Changes in Rat Mammary Tissue Architecture Following Pregnancy/Lactation Exposure to Glyphosate Alone or with 2,4-D and Dicamba

Abstract

The current study aimed to assess the possible endocrine disruptor effects on rat mammary tissue and reproductive organs during pregnancy and lactation when exposed to low doses of glyphosate and its combination with 2,4-dichlorophenoxyacetic acid (2,4-D) and dicamba. The study involved the exposure of pregnant Wistar rats to various regulatory-relevant doses of glyphosate, ranging from gestational day 6 until fine of the lactation period. Glyphosate doses corresponded to the European Union’s glyphosate-acceptable daily intake (ADI; 0.5mg/kg bw/day) and no observed adverse effect level (NOAEL; 50mg/kg bw/day). The dose of the mixture of glyphosate, dicamba, and 2,4-D was at the European Union ADI for each herbicide namely 0.5, 0.002, and 0.3mg/kg bw/day, respectively. In the animals exposed to glyphosate NOAEL serum estradiol levels were increased compared to untreated animals, along with an upregulation of TNF-?, MMP-2, and MMP-9 as measured in mammary gland homogenates compared to non-treated animals. Moreover, in this group, a focally acute inflammatory infiltrate was observed in the mammary gland. Our study showed that short-term exposure to glyphosate at doses that are set as safe by regulators and thus without risk corroborated with a particular physiological state as gestation and lactation, can give rise to inflammatory changes in breast tissue in rats. These findings support the need for further evaluation of glyphosate and mixtures of glyphosate with other pesticides for public health protection, especially for those categories vulnerable to the potential endocrine disruptor properties of these pesticides such as pregnant women, newborns, and children.

Introduction

Human exposure to pesticides, including those for domestic use, is currently a major public health concern, as an increasing number of studies highlight a link between long-term exposure to these substances and the pathogenesis of many chronic diseases, including neurological [1, 2], metabolic (including obesity) [3], cardiovascular [4], immune [5] and reproductive disorders [6].

Furthermore, they are also implicated in the risk of developing various cancers [7, 8], owing to these substances' capacity to cause oxidative stress [9], genotoxicity [10], and epigenetic changes at the level of various genes [11, 12].

Numerous plant protection products have been recognized to possess an endocrine disrupting chemical (EDC) capability. EDCs mainly disrupt natural hormones through their high ability to attach to androgen or estrogen receptors [13].

In addition, EDCs are capable of binding to and activating various hormone receptors (ARs, ERs, AhRs, PXRs, ERRs) and subsequently mimicking the effects of natural hormones (agonistic effects).

EDCs are also able to bind to these receptors without activating them and thus acting as antagonists resulting in an inhibition of their activity [13].

Finally, EDCs can reduce the concentration of natural hormones by interfering with their synthesis, transport, metabolism and excretion [14].

Monsanto introduced glyphosate-based herbicide (GBH) to the market in 1974 under the brand name Roundup®.

It is a non-selective, broad-spectrum herbicide that works by preventing the synthesis of vital aromatic amino acids by inhibiting the shikimate pathway's 5-enolpyruvylshikimate-3-phosphate synthase.

Genetically modified (GM) glyphosate-tolerant crops were introduced and quickly adopted in North and South America beginning in 1996, which resulted in a significant rise in GBH.

These herbicides are currently the most widely used worldwide [15].

Concerns have been raised recently regarding the safety of glyphosate and commercial GBH formulations, which has been the focus of much debate due to its widespread distribution in the environment and disagreements on its toxicity [16].

One area of conflicting evidence and interpretation of glyphosate (GLY) toxicology has centered the possible of this compound acting as an EDC [17, 18].

Regulators in the US [19] and Europe [20] believe that there is in sufficient data to substantiate glyphosate-induced endocrine disruption.

These findings are supported by long-term in vivo toxicity studies as well as a number of in vitro tissue culture-based investigations as well as short-term screening in vivo looking at possible interactions with the thyroid, estrogen, and androgen pathways or interfering with steroid hormone synthesis.

On the other hand, glyphosate and GBHs may have different endocrine and reproductive effects in different model systems, as demonstrated by a number of studies, most of which concluded that GBH was more harmful than glyphosate alone.

