Review of the earth open source (eos) report " roundup and birth defects: is the public being kept in the dark?"

Download 0.79 Mb.
Date conversion21.02.2016
Size0.79 Mb.
1   ...   9   10   11   12   13   14   15   16   ...   19


A4.1 Effects of a glyphosate-based herbicide formulation on gene expression in vitro

Hokanson et al (2007): In a study of the effects of glyphosate on the expression of oestrogen-regulated genes, MCF-7 human breast adenocarcinoma (oestrogen sensitive) cells were exposed to an unidentified home garden herbicide containing 15% glyphosate (no additional details provided) with or without 3.0 X 10 M 17β-estradiol (oestrogen). Cells were incubated for 18 hours with the herbicide at final glyphosate concentrations of 0.23, 0.023, 0.0023, or 0.00023%. Following purification of cellular RNA and generation of cyanine 3- and 5-labelled anti-sense RNA, the activity of 1550 genes was then measured by DNA microarray analysis using RZPD chips.

According to the study authors, 680 of the 1550 investigated genes were dysregulated by exposure to the herbicide. However, they did not state by how much the affected genes’ activity differed from control levels, or at what glyphosate concentrations. The study authors listed a sub-set of 29 genes whose activities were up- or down-regulated by greater than 2-fold, of which seven were tested further by quantitative real-time PCR to corroborate the results of DNA microarray analysis.

Only three of the 1550 genes fulfilled the criteria for significant dysregulation, when appraised by both methods. In the presence of glyphosate at 0.00023%, DNA microarray analysis indicated that HIF1 was up-regulated by 2.2-fold, while CXCL12 and EGR1 were down-regulated to 0.46 and 0.49 of control activity. qrtPCR expression analysis showed that HIF1 was up-regulated by over two-fold whereas CXCL12 and EGR were down-regulated by over 50%. For each gene, cell treatment with oestrogen alone yielded expression levels that were intermediate between those observed in control cells and cells exposed to oestrogen and herbicide combined.

According to the study authors, the HIF1 gene primes cells for the initiation of apoptosis under hypoxic conditions, and therefore plays a key role in cell death resulting from cerebral and myocardial ischemia. They raise the possibility that elevated levels of HIF1 [protein] may initiate apoptosis in the absence of hypoxia, promoting a variety of hypoxia-initiated patho-physiological states including ischemia of the myocardium, brain and retina; pulmonary hypertension, pre-eclampsia and intrauterine [foetal] growth retardation.

The CXCL12 gene product (also known as stromal cell-derived factor 1 and pre-beta cell growth-stimulating factor) is a lymphocyte chemoattractant, may be involved in lymphocyte activation, and is reportedly critical for the mobilisation of cells of the haematopoietic tissues into peripheral blood. Hokanson et al suggest that altered [decreased] levels of CXCL12 may contribute to disruption of immune surveillance and basal extravasation of mono- and lymphocytes.

Among the biological effects attributed to EGR1 are regulating the expression of transforming growth factor beta-1, involvement in the suppression of [cellular] growth and transformation, and the regulation of apoptosis, endothelial cell growth, neovasculatisation, tumour initiated angiogenesis and tumour growth. The study authors consider that [decreased] levels of EGR1 may potentially affect the rate of initiation of apoptosis and alter the level of vascularisation associated with tumour formation.


This paper is of limited value: it does not identify which components of the glyphosate-based herbicide formulation are responsible for altering gene expression, does not identify any mode of action of those components, does not provide evidence that the observed changes in gene expression are anything other than homeostatic regulation, and does not establish that the effects observed in MCF-7 cancer cells in vitro would be representative of those that would occur in non-cancerous mammalian cells (especially within tissues or at the whole animal level). Other than retardation of foetal growth, the postulated effects of HIF1, CXCL12 and EGR dysregulation have not been reported in toxicology studies in laboratory animals, and there appears to be no justification for extrapolating from the study’s findings to predicting adverse effects on human health.

Mink et al (2011) have reviewed epidemiological studies relevant to some of the non-cancer end-points that Hokanson et al speculate may be affected. In the study populations, there was no statistically and/or biologically association between exposure to glyphosate and retinal degeneration (Kirrane et al, 2005), myocardial infarction (Dayton et al, 2010 and Mills et al, 2009) or depressed birthweight (Sathyanarayana et al, 2010). Furthermore, epidemiological evidence of associations between glyphosate exposure and cancer is weak and conflicting (DoHA, 2005). A recent review (Mink et al, 2012) of epidemiological studies relevant to cancer end-points considered seven cohort studies and fourteen case-control studies looking at possible associations between glyphosate and one or more cancer outcomes; there was no consistent pattern of positive associations to indicate any causal relationship between total cancer (in adults or children) or any site-specific cancer and exposure to glyphosate.

