Influences of glyphosate residues in combination with varying concentrate feed proportions in the ration on the health of German Holstein cattle

For many decades, glyphosate has been the most used active substance in common non-selective herbicides worldwide. Although 5-enolpyruvylshikimic acid-3-phosphate synthase, the essential enzyme inhibited by glyphosate, is only present in plants, bacteria, and fungi, the intensive usage triggered public concerns regarding putative effects of glyphosate on health of humans and animals. Since glyphosate residues are commonly detected in ruminant feed like imported soybeans, these residues are consequently ingested leading to oral exposure of cattle and possibly in utero exposure of their offspring. Whereas in Germany, restrictions on usage of glyphosate-based herbicides were recently enacted, pre-harvest application has been, and still is, a common legal agricultural practice in other countries.
To analyze effects of glyphosate residues resulting from formerly realistic and common agricultural exposure conditions on cattle, controlled in vivo feeding studies with dairy cows and growing bulls were conducted. Since fiber content-dependent effects of glyphosate on ruminal microbial populations were postulated in literature, varying fiber contents and thus varying concentrate feed proportions (CFP) were considered in the rations of the animals. Next to putative fiber content/concentrate feed-dependent effects of glyphosate feeding different CFP were contemplated as a common feeding practice for considering animal specific requirements (ongoing lactation, growth…). For both feeding trials, a 2x2 factorial randomized block design was used. First, animals were blocked by characteristic performance parameters, like number of lactations (dairy cows, Paper I), or body weight gain (bulls, Paper II), followed by random allocation to one of the four feeding groups. Secondly, animals were fed rations either with (glyphosate groups; GLY) or without glyphosate residues (control groups; CON) in combination with high (HC groups) or low (LC groups) CFP for several weeks (dairy cows: 16 weeks; bulls: 15-19 weeks). After terminating feeding experimental rations and a depletion period, those dairy cows, which had been pregnant during the feeding trial, and their calves, which were exposed in utero, were examined directly after parturition and before calves had colostrum intake regarding persistent and drastic teratogenic effects of glyphosate (Paper III).
During feeding experimental rations for 16 weeks to the 61 dairy cows completing the trial, blood samples were collected every two weeks and analyzed for liver health-related blood parameters, like enzyme activities indicative for liver damage or levels of parameters associated to hepatic function. Additionally, liver biopsy was conducted three times and tissue samples were histopathologically evaluated. At the end of trial, RNA of liver tissue was isolated and processed for RNA sequencing and quantitative real time polymerase chain reaction (qRT-PCR) approaches. Neither analysis of biochemical blood parameter-related to liver health, nor histopathological analysis of liver tissue revealed adverse influences of daily average glyphosate exposure of 122 μg/kg BW/d in GLY groups on dairy cows. In contrast, high CFP alone or in combination with time influenced most analyzed blood parameters leading to elevated enzyme activities as well as slightly elevated histopathological scores. Regarding hepatic transcriptome, on the one hand, glyphosate-responsiveness was detected for seven differently expressed genes (DEGs), while only two of them could be validated by qRT-PCR additionally. This fact, in combination with weak fold changes, low read counts, and low DEG number compared to underlying genome size, indicate a false-positive detection of glyphosate-responsive DEGs. On the other hand, 167 CFP-dependent DEGs were observed (CONHC vs. CONLC: 48 induced genes, 32 repressed genes; GLYHC vs. GLYLC: 56 induced genes, 30 repressed genes). Of these 167 DEGs, 21 genes were enriched in the metabolic or signaling pathways “carbon metabolism”, “complement and coagulation cascade, “metabolism of xenobiotics by cytochrome P450”, and “chemical carcinogenesis”. Additionally, expression data of CFP-responsive DEGs showed correlations with performance and blood data of dairy cows, as well as with feed and nutrient intake. In summary, results did not indicate adverse effects of glyphosate residues in feedstuff resulting from formerly realistic exposure conditions on liver health of dairy cows, while results of different CFP over time indicated enhanced hepatic metabolic activity upon feeding high CFP.
After terminating feeding experimental rations and a depletion period of 58 up to 158 days (mean±SE: 107±4), blood samples were taken from 39 cows after parturition. Moreover, blood of newborn calves was collected before their first colostrum intake and calves were examined for evidence of teratogenic effects of in utero glyphosate exposure like, anatomical malformations or physiological misfunction. For a few parameters regarding blood metabolites, hematology, functional properties of blood cells, or gene expression in leukocytes of dams and calves, weak effects of glyphosate, CFP, or their interactions were observed. Putative type I statistical errors as consequence of conducting large numbers of tests must be considered for this study. Furthermore, levels of most putatively affected parameters were comparable to ranges already observed in other animals kept at the same research institute. Importantly, no evidence for drastic teratogenic effects of glyphosate, like malformations in newborn calves, could be detected after parturition. For dams, putative effects of glyphosate or CFP were not consistent between early gestation as analyzed during feeding experimental rations and parturition, indicating physiological challenges and adaptions, which are highly dynamic during this period. However, to finally exclude putative teratogenic or persistent glyphosate effects on health of dams and their newborn calves, glyphosate exposure studies with larger number of dams during entire and/or late gestation and consequently without depletion period would be needed.
Performance data of 47 growing bulls were collected continuously (dry matter, water intake) and weekly (body weight) throughout the trial, while data of blood samples were taken at the beginning of the feeding experiment, after 7, and after 15 weeks. At slaughtering, organ weights were measured and urine samples were collected for glyphosate residue detection. Upon average daily exposures of 128.6 (GLYHC), 213.7 (GLYLC), 1.3 (CONHC), and 2.0 μg glyphosate/kg body weight/d (CONLC), many parameters related to performance, or slaughtering characteristics, like average daily gain or relative liver weights were unaffected. For some hematological traits, functional properties of blood cells, proportions of leukocyte phenotypes, or chemical-clinical traits putative effects of glyphosate alone or in combination with CFP and/or time were observed. These effects and interactions, however, were mostly not consistent between parameters, rather weak and thus, probably not reproducible. Again, CFP and time influenced many assessed parameters markedly, like higher dry matter intake and relative liver weights in HC groups compared to LC groups. Finally, no animals showed any conspicuous clinical symptoms throughout the trial. However, lifetime exposure studies with a larger number of animals and sample collections would further extend the knowledge about putative adverse effects of glyphosate on growing bulls.
When comparing individual studies, effects of glyphosate occurred in inconstant matters. Different results between studies might be explained by different physiological focuses, like ongoing lactation (dairy cows), parturition with related physiological changes (dams, newborn calves), and the fattening process (bulls). In general, high individual variation was observed for data of bulls, like for glyphosate concentration in urine and especially for data of newborn calves in particular. Finally, glyphosate residues in feedstuffs resulting from formerly common agricultural practices in Germany did not adversely affect assessed parameters related to liver health of dairy cow and parameters related to performance and slaughtering, or many blood parameters of growing bulls throughout fattening phase. In newborn calves and dams, no evidence for drastic teratogenic, other clear or persistent effects of glyphosate on analyzed parameters could be observed following glyphosate exposure, depletion period, and parturition.