In my series 17 Questions about Glyphosate, question 2. looks at glyphosate and health effects, and also at what role surfactants might play in some of the results. *Added: Information about glyphosate breakdown product AMPA, and a section on endocrine disruptor claims and Seralini, et al. Question 3. looks at the colourful claims computer scientist Stephanie Seneff has produced by running samples of the scientific literature through her computer algorithms. These include ideas about glyphosate being behind celiac disease, autism, obesity, Parkinsons, diabetes, mental health problems, gastrointestinal problems, school shootings… you name it.
2. Could glyphosate have other health effects? What about the surfactants in RoundUp, or glyphosate breakdown products?
A number of different health concerns have been raised for glyphosate, not just worries about cancer. There is a trend in many of these concerns. Often the chain of conclusions is rather loose, and the argument relies on “glyphosate can be detected in x” (see post sections 7-11 for more about urine, breastmilk, wine, and wheat) – leaving it up to the reader to draw the conclusion that this detection must be significant and should make us worry about health effects, even though the evidence on those is left out or only loosely implied.
The other tactic is to claim that high levels of glyphosate could have an effect x, so we must be wary of the substance in any form, leaving aside the discussion about what the realistic levels of exposure are and how they relate to that risk. This idea of absolute avoidance creeps into our thoughts quite easily when it comes to new ‘invisible’ chemical risks which we are not personally in control of, whereas with familiar risks, such as salt (which is more toxic than glyphosate) and alcohol (which is definitely carcinogenic in the amounts commonly ingested), we tend to naturally understand that moderation is key.
The important question is whether there is a realistic basis to worry about health risks. The best way to find guidance to that question is to rely on the collected wisdom of the best and most comprehensive recent reviews on the research (examples here and here, or you can check out the collection of all glyphosate review papers in Food And Farm Discussion Labs wiki page). There have been at least eight reviews looking at human health and glyphosate science in the last 20 years, and all of them find no harm to human health caused by normal glyphosate use. Another good place to begin is at Biofortified, where plant geneticist Anastasia Bodnar has written an overview of the research landscape:
…three recent reviews that summarize the literature on glyphosate and humans: Epidemiologic studies of glyphosate and non-cancer health outcomes, Epidemiologic studies of glyphosate and cancer, and Developmental and reproductive outcomes in humans and animals after glyphosate exposure. These reviews looked at epidemiological studies, ones that look at disease incidence in large numbers of humans with varying levels of exposure to G or that look at exposure to G in a population that has a disease.
Now, epidemiology isn’t perfect, but with carefully designed studies it can be a powerful way to look for connections in real human populations. Even better when we can look at reviews that put multiple studies all in one place. These reviews cover a lot of studies that find there is no correlation between glyphosate exposure and cancer or non-cancer diseases.
She also cautions for perspective when it comes to single studies conducted in the lab with cell cultures:
There are occasionally alarm-inducing papers like Glyphosate induces human breast cancer cells growth via estrogen receptors. This paper, and others like it, tend to use human cells in a petri dish rather than whole animals. I had the misfortune to do some research on cultured human cells myself and let me tell you, those are some tricky buggers to work with. Even when everything is working perfectly, it’s still very hard to tell if the results you are getting will hold true when repeated in a whole animal model. Something that causes a reaction in naked cells may not react the same when applied to your skin or taken in through your digestive system (both of which have evolved to keep you safe from many things).
Only a combination of animal models and cell studies can give us the full picture (even better if we can pair these up with some epidemiology).
What about the surfactants (soap-like substances) in pesticides?
To allow for better spread, glyphosate mixtures usually contain surfactants. In a wide array of studies, the substance under study is a mix of glyphosate and surfactants – most commonly one called polyethoxylated tallow amine (POEA). Surfactants are substances such as soaps and other amphiphilic molecules, which have the ability to change surface tension properties between liquids (or liquids and solids). They also help solubilise fats and proteins, that is, they act as detergents. Sometimes there may also be other pesticides added (combination pesticide mixtures). Each agent will have an effect of its own, which of course becomes pronounced if the target animal or cell culture is exposed to high enough concentrations for a long enough period of time.