A4.2 Cytotoxicity of glyphosate, AMPA and glyphosate-based herbicides in vitro

Benachour et al (2007): Human embryonic kidney (HEK) 293 and human choriocarcinoma-derived placental JEG3 cells were exposed for 1 – 72 hours in vitro to Roundup Bioforce (360 g/L glyphosate acid present as 480 g/L glyphosate isopropylamine salt, no other constituents identified; Monsanto, Anvers, Belgium) at up to 2% in the incubation medium, or glyphosate at equivalent concentrations (up to 42 mM). Cell viability was measured by the MTT assay, based on the cleavage of MTT by the mitochondrial enzyme succinate dehydrogenase (SDH). When the effects of the test formulation and glyphosate were compared, glyphosate solutions were adjusted to ca pH 5.8, the pH of a 2% Roundup solution.

Roundup Bioforce showed greater concentration- and time-dependent cytotoxicity against both cell lines than glyphosate at equivalent concentrations, suggesting that adjuvants in the formulation were contributing to cellular injury. JEG3 cells were more resistant to Roundup Bioforce than HEK293 cells, but both types were of similar susceptibility to glyphosate.

Table 4.1: EC50s* (% in serum-containing medium) of Roundup Bioforce and equivalent concentrations of glyphosate for viability of HEK293 and JEG3 cells.

Test compound

1 h

24 h

48 h

72 h

HEK293 cell line

Roundup Bioforce










JEG3 cell line

Roundup Bioforce










*EC50 (not the LD50 as claimed by the study authors-10) = the concentration required to cause a 50% decrease in mitochondrial SDH activity. As the data were provided in graph form, all values are approximate.

Effects of Roundup Bioforce and glyphosate on the activity of aromatase (CYP19; an enzyme catalysing the conversion of androgens to oestrogens) were measured in HEK293 cells transfected with human aromatase cDNA, human placental cell microsomes and equine testicular microsomes. The HEK293 cells were exposed to the test compounds for 24 hours at up to 0.2% Roundup or 1% glyphosate, while microsomes had a 15-minute exposure period at up to 10% Roundup or 2% glyphosate. The assay quantified the release of tritiated water from [1β-H]-androstenedione.

Both the formulation and active constituent weakly inhibited aromatase activity in vitro. Under pH-adjusted conditions at 37 3oC, glyphosate had IC50s of ca 1.0% and 0.8% against aromatase in placental microsomes and HEK293 cells, respectively. Over its tested concentration range (0.01 – 0.2%), Roundup Bioforce inhibited aromatase by ca 20% in HEK293 cells. Roundup Bioforce had an IC50 of ca 4% against aromatase activity in human placental and equine testis microsomes, at 25 oC and physiological pH.


The concentrations of Roundup and glyphosate required for cytotoxicity and aromatase inhibition were similar to those present in herbicidal spray mixtures (1 – 2% formulation or 21 – 42 mM glyphosate), orders of magnitude higher than would be attained within cells or tissues in vivo under physiological conditions. Over the more biologically relevant concentration range 0.001 – 100 µM, glyphosate has no effect on steroid hormone production in the H295R steroidogenesis assay, developed by the OECD as an in vitro screening assay for endocrine disrupting chemicals (Hecker et al, 2010). Given that surfactants inhibit aromatase activity by disrupting mitochondrial membranes (Levine et al, 2007), the reported effects of Roundup Bioforce in HEK293 cells and microsomes are likely to be experimental artefacts. Another confounding factor would have been the pH of the incubation medium, which was below the physiological range during the cell viability assays.

Benachour and Seralini (2009) evaluated the in vitro cytotoxicity of glyphosate (Sigma-Aldrich), the glyphosate metabolite AMPA (Sigma-Aldrich), four glyphosate-based herbicide products (see table below), and the surfactant polyethoxylated tallow amine (POEA; a component of some glyphosate formulations) to human umbilical cord vein endothelial cells (HUVEC)14 and the human choriocarcinoma-derived placental (JEG3) and human embryonic kidney (HEK293) cell lines.

Table 4.2: Glyphosate-based herbicides studied in Benachour & Seralini (2009)

All products were manufactured by Monsanto, Anvers, Belgium

Product Name

(Abbreviation used in evaluation)

Glyphosate concentration (g/L)

Roundup Express (R7.2)


Roundup Bioforce* (R360)

Roundup Extra 360*


Roundup Grands Travaux (R400)


Roundup Grands Travaux Plus (R450)


*The study authors treated both products as being the same formulation.

No further information on product composition was provided.

Cells were exposed for 24 hours in serum-free medium to each individual test compound at 14 concentrations ranging from 10 ppm to 20 000 ppm (0.001% to 2%). Cells were also exposed to POEA at 1 and 5 ppm, and AMPA at 4, 6, 8 and 10%. Using sub-toxic concentrations of glyphosate, AMPA and POEA, evidence of additive or synergistic toxicity was sought in HEK293 and JEG3 cells exposed to combinations of POEA 1 ppm + glyphosate or AMPA 5000 ppm, and glyphosate 4000 ppm + AMPA 1000 ppm. HUVEC cells were exposed to POEA 1 ppm + glyphosate or AMPA 500 ppm, and glyphosate 400 ppm + AMPA 100 ppm.

