Surfactant co-formulants in glyphosate-based herbicides: current gaps, and paths forward in human biomonitoring

Abstract

Polyethoxylated tallow amine (POEA) surfactants in glyphosate formulations are understudied. They may constitute greater health risks than glyphosate itself. Lack of validated biomarkers of exposure and metabolism, as well as analytical methods for measuring POEA, limit study of a formulation’s toxicity and associated risk.

Glyphosate-based herbicides (GBHs) are mixtures of chemicals rather than a single compound.¹ The commercial product includes "active" ingredients that are subject to strict regulations for home and agricultural use. As a result, the main focus on human exposures has been limited to glyphosate, the active ingredient of GBHs. Little information is available about additives in the formulation. They facilitate glyphosate absorption and perfusion into plants. The most often employed co-formulants are polyethoxylated tallow amine (POEA) surfactants.¹ Recent studies suggest that POEA additives in GBHs, rather than the glyphosate itself, exert significant endocrine-disrupting effects in human cells.² However, there is an enormous knowledge gap about human exposure to POEA co-formulants including their toxicokinetics and appropriate biological matrices for biomonitoring (Fig. 1).

Figure 1

Non-ionic POEA surfactants are synthesized from fatty acids with 12-18 carbons derived from animal fat, among them oleic acid, palmitic acid, stearic acid, palmitoleic acid, and myristic acid. The fatty acids are ethoxylated into tallow amines by adding ethylene oxide. One alkyl chain (R) and two repeating ethoxylate (EO) units make up the resulting tertiary amine.² The tallow amine R chain is primarily composed of C₁₆H₃₃, C₁₈H₃₅, or C₁₈H₃₇ moieties. The two EO units (m and n) add up to 2 to 15 moles of ethylene oxide, but some GBHs contain POEA with higher amounts of polyethoxylation.² There is a great deal of uncertainty regarding the structures and quantities of co-formulants because their composition is often proprietary. Yet this complexity creates several problems for quantification and biomonitoring.³ Minor compositional changes can have a significant impact on properties of POEAs. Toxicity and environmental persistence can vary with length of the alkyl and ethoxylated chains. Therefore, a biomonitoring method must overcome obstacles in the type of biospecimen, cleanup, and analytical sensitivity for detection at biologically relevant concentrations.

Choice of biospecimen to be used for optimal biomonitoring requires knowledge of pharmacokinetics – absorption and disposition in the body –to identify the compartment where an absorbed analyte or metabolite exists in greatest concentration. POEAs have a structure similar to aliphatic alkyl polyethoxylates, also known as alcohol polyethoxylates. Therefore, it is likely that POEA metabolites are excreted from the body within a few days, similar to pharmacokinetics of alkyl polyethoxylates.⁴ However, knowledge about oxidized metabolites of POEA would improve quantitative biomonitoring. It would also provide insight into the metabolism and urinary excretion fraction (FUE) to back-calculate exposures to POEA, as has been done for nonylphenol ethoxylates.⁵ POEAs with shorter EO chains were more frequently found in human plasma than those with longer chains,⁶ which was ascribed to (i) the addition of POEAs with a low number of EO units to the GBH and (ii) the possibility that POEAs might go through non-oxidative metabolism to yield shorter EO chains, similar to alkyl polyethoxylates.

An instrumental method should detect POEA homologues and metabolites.² An initial obstacle is that analytical standards for native or radiolabeled individual homologues or their mixtures are not available. There is also a lack of stable-isotope-labeled POEA homologues to use as internal standards in bioassays. Although not ideal, standards of homologues are often used for mass spectrometric (MS) quantitation, as they are assumed to have similar MS response factors and chromatographic behavior. Quantification accuracy suffers when calibration curves are based on "total" quantities rather than individual homologues. Internal standards are essential to avoid or correct for analyte concentration losses during sample pre-treatment, injection, chromatographic separation, and mass spectrometric detection. Reduction of matrix interference with chromatographic efficiency and MS detection is necessary to provide increased sensitivity for measuring biologically relevant concentrations. This requires adequate sample cleanup and extraction to preconcentrate biomarkers.

The instrumental analysis should monitor functional groups of POEA metabolites. Using functional group-specific neutral loss from the parent ion, MS can find out how a chemical change from a parent ion to a product ion. As class-diagnostic product ions (Prod) or neutral losses (NL) are often created by chemical properties, these approaches allow profiling or screening for discovery analysis.⁷ As an example, the NL scan of m/z 282.0 and the Prod scan of m/z 400.6 in positive mode can detect homologues of a POEA with ten to fourteen EO units by following the fragmentation pattern.⁸ Based on a method developed for urinary POEA biomarker measurement, a higher detection rate and levels of POE (2) were found in urine from an animal study compared to POE with longer chains, glyphosate, or its metabolite aminomethylphosphonic acid (AMPA).⁹

Exposure assessment to POEA is a public health need. Formulation of GBHs varies widely globally and even among those from the same manufacturer; their exact compositions are often unknown. Regulatory posture treats herbicide co-formulants as inert, so their presence in the environment, food, feed, animals and people is not formally monitored. Methods are needed to monitor residues in the food supply, as well as people, cattle, and other animals. In order to make sure that herbicide exposure does not have "unacceptable adverse effects," guideline values are derived from toxicity tests. Usually, the only information provided on product labels is concentration of an active ingredient. Recognizing these differences and figuring out their effect on GBH risk profiles is important. More studies are urgently needed on the potential contribution of co-formulants to GBHs to human exposure and to the toxicity of herbicides.