Not surprisingly, cells do not like direct exposure to large concentrations of agents that solubilise their structural components, such as surfactants. Neither do aquatic animals. We can tolerate washing our hands with soaps because of the barrier-function of our skin, and the small amounts that might be ingested (from our own soaps, or from pesticide residue) are so vanishingly small as to be of no concern. But we would also most certainly experience some adverse effects if we actually drank soap, and we do not tend to require that our soaps should be safe if ingested in large amounts, because this is not their intended use. This is the relevance we need to look for in studies: that they look at realistic exposures (whether it’s about soap or glyphosate). It is the dose that makes the poison.
Glyphosate breakdown product AMPA
The major breakdown product of glyphosate is aminomethylphosphonic acid (AMPA), which has been found to have similarly low profile of toxicity as glyphosate according to the report from Food and Agricultural Organisation of the WHO. One review from 2000 has looked at both glyphosate, the surfactant POEA, and AMPA specifically. They conclude:
The oral absorption of glyphosate and AMPA is low, and both materials are eliminated essentially unmetabolized. Dermal penetration studies with Roundup showed very low absorption. Experimental evidence has shown that neither glyphosate nor AMPA bioaccumulates in any animal tissue. No significant toxicity occurred in acute, subchronic, and chronic studies. […]
Therefore, it is concluded that the use of Roundup herbicide does not result in adverse effects on development, reproduction, or endocrine systems in humans and other mammals. […]
It was concluded that, under present and expected conditions of use, Roundup herbicide does not pose a health risk to humans.
A quick a nod to the topic of cancer, which I delved into more in the first post of this series: the field is complicated, as even among our common foodstuffs, you can find evidence for a great number that have demonstrated both a cancer-causing and a cancer-protective effect. Keeping that in mind, perhaps it is not so surprising what I found when I dived into the studies on PubMed. One of the most recent studies on AMPA actually looks at the substance as a prospective cancer treatment:
these results demonstrate that AMPA can inhibit prostate cancer growth and metastasis, suggesting that AMPA may be developed into a therapeutic agent for the treatment of prostate cancer.
This goes to show that single studies may raise alarm or give hope about any number of substances as either harmful or beneficial. Time and accumulating number of quality studies that either support or reject those views, are the critical factor in making sense of the science. So far this process has found neither glyphosate, nor its common companion surfactants, or breakdown products, to pose a risk for us or the environment through normal exposure.
Could glyphosate be an endocrine disruptor? Enter Seralini et al.
The French researcher Seralini has published a paper that claims glyphosate to be an endocrine disruptor. This is a good example of what Anastasia Bodnar cautioned about above: the study was done entirely on cell lines, which gives us little idea about the relevance when it comes to food residues. Exposing cells cultures directly to high levels of glyphosate and its formulations with surfactants will most likely have adverse effects on the cell lines, which they observed.
There are some red flags here too. Three years after this study, Seralini published another one, which he is probably most famous for. It was a genetically engineered corn feeding study on a specific kind of laboratory rat breed which is naturally prone to tumours. He kept the rats alive and suffering well past any ethically defensible time-frame (two years), and had his study redacted due to its confusing study design, flawed statistics, and a conclusion not supported by the data. If you are interested, you can read more about the affair in Nature or Wikipedia.
It may come as no surprise, then, that even his earlier endocrine disruption study has not lead to more research that would have confirmed the suggested endocrine effects. A 2015 report from Environmental Protection Agency (EPA) concludes that there is no convincing evidence that would warrant more extensive testing:
Based on weight of evidence considerations, mammalian or wildlife EDSP Tier 2 testing is not recommended for glyphosate since there was no convincing evidence of potential interaction with the estrogen, androgen or thyroid pathways.
It is one thing to suggest that an adverse effect observed in a a single cell line study would have greater relevance. If the results are not confirmed in other ways, however, and if the data simply does not get support in any of the epidemiological studies from decades of human populations exposure to the pesticide residues, then the suggestion is probably not that relevant after all.
