Thanks for your question and affording us the opportunity to respond to this issue. Actually, the manuscript you are referring to by Shehata et al. is the second publication from this group (Monika Krüger lab, 2012). In both publications, the authors employed an unrealistic model system (in vitro), utilizing a few isolated bacterial strains to represent the complex mixed microbial population of the animal gut, and subjected them to high concentrations of glyphosate formulations. Most critically, in-vitro studies of an isolated organism(s) cannot begin to model the complexity of intestinal or rumen flora which is different in every individual based on physiology of each person or animal and the foods/feeds consumed.
The selected pathogens and a very limited number of non-pathogenic microorganisms constitute only a tiny fraction of normal intestinal flora in any given species and are not at all representative of gut flora or the interactions between bacterial species. These few selected strains cannot be taken to represent the overall bowel or rumen flora of poultry or livestock. While the test system employed is not terribly relevant to the real-life environment, any findings observed must also be considered in the context of these other important influences on gut flora.
The authors do not provide data adequate to support the conclusion that glyphosate in animal feed produces clinically relevant alterations of microbial flora in poultry and livestock. Nor is there any clinical evidence to suggest that the widespread use of glyphosate-tolerant feed components in animals has resulted in an increase in livestock disease or human illness or, indeed, that such alterations even occur. This conclusion is derived from actual studies on real animals. Such animal feeding studies routinely performed in poultry and cattle, fail to demonstrate an alteration in nutritional performance as a result of the feeding of glyphosate-tolerant feed components, and limited studies of high concentrations of glyphosate in sheep failed to demonstrate alterations of rumen physiology.
This is important because the myth that cows digest grass is not true. Instead, the bacteria in their rumens digests cellulose and other fibrous fractions. The cow takes advantage of the products of this degradation as well as they digest the bacterial cells. Such findings indicate that, if any alterations in microbial flora do occur at relevant concentrations of glyphosate, they are of no observable clinical significance. Importantly, Shehata et al. provided no evidence to support the existence of an in-vivo impact on bowel flora. There is no evidence that the use of glyphosate or of GM crops has resulted in an increased rate of human or animal infection with enteric pathogens (defined in reference to humans- animals carry these agents as normal flora), or that the rate of either human or animal botulism has increased.
So you might ask, well okay this might not be a realistic model, but why do we see any effect at all with glyphosate on bacteria? That is a good question, and is best answered by looking at the formulation of glyphosate that was applied. In both studies, the authors determined inhibitory concentrations using Roundup UltraMax®, a formulated product containing a surfactant blend as well as glyphosate and other minor formulation components.
It is important to remember that surfactants are detergents, like the ones found in common household products such as baby shampoo or hand soap. While the impact of the surfactant (or glyphosate alone) is not clear, as neither were tested alone, surfactants are routinely used to clean and disinfect feed and food processing facilities and maintain hygiene in animal facilities, and often have bacteriostatic or bacteriocidal effects at concentrations within the range of values reported for glyphosate.
So just how do animals like cows get botulism? Well bovine botulism is the result of pre-formed toxins in poorly-fermented feed or in improper feed sources (cattle grazing unintentionally on dead animal carcasses), not the result of the growth of C. Botulinum in animal gut, suggesting that the results of these publications are likely irrelevant to the occurrence of bovine botulism.