Glyphosate-based herbicide: a risk factor for demyelinating conditions of the peripheral nervous system?

Glyphosate is a broad-spectrum herbicide originally introduced to the market in 1974 by the agrochemical company Monsanto. More than 40 years down the line, glyphosate has become one of the most economically meaningful herbicides, with a global use of more than 1.8 million pounds in 2014 (Benbrook, 2016). In non resistant plants, glyphosate is widely believed to exert its herbicidal effect via inhibition of the 5-enolpyruvylshikimate-3-phosphate synthase, an enzyme of the shikimate pathway required for the biosynthesis of aromatic amino acids in plants and most microorganisms such as fungi, bacteria and some protozoans. As the shikimate pathway has no known physiological function in mammals, glyphosate was considered safe for humans. However, this view has been challenged by several studies conducted by researchers from various fields, leading to the assumption that agricultural spreading of glyphosate might bear health risks, including but not limited to carcinogenic, inflammatory and endocrine disruptive effects (Mesnage et al., 2015). While the potential health risks associated with exposure to residues of pure glyphosate from food and environmental contaminations remain a matter of controversial debate, glyphosate-based herbicide (GBH) formulations have almost unequivocally been demonstrated to exert concentration-dependent cytotoxicity. Here, one point particularly worthy of note is that glyphosate is always used as herbicide formulation and not as a pure substance. A significant portion of GBH products contain the isopropylamine salt of glyphosate alongside mostly unspecified mixtures of surfactants and auxiliary agents. These may include petroleum distillates and polyoxyethyleneamines (POEAs) (Defarge et al., 2017). POEAs have been found to contribute substantially to the toxicity of GBH formulations. Following a risk assessment by the European Food Safety Authority, POEAs have been banned as co-formulants for herbicides in the European Union since mid-2016. Nevertheless, outside the European Union, POEAs are still commonly used as surfactants in GBH formulations. Moreover, a recent investigation found several herbicide formulations to contain heavy metals (Defarge et al., 2017). While numerous studies have focused on general cytotoxicity and carcinogenicity of glyphosate or GBH, very little is known about specific effects of these compounds on the nervous system and no studies have been conducted so far to investigate their impact on the peripheral nervous system (PNS). For this reason, we decided to study and compare the effects of both pure glyphosate and a commercially available GBH product (Roundup LB Plus, Monsanto) on murine embryonic dorsal root ganglia (DRG) explant cultures as an ex vivo model of neurite outgrowth and myelination in the PNS (Szepanowski et al., 2018a). Following explantation, these DRG cultures were initially kept under neurite-stimulating conditions before myelination of the outgrown neurites was subsequently induced. Glyphosate was then added to the culture medium either directly at the start of myelination-induction or following an initial myelination period of 10 days. Finally, the cultures were stained with Sudan Black and the myelin content was determined by measuring the total number of internodes per neurons present in individual culture wells (Figure 1). Unexpectedly, we did not find any significant impact of pure glyphosate, even at considerably high concentrations, on the DRG cultures under both culture conditions: Pure glyphosate did not interfere with myelination or cause demyelination in cultures with pre-existing myelin. In sharp contrast, treatment of cultures with the GBH product significantly impaired myelination and was associated with a concentration-dependent demyelinating effect. Of note, the observed effects of GBH on DRG cultures appeared to be exclusively related to an impairment of myelin rather than neurite integrity. Neither glyphosate nor GBH treatment were associated with reduced neurite outgrowth kinetics and neither affected neurite density following exposure times of up to 10 days. As the GBH contained glyphosate as isopropylamine salt rather than pure glyphosate, we next tested whether a mixture of pure glyphosate and isopropylamine would exert a similar demyelinating effect to that of the GBH formulation. We found a slight, although non-significant reduction in the number of internodes per neurons between pure glyphosate or glyphosate plus isopropylamine. Thus, isopropylamine may have a minor damaging effect on myelin, possibly by saponification of lipids, but its impact was not comparable to that of GBH treatment. These findings suggest that undisclosed additives in the GBH product might be responsible for the observed demyelinating effect. To better understand relevant mechanisms that might explain the observed differential effect of pure glyphosate and GBH on myelin integrity, we next assessed the influence of GBH on markers of oxidative/ nitrosative stress and cell damage, such as malondialdehyde, nitric oxide and lactase dehydrogenase activity. However, we could not find specific effects of GBH on any of these markers. Since these measures for cellular damage did not yield any results specifically related to GBH, we focused on another mechanism that might contribute to demyelination and also prevent myelination of unmyelinated cultures, namely inflammatory Schwann cell activation. Schwann cells, as myelin-forming cells of the PNS, are generally accepted to be immunocompetent cells, resembling several features of classical innate immune cells (Ydens et al., 2013). Similar to macrophages, Schwann cells express a wide range of pattern recognition re-