Science-based Medicine has also summarised the topic of glyphosate’s health effects in their piece Glyphosate – The New Bogeyman, coming to a conclusion very similar to the one Anastasia Bodnar made above:
…numerous published systematic reviews show clear evidence that glyphosate has very low toxicity. More careful study when it comes to any agent being used as heavily as glyphosate is always welcome. Science is complicated, and it is always a good idea to consider factors that may have been previously missed. However, failure to show any adverse effect from glyphosate in epidemiological studies is very reassuring. Given its widespread use, any adverse effect must be tiny or non-existent to be missed by the evidence we have so far.
The evidence, however, will not stop ideologues from cherry picking, misusing evidence, presenting pure speculation as if it were evidence, assuming causation from correlation, and generally fearmongering about a safe chemical in order to grind their ideological axe.
3. What about studies claiming glyphosate causes celiac disease, autism, and obesity, etc? A look at Seneff et al.
For scientists, it may be easy to navigate the landscape of scientific publications, but most people have no experience and knowledge of how to easily discern whether the claims being made are incredibly far-fetched or fundamentally lacking in evidence; if it is so that the authors are not actually qualified in the field in question; or that the publishing journal itself is of a questionable nature.
To get a better understanding of the latter factor, you can read more about predatory journals in discussed in this piece in Nature. There are also questionable publishing groups which are buying up small but previously reputable journals in order to increase their revenues from having less successful scientists pay for the semblance of credibility they get from publishing their pieces in what technically qualifies a scientific publication. These publications are not very hard for professional scientists to spot, but they might fool any number of outsiders to the field. This is why it is best for outsiders to the field rather to rely on the views supported in large review papers which work to summarise a topic.
In the thousands of small ditches between the open waterways of mainstream scientific publishing, you can find all kind of puddles of muddled semi-scientific attempts. As an example, there is a paper that looks at results from a fish study. The analysis is written by an electrical engineer S. Seneff and a retired consultant A. Samsel, and they use computer science methods such as Natural Language Parsing to analyse papers in order to ‘help her figure out the story’ behind them. Using these methods, they claim that they can see that glyphosate causes celiac disease. The claim is based on an Indian fish study, where fish were kept in water which was supplied with many times field-realistic levels of a glyphosate mixture for weeks. Importantly, the mixture included surfactants, which are not allowed in aquatic use in the first place, because surfactants are very harmful to fish. There was no control for the surfactant. And the connection to celiac disease? It was their ‘logical conclusion’, seeing as the fish had seriously damaged intestines after six weeks in this surfactant bath. I think I do not need to point out that general intestinal damage in fish after 1.5 month long potent pesticide-soap-bath hardly equals proof about celiac disease. You can read more from Steve Savage in a fishy attempt to link glyphosate and celiac disease.
Seneff, together with Samsel, does not stop there, but they also claim that glyphosate is behind a whole host of ‘modern diseases’ (they seem to subscribe to the strange notion that most diseases did no exist before modern times). According to Seneff’s analysis, glyphosate is also behind the development of, hold on to your hats: gastrointestinal disorders, cancer, obesity, diabetes, heart disease, autism, Alzheimer’s disease, depression, anxiety syndrome, Parkinson’s disease, prion diseases, infertility, and birth defects.In an interview with Jeffrey Smith (around the 28 minute mark)In an interview with Jeffrey Smith (around the 28 minute mark), Seneff goes as far as to blame glyphosate (RoundUp) for school shootings and the Boston Bombings. I don’t even know what to say to that. Maybe we should also blame glyphosate for climate change?
The red flags could not be much more numerous. Let’s list some: 1. Seneff is not a chemist, a biologist or a medical doctor, so she is publishing outside of her field. 2. The claim that any one specific thing is the cause of a wide variety of startlingly different kinds of diseases. (Same warning goes for any one things that supposedly cures every number of different diseases.) 3. She doesn’t produce any actual new data, just reinterpretes old data. 4. She publishes in low-quality, pay-to-play predatory journals, some of which are not even connected to the field of biology. Seneff’s article on glyphosate and gut microbiota is actually used as a good example of how to spot a bogus scientific journal. In the piece they write:
This article is attributing pretty much all the chronic diseases of the modern world to a single agent, glyphosate. Wouldn’t it be fantastic if just by getting rid of one chemical we could be as healthy and happy as we have never been?