After incubation, cytotoxicity was assessed by the following criteria: Adenylate kinase (AK) activity in the incubation medium, as a biomarker of cytoplasmic membrane rupture (assumed to result from cellular necrosis, either primary or secondary after apoptosis); Intracellular succinate dehydrogenase (SDH) activity, assayed by the MTT test as a measure of mitochondrial respiration rate; and Intracellular caspase 3/7 activity, as indicators of apoptosis. Results from the cytotoxicity assays were presented in graphical form alone, and therefore only approximate quantitative values are available.


Cytotoxicity, assessed by impact on mitochondrial respiration rate: In all three cell types, the concentration of glyphosate causing a 50% decrease in SDH activity (ie, the EC50, and not the LD50 as claimed by the study authors15) was ca 10 000 ppm. The metabolite AMPA was markedly less toxic, having EC50s of ca 40 000, 100 000 and >100 000 ppm in HEK293, JEG3 and HUVEC cells, respectively. By contrast, POEA was highly cytotoxic, demonstrating a lowest EC50 of ca 3 ppm (see following table). All Roundup formulations were more toxic than the active constituent. Moreover, their EC50s were not linearly proportional to the concentration of glyphosate in the products or incubation medium. This is consistent with other formulation components being cytotoxic and/or potentiating the toxicity of the active constituent.

Table 4.3: Concentrations of glyphosate and other test compounds causing a 50% decrease in intracellular succinate dehydrogenase activity in HUVEC, JEG3 and HEK293 cells

Test compound

Approx EC50


Glyphosate concentration (ppm)

in medium at the EC50


10 000

10 000


>40 000



3 – 30


R 7.2

6000 – 9000

42 – 63


2000 – 3000

720 – 1080







Cell membrane integrity: AMPA, POEA and the Roundup formulations caused increases in extracellular AK activity, consistent with leakage or rupture of cell membranes. By contrast, cells exposed to glyphosate alone released little or no AK, even in the presence of marked depression in mitochondrial respiration. The study authors interpreted this as evidence that glyphosate does not mediate cell death by necrosis, in contrast to AMPA, POEA and Roundup formulations.

Interactions between glyphosate, AMPA and POEA, assessed by effects on cell membrane integrity: Combinations of glyphosate + POEA, glyphosate + AMPA and AMPA + POEA (see above) were clearly more cytotoxic to HUVEC and HEK293 cells than the individual chemicals, causing about 2-fold and 4 to 8-fold more extensive release of AK from the two respective cell types. However, for reasons unknown, additive or synergistic toxicity was not observed in JEG3 cells.

Apoptosis: At incubation concentrations of 50 ppm and above, glyphosate and R360 induced transient but marked increases in intracellular caspase 3/7 activity within HUVEC cells. The effect was first observed after 6 hours of exposure. After 12 hours, caspase activity peaked at 20 – 30 times control levels. Reversibility was well advanced by 18 hours and complete at 24 hours. Similar but much weaker responses occurred in HEK293 and JEG3 cells, within which caspase 3/7 activity increased by no more than 2 or 3-fold. These cell lines were markedly less sensitive than HUVEC cells, requiring glyphosate and R360 concentrations of at least ca 10 000 and 1000 ppm, respectively, for induction of caspase activity. Cell death, loss of adhesion, shrinkage and fragmentation were confirmed microscopically in all cell types after 24 hours exposure to 50 ppm R400. DAPI staining revealed DNA condensation in HUVEC, HAK293 and JEG3 cells exposed to glyphosate or R360 at 5000 ppm.

No findings were presented on the influence of AMPA and POEA on caspase activity or cell morphology.


The French Agency for Food Safety (AFSSA, 2009) has reviewed Benachour and Seralini (2009), commenting that:

  • During exposure to the test compounds, cells were incubated for 24 hours in medium without serum, which could lead to disturbance of their physiological state.

  • The glyphosate tested in the study was glyphosate acid, whereas glyphosate isopropylamine salt was present in the commercial formulations tested. No precise information regarding pH was given, except at the highest concentrations [where the pH was adjusted to 5.8].

  • Cytotoxicity and induction of apoptosis may have been due to pH and / or variations in osmotic pressure at the highest concentrations tested.

  • Surfactant effects and increased osmolality are known to increase membrane permeability, causing cytotoxicity and induction of apoptosis.

  • The test cells were exposed at extremely high concentrations of the test compounds under physiologically abnormal conditions.
1   ...   9   10   11   12   13   14   15   16   ...   19

The database is protected by copyright © 2016
send message

    Main page