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

A4.7 Dermal carcinogenicity of a glyphosate-based herbicide in mice

Download 0.79 Mb.
Size0.79 Mb.
1   ...   11   12   13   14   15   16   17   18   19

A4.7 Dermal carcinogenicity of a glyphosate-based herbicide in mice

George et al (2010): Carcinogenicity bioassay: The biological activity of Roundup Original* (a commercial formulation containing 360 g/L glyphosate acid equivalent as the isopropylamine salt, with 15% POEA; no other components identified; manufactured by Monsanto Co., St Louis, MO USA) was tested in a mouse two-stage initiation / promotion model of dermal carcinogenesis. Eight groups of 20 male Swiss mice (12 – 15 g initial bodyweight; from the Indian Institute of Toxicology Research breeding colony) were treated according to the following scheme:


Treatment protocol




Roundup*, 25 mg/kg bw, 3X/wk for 32 wk


DMBA, 52 µg/mouse, single dose then TPA, 5µg/mouse, 3X/wk for 32 wk


Roundup, 25 mg/kg bw, single dose then TPA, 5µg/mouse, 3X/wk for 32 wk


Roundup, 25 mg/kg bw, 3X/week for 3 wk then TPA, 5µg/mouse, 3X/wk for 32 wk


DMBA, 52 µg/mouse, single dose


TPA, 5µg/mouse, 3X/wk for 32 wk


DMBA 52 µg/mouse single dose then Roundup, 25 mg/kg bw, 3X/wk for 32 wk

*The study authors include Roundup Original, but not pure glyphosate, in the list of experimental materials. They state that mice were treated with “glyphosate 25 mg/kg bw”. It is unclear whether they mean “Roundup at 25 mg/kg bw” [in which case the dose of glyphosate would be 9 mg/kg bw], or “sufficient Roundup to deliver a glyphosate dose of 25 mg/kg bw”. I have assumed the former, and use the name “Roundup” to preserve the distinction between the active constituent and commercial formulations bearing this trade name.

DMBA = 7,12-dimethyl benz[a]anthracene

TPA = 12-O-tetradecanoyl-phorbol13-acetate

The initiator (DMBA), promoter (TPA) and Roundup formulation were applied to the clipped intact dorsal skin. According to the study authors, “Vehicle for glyphosate, DMBA and TPA were 50% ethanol and acetone, respectively” [sic]. Animals were weighed and examined weekly for the presence of tumours. All mice were sacrificed after 32 weeks of treatment.

Proteomic study: Groups of four male mice (which had not been used for the carcinogenicity bioassay) were treated dermally once with Roundup (50 mg/kg bw), DMBA (104 µg/mouse) or TPA (10 µg/mouse). The study authors did not state whether vehicles were used. A further four untreated animals served as controls. At 24 hours post-treatment, mice were sacrificed and skin samples from the treatment sites were excised, homogenised, lysed, sonicated, centrifuged and pooled for each respective group. Proteins in the supernatants were then separated by two-dimensional gel electrophoresis (2-DE), with the first dimension on immobilised pH gradient strips (pH 3 – 10) and the second dimension on polyacrylamide gel. Analysis was performed in triplicate. Protein expression levels were measured using PDQuest software, and protein spots that varied > two-fold from control were identified by matrix-assisted laser desorption / ionisation time-of-flight and liquid chromatography / mass spectrometry. The identity of some proteins was confirmed by immunoblotting.


Carcinogenicity bioassay: All mice survived until scheduled termination. All 20 positive controls (Group 3 animals treated with DMBA and TPA) developed skin tumours (squamous cell papillomas), with some animals bearing multiple tumours (see Table). Skin tumours also developed on eight / 20 mice receiving DMBA and Roundup. Compared with TPA, Roundup induced the formation of fewer (by 85%), smaller tumours, which first appeared after a more prolonged (by 2.5-fold) treatment period. There was no comment on whether the tumours were preceded or accompanied by dermal irritation or other visible abnormalities at test sites. No dermal tumours were observed on mice from Groups 2, 4, 5, 6 or 7. Therefore, Roundup behaved as a tumour promoter in this experimental model, but not as an initiator or complete carcinogen.