Need I mention, once more, that several comprehensive reviews by actual experts in human disease and biology have not found any epidemiological connection between any of these diseases and glyphosate?
To address concerns about several of these hypothetically connected diseases listed above, the US National Academies of Science included in their 2016 report a comparison of the rates of disease incidence between USA (where glyphosate use is much more abundant) and other countries such as UK. For the prevalence of celiac disease, autism and food allergies, they found no difference between USA and UK:
Celiac-disease detection began increasing in the United States before the introduction of GE crops and the associated increased use of glyphosate; the disease appears to have increased similarly in the United Kingdom, where GE foods are not typically consumed and glyphosate use did not increase. The similarity in patterns of increase in autism spectrum disorder in children in the United States and the United Kingdom does not support the hypothesis of a link between eating GE foods and the prevalence of the disorder. The committee also did not find a relationship between consumption of GE foods and the increase in prevalence of food allergies.
A word about autism
To lift out one of the topics on that list, I would like to make a comment about what the scientific evidence has to say about the causes of autism. The Autism Science Foundation gives good overviews of autism research, and if you look at their reports on studies about autism risk factors and environmental factors, the list of possible connections to autism is really long (note, it does not contain glyphosate):
Maternal age, paternal age, mother’s influenza/untreated fever during pregnancy, diabetes, air pollution, cigarette smoke, lead, methylmercury, polychlorinated biphenyls, arsenic, and toluene, manganese, fluoride, chlorpyrifos, dichlorodiphenyltrichloroethane, tetrachloroethylene, and polybrominated diphenyl ethers, organophosphate pesticides, maternal childhood abuse, placental folds…
Autism is fundamentally a genetic disease, and autism spectrum is viewed as a neural ‘type’, not a disease. It is something that is largely hereditary, it develops during gestation, and it can be exacerbated by additional (related) new point mutations connected to neural development before or during fetal development. There is a lot of research into what could cause potential epigenetic and other factors that would make some of the more difficult aspects of autism more pronounced.
Considering the number and diversity of these factors, it’s best to wait for comprehensive meta-analyses or reviews on these topics before making too strong conclusions. One of the recent meta-analyses was about maternal age:
The results of this meta-analysis support an association between advancing maternal age and risk of autism. The association persisted after the effects of paternal age and other potential confounders had been considered, supporting an independent relation between higher maternal age and autism.
The topic is complex enough without somebody manufacturing hypotehtical connections and creating more noise for the worried parents and dedicated researchers to wade through, and making strong claims without evidence and spreading them in the media like Seneff does, is very irresponsible behaviour.
If you would like to read more about all the wild claims and innumerable logical lapses in Seneff’s papers, there are many good places to turn to. Here are a couple: at Biofortified Medical Doctors Weigh in on Glyphosate Claim, Orac on Oh no GMOs are going to make everyone autistic, or even Snopes in their piece on Glyphosatan – Unsupported claims assert that one in two children will be autistic by 2025 due to the use of glyphosate (Roundup) on food crops.
If you are interested in the other health effect claims, not discussed under the scope of this post, you can find two more topics over at: 1. Does Glyphosate Cause Cancer? and 4. Does Glyphosate Harm Gut Bacteria?, or at the parent page 17 Questions about Glyphosate.
New innovative research is always welcome, especially for a substance as widely used as glyphosate, but we should always strive to honestly evaluate the evidence before forming our views on a topic. As the numerous examples above demonstrate, the greatest glyphosate-resistance around may indeed be one of a more psychological kind: it has become a fix idea in many minds that glyphosate must be behind a whole host of ills in our world. Trying desperately to fit the evidence into the idea, rather than allowing our ideas to be shaped by the evidence, is what has resulted in this process of claim-whack-a-mole. I have no doubt that next month some new variation of glyphosate-sensationalist news will give wings to yet another far-fetched or misleading claim. The game might never come to a real conclusion, for it may be that for many, the only acceptable kind of world is one where glyphosate can only be a bad guy.
If you are interested in other environmental or health topics, you can find my other pieces and further resources under Farming and GMOs, The Environment, and Vaccines and Health. If you would like to have a discussion in the comments below, please take note of my Commenting policy. In a nutshell:
- Be respectful.
- Back up your claims with evidence.