNA from the chloroplasts. We're left wondering whether this has any actual clinical significance if it's even a true finding.
This newest African Biotech study (2) claims to demonstrate horizontal gene transfer, from GM crops to the rats studied - proving that would be a rather tall order and something that would make it to a big name journal. I was skeptical from the start, given its publication in a journal that's relatively low impact and open access. The author also has another study out this year (3) that claims to show organ damage from feeding rats a GMO containing diet - but a quick look at that study reveals a number of sloppy flaws that characterize many studies that purport to find a negative effect from GMO feeding: they don't tell us the number of control animals, the diets between the two are different, and the histology is only presented as a few slides showing organ damage.
This newest study, unfortunately, is not free of methodological issues*. The authors used 29 male Wistar rats, and broke them into a non-GMO and GMO feed group. From the start, we've got some issues. The OECD protocols for performing short-term(90days) feeding studies in lab animals suggest that 10 animals be used per group, per sex (4). The rats in the GMO feed group, we are told, were fed a diet containing 60% maize and 34% soybean - the diets were confirmed to contain genetically modified materials via amplification of a sequence of the Cauliflower mosaic virus. We're not told what percentage of the diet contained GMO foods; a well-designed GE study will mention the percentage of the feed that is GE (usually about 1/3). The rats fed in the non-GMO group, we are told, are fed a 'well-balanced diet' - what does that mean?. This is a sloppy design from the get go.
We're told that the animals were further divided into groups, with the GMO diet groups containing 6,6, and 7 individuals. This means that 19 individuals were in the GMO group, and 10 were in the control group - again, odd to have this sort of pairing off (especially when the results only show a few individuals from both groups). The animals were split into 6, 6, and 7 individuals to be euthanized at 30, 60 and 90 day time points. Control animals were euthanized at the end of the experiment. Blood, liver, and brain tissues were collected. Oddly, brain and liver tissue were only stored at -20degrees celsius (I've never stored tissue at less than -80). DNA was isolated from both the diet, and the tissues, and 3 segments of the cauliflower mosaic virus (CAMV) promoter were amplified. I really wish we had some sort of positive control to show that these primers were specific - the authors mention that the primers were tested for specificity by running a BLAST of the sequences against the bacteria, mouse and rat genomes; however, they don't run them against the maize or soy genomes, so i'm not sure if these sequences would amplify other DNA sequences (since we don't know the control diet, we can't confidently say that the primers were specific because they didn't amplify sequences in the control diet - the other GMO article I mentioned from this group used a wheat, not corn or soy diet, in the control group).
The results, not surprisingly, confirm the presence of the larger fragment of the CAMV promoter in the GE diet. The authors state that they are able to detect CAMV promoter fragments in some tissues of some animals, and only "in some cases in the present work, the three segments were all amplified in the same animal". The authors list that the frequency of transfer of the GM-target sequences from the GM diet into the examined tissues is 33.3 to 37% after 30 to 60 day feeds, and reached 50% after 90 days.
The paper has a major flaw in its interpretation of the results. The authors boldly state that the DNA fragments are incorporated into the genomes of the target tissues (Brain and liver). They don't actually show this though; we would need something like fluorescent in-situ hybridization to determine where those DNA fragments are located. Right now, we're likely just amplifying the DNA fragments that were in the blood vessels of those tissues. If the fragments were incorporated (highly unlikely as the CaMV promoter has never been shown to infect mammals, only plants), it wasn't shown here. Previous literature by Mazza et al, 2005, investigating the uptake of small DNA fragments (from GE plants and non-GE) concluded that the blood was the major organ that uptook DNA fragments, followed by blood rich tissues (5). Mazza et al concluded that "the risk that a transgene is transferred to an organism is not higher than the risk for the transmission of other plant genes and therefore consumption of GMO should not raise concerns about safety in animal production as well as, based on physiological similarities in human nutrition". This is consistent with the reports of all major scientific bodies, that genetic engineering doesn't introduce unique threats relative to conventional breeding. Despite the title, the paper doesn't actually show evidence of gene flow.
A minor issue that I have with the paper is the lack of discussion surrounding contamination. It's not mentioned what steps were taken to prevent contamination. While the addition of the control group, which showed no amplification of the CAMV segments, adds some confidence, we're not sure whether care was taken. Seeing as the authors did test the feed for the presence of DNA fragments, it's highly possible that their results are skewed via contamination. Kevin Folta recently pointed out an issue with their amount of starting DNA (very large, 100ng) and that they used 40 cycles of PCR to amplify the sequences (you should be able to amplify by 30, anymore and you're probably getting contamination). Another minor issue - the gels in this paper are pretty weak (very faint lines, kinda messy), and the graphs are terribly made (not even a standard error bar!).