Table 4.15: Tumour formation on the skin of treated and control mice






Days until 1st tumour





no. of tumours

Mean no. tumours / mouse

Mean tumour vol (mm3)/TBM^



0 / 20








20 / 20*







DMBA + Roundup

8 / 20*






^TBM = Tumour bearing mice NA = Not applicable

*p<0.05 vs untreated controls (ANOVA)

Proteomic study: As revealed by 2-DE, single doses of Roundup, TPA or DMBA caused a >two-fold increase or decrease in the expression of 22 proteins. Expression levels of 13 of these proteins were said to be affected similarly by Roundup and TPA, but quantitative data were provided for only nine of these (see Table). DMBA up-regulated four of this sub-set of proteins similarly to Roundup and TPA, but had little or no effect on the expression of superoxide dismutase 1 (see Table). Use of the Western blotting technique confirmed that Roundup and TPA both up-regulated calcyclin and calgranulin-B by ca three- and four-fold, respectively, and down-regulated superoxide dismutase by about ten-fold (all p<0.05 vs control). Western blotting also demonstrated that DMBA did not influence the expression levels of these particular proteins.

Table 4.16: Expression levels of skin proteins in mice


Difference from untreated control




Translation elongation factor eEF1A1




Carbonic anhydrase III








Annexin II




Fab fragment of anti-VEGF antibody








Superoxide dismutase 1




Stefin A3




Calgranulin-B (two “spots” corresponding to the same protein)







*p<0.05 vs control ND = Not detected using 2-DE


The study authors concluded that glyphosate is a tumour promoter in mouse skin which, based on the similarities in protein expression profile, has a mechanism similar to TPA. They noted that several of the proteins whose activity levels were up-regulated have biologically significant roles in cell proliferation20, while suggesting that down-regulation of superoxide dismutase (which protects cells against reactive oxygen intermediates) could potentiate tumour formation.


In the reviewing toxicologist’s opinion, the carcinogenicity bioassay was not controlled adequately. The test compound was a mixture containing glyphosate, POEA and possibly other adjuvants, and yet no animals were treated with glyphosate, POEA or other formulation constituents in isolation. Therefore, the study could not identify which formulation constituent(s) promoted the growth of tumours in Group 8, show that tumour promotion was caused by any single chemical, or exclude the possibility that promoting activity arose from an interaction between two or more formulation components.

The study reporting would have been strengthened if the authors had commented on whether Roundup Original caused irritation or other effects on the skin where it was applied. This would have been of particular interest because POEA is a severe dermal irritant (Birch, 1977), consistent with the properties of surfactants in general, which interact with and solubilise lipid components of the skin and mucous membranes (Williams et al, 2000). The presence or absence of dermal responses such as inflammation, de-fatting, cell proliferation, scabbing, scarring or fissuring could have assisted in identifying the mechanism(s) by which Roundup promoted the formation of tumours. In this context, it is notable that POEA is not a mutagen (Stegeman and Li, 1990; Williams et al, 2000).

Another point deserving comment is that the proteomic analysis was carried out only at 24 hours after a single application of DMBA, Roundup or TPA. This is fundamentally different from the carcinogenicity bioassay, which involved repeated dosing over 32 weeks after DMBA application. No analysis was performed on skin from test sites during or at the end of the treatment period, on the tumours themselves, or on skin that had been treated with both DMBA and Roundup or TPA. The study did not demonstrate that the changes in protein expression observed after one dose of DMBA, TPA or Roundup were sustained throughout the experimental period, were a toxicological endpoint rather than homeostatic regulation, or were causally associated with the eventual development of tumours. Furthermore, the study could not detect changes in the expression of additional proteins after repeated treatment. Consequently, it is uncertain that the promoting activity that the study authors attributed to glyphosate is mechanistically similar to that of TPA.

Overall, this study has shown that Roundup Original is a tumour promoter on mouse skin, its activity is weaker than that of the positive control, TPA, and is dependent on prior induction with the initiator DMBA. The causative agent(s) and its (or their) mode of action remain unidentified. Given that Roundup Original is not a complete carcinogen, it is unlikely to pose a carcinogenic hazard for persons exposed dermally.

Share with your friends:
1   ...   11   12   13   14   15   16   17   18   19

The database is protected by copyright © 2020
send message

    Main page