While some on twitter have thrown this paper around and tried to show it as illustrating harm of GMOs, or new reason to invoke the precautionary principle, it doesn't. There are no functional outcomes in the paper, and further, it doesn't do what it says it does (show incorporation of foreign DNA into the host DNA). I'd be particularly worried if long term/intergenerational feeding studies showed some level of harm from the consumption of GMOs, but they don't (6). Most importantly, for those using this study to raise concerns about GMOs, one should note that the CAMV promoter isn't unique to GE foods - it's a common virus found in nature that infects many plants in the cabbage family, including cauliflower and brussels sprouts - AcademicsReview covers that here. If the results of this paper concern you, you might also be concerned with eating conventional produce #labelit?
Sidenote: If one Googles the CAMV promoter, they'll find a slew of websites, like 'independentsciencenews', claiming that regulators found a new viral sequence potentially not suitable for human consumption. The European Food Safety Authority responded to these claims here , noting that the sequence was known about, and that assessments have included the potential for unintended consequences, despite none being known.
*I receive no compensation and have no affiliation with Monsanto or any biotech company.
1. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0069805
2. http://www.academicjournals.org/article/article1417172424_Oraby%20et%20al.pdf
3. http://journals.tubitak.gov.tr/havuz/biy-1406-61.pdf
4. http://www.jasbsci.com/content/4/1/37
5. http://www.ncbi.nlm.nih.gov/pubmed/16245168
6. http://www.sciencedirect.com/science/article/pii/S0278691511006399
Before we break down the study, it's important to note that it follows up on a 2013 study by Spisak, et al, in PloS One (1), that caused some controversy, because it's results suggested that whole plant genes crosse from the digestive tract into the human blood stream. These findings were immediately jumped upon by the typical anti- GMO media (Natural News, Collective Evolution, etc etc here, here, here, here) where the story incited fear that trans-genes in plants could be floating around in our blood stream. The actual study wasn't specific to GMOs; actually, GMOs are only briefly mentioned. The findings were controversial because the idea that whole genes can pass across the intestinal tract is generally considered physiologically impossible - DNA is digested into short fragments and its constituent nucleotides before absorption. The authors of the Spisak paper even mention that the mechanism of this absorption is unknown. Unfortunately, the Spisak paper didn't include a negative control, so we're not sure if this finding is due to contamination, rather than actually having whole plant DNA in human blood. Making the study even weirder, they found high levels of plant and tomato DNA in all of their samples and only looked at DNA from the chloroplasts. We're left wondering whether this has any actual clinical significance if it's even a true finding.
This newest African Biotech study (2) claims to demonstrate horizontal gene transfer, from GM crops to the rats studied - proving that would be a rather tall order and something that would make it to a big name journal. I was skeptical from the start, given its publication in a journal that's relatively low impact and open access. The author also has another study out this year (3) that claims to show organ damage from feeding rats a GMO containing diet - but a quick look at that study reveals a number of sloppy flaws that characterize many studies that purport to find a negative effect from GMO feeding: they don't tell us the number of control animals, the diets between the two are different, and the histology is only presented as a few slides showing organ damage.
This newest study, unfortunately, is not free of methodological issues*. The authors used 29 male Wistar rats, and broke them into a non-GMO and GMO feed group. From the start, we've got some issues. The OECD protocols for performing short-term(90days) feeding studies in lab animals suggest that 10 animals be used per group, per sex (4). The rats in the GMO feed group, we are told, were fed a diet containing 60% maize and 34% soybean - the diets were confirmed to contain genetically modified materials via amplification of a sequence of the Cauliflower mosaic virus. We're not told what percentage of the diet contained GMO foods; a well-designed GE study will mention the percentage of the feed that is GE (usually about 1/3). The rats fed in the non-GMO group, we are told, are fed a 'well-balanced diet' - what does that mean?. This is a sloppy design from the get go.
We're told that the animals were further divided into groups, with the GMO diet groups containing 6,6, and 7 individuals. This means that 19 individuals were in the GMO group, and 10 were in the control group - again, odd to have this sort of pairing off (especially when the results only show a few individuals from both groups). The animals were split into 6, 6, and 7 individuals to be euthanized at 30, 60 and 90 day time points. Control animals were euthanized at the end of the experiment. Blood, liver, and brain tissues were collected. Oddly, brain and liver tissue were only stored at -20degrees celsius (I've never stored tissue at less than -80). DNA was isolated from both the diet, and the tissues, and 3 segments of the cauliflower mosaic virus (CAMV) promoter were amplified. I really wish we had some sort of positive control to show that these primers were specific - the authors mention that the primers were tested for specificity by running a BLAST of the sequences against the bacteria, mouse and rat genomes; however, they don't run them against the maize or soy genomes, so i'm not sure if these sequences would amplify other DNA sequences (since we don't know the control diet, we can't confidently say that the primers were specific because they didn't amplify sequences in the control diet - the other GMO article I mentioned from this group used a wheat, not corn or soy diet, in the control group).
The results, not surprisingly, confirm the presence of the larger fragment of the CAMV promoter in the GE diet. The authors state that they are able to detect CAMV promoter fragments in some tissues of some animals, and only "in some cases in the present work, the three segments were all amplified in the same animal". The authors list that the frequency of transfer of the GM-target sequences from the GM diet into the examined tissues is 33.3 to 37% after 30 to 60 day feeds, and reached 50% after 90 days.
The paper has a major flaw in its interpretation of the results. The authors boldly state that the DNA fragments are incorporated into the genomes of the target tissues (Brain and liver). They don't actually show this though; we would need something like fluorescent in-situ hybridization to determine where those DNA fragments are located. Right now, we're likely just amplifying the DNA fragments that were in the blood vessels of those tissues. If the fragments were incorporated (highly unlikely as the CaMV promoter has never been shown to infect mammals, only plants), it wasn't shown here. Previous literature by Mazza et al, 2005, investigating the uptake of small DNA fragments (from GE plants and non-GE) concluded that the blood was the major organ that uptook DNA fragments, followed by blood rich tissues (5). Mazza et al concluded that "the risk that a transgene is transferred to an organism is not higher than the risk for the transmission of other plant genes and therefore consumption of GMO should not raise concerns about safety in animal production as well as, based on physiological similarities in human nutrition". This is consistent with the reports of all major scientific bodies, that genetic engineering doesn't introduce unique threats relative to conventional breeding. Despite the title, the paper doesn't actually show evidence of gene flow.
A minor issue that I have with the paper is the lack of discussion surrounding contamination. It's not mentioned what steps were taken to prevent contamination. While the addition of the control group, which showed no amplification of the CAMV segments, adds some confidence, we're not sure whether care was taken. Seeing as the authors did test the feed for the presence of DNA fragments, it's highly possible that their results are skewed via contamination. Kevin Folta recently pointed out an issue with their amount of starting DNA (very large, 100ng) and that they used 40 cycles of PCR to amplify the sequences (you should be able to amplify by 30, anymore and you're probably getting contamination). Another minor issue - the gels in this paper are pretty weak (very faint lines, kinda messy), and the graphs are terribly made (not even a standard error bar!).
While some on twitter have thrown this paper around and tried to show it as illustrating harm of GMOs, or new reason to invoke the precautionary principle, it doesn't. There are no functional outcomes in the paper, and further, it doesn't do what it says it does (show incorporation of foreign DNA into the host DNA). I'd be particularly worried if long term/intergenerational feeding studies showed some level of harm from the consumption of GMOs, but they don't (6). Most importantly, for those using this study to raise concerns about GMOs, one should note that the CAMV promoter isn't unique to GE foods - it's a common virus found in nature that infects many plants in the cabbage family, including cauliflower and brussels sprouts - AcademicsReview covers that here. If the results of this paper concern you, you might also be concerned with eating conventional produce #labelit?
Sidenote: If one Googles the CAMV promoter, they'll find a slew of websites, like 'independentsciencenews', claiming that regulators found a new viral sequence potentially not suitable for human consumption. The European Food Safety Authority responded to these claims here , noting that the sequence was known about, and that assessments have included the potential for unintended consequences, despite none being known.
*I receive no compensation and have no affiliation with Monsanto or any biotech company.
1. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0069805
2. http://www.academicjournals.org/article/article1417172424_Oraby%20et%20al.pdf
3. http://journals.tubitak.gov.tr/havuz/biy-1406-61.pdf
4. http://www.jasbsci.com/content/4/1/37
5. http://www.ncbi.nlm.nih.gov/pubmed/16245168
6. http://www.sciencedirect.com/science/article/pii/S0278691511006399
Comments
Post a Comment