The discussion around the endocrine disruptive properties of glyphosate and other herbicides is crucial in the context of chemical regulation, since several nations and territories such as the European Union (EU) and Brazil, have adopted a risk-based approach to endocrine disrupters in pesticide products [21].

The effects of combining GLY with other herbicides on health are not well understood.

This is becoming of particular concern alongside the widespread use of GM crops that are tolerant to combinations of herbicides, in particular, GLY together with dicamba and 2,4-dichlorophenoxyacetic acid (2,4-D).

Surveys have already shown a very large increase in exposure to a combination of these pesticides, which is anticipated to escalate even further [22].

Considering this and the new toxicological trend related to the evaluation of potential threats determined by exposure to chemical combinations using real-life risk simulation [23, 24, 25, 26], in this study we aimed to evaluate the effects of low, environmentally and regulatory relevant doses of glyphosate and its combination with 2,4-D and dicamba on reproductive organs, sex hormone and breast biology of female rats exposed during pregnancy and lactation.

We previously showed that GLY caused a dose-dependent increase in thyroid hormone levels and malfunction of kidney tubules and glomeruli, while its mixture with dicamba and 2,4-D resembled the same effects produced by GLY alone at a dose 100 times higher, indicating a synergistic effect from exposure to the three herbicides [27].

Materials and Methods

Animal experiment

Wistar rats of 3-months of age were used.

The animals were given filtered tap water and standard rat chow ad libitum and housed at a regulated temperature of 21±2°C and humidity of 50±10% for a two weeks acclimatization period previous to study conditions.

After acclimatization, the female rats were mated.

On gestation day (GD) 1, they were separated from males into cages with 2 or 3 animals per cage.

The animals were randomized, starting on GD 6, to one of four experimental groups (n=5 per group).

From GD 6 to the weaning date [postnatal day (PND) 28], these groups were administered the following treatments in drinking water:

-Untreated group: animals received water and food ad libitum.

-EU glyphosate ADI dose group: 0.5mg/kg bw/day glyphosate [28].

-EU glyphosate NOAEL dose group: 50mg/kg bw/day glyphosate [28].

-Mixture group: 0.5mg/kg bw/day glyphosate, 0.02mg/kg bw/day 2,4-D [29] and 0.3mg/kg bw/day dicamba [30], which corresponds to current EU ADI values.

This study's detailed methodology has been previously published [31].

The animal study was approved by the Ethical Committee of the UMF of Craiova, Romania,No. 120/19.11.2020.

On PND28 the animals were sacrificed by exsanguination after previous anaesthesia with anesthetized a combination of ketamine and xylazine [31].

The serum was separated and kept at-70°C until needed for further measurements.

Assessment of serum hormone concentrations

ELISAs was used to measure the levels of serum progesterone, estrogen and testosterone in accordance with the manufacturer's instructions (DRG Instruments GmbH, Marburg, Germany).

Testosterone (cat no. EIS-1559), progesterone (cat no. EIA-1561), and estradiol (cat no. EIA-2693) kits were used.

Histopathological evaluation of the ovary, salpinx, uterus and mammary gland

The ovary, salpinx, uterus and mammary gland of each dam were collected, cleaned with physiological saline, and dried with clean paper before weighing.

The tissues were partially preserved for a full day in a solution of paraformaldehyde 4% and then dried via a series of ethanol solutions with escalating concentration.

Following dehydration, the tissues were embedded in paraffin and fixed in xylene for two hours.

A microtome was used to cut paraffin blocks into 25µm sections, which were then stained with hematoxylin/eosin.

A Panthera L light microscope (Motic Europe, S.L.U.), was used to examine the slides.

Histopathological alterations were graded based on the severity of the changes and the quantity of animals impacted.

The following damage intensity ratings were assigned: (-) non-existent, (+) mild, (++) moderate, and (+++) severe.

ELISA determination of tumor necrosis factor (TNF-α), matrix metalloproteinase-2 (MMP-2), and matrix metalloproteinase-9 (MMP-9) measurement in mammary gland homogenates

The levels of MMP-2 and TNF-α in mammary gland homogenates were determined using rat MMP-2 (cat no. CSB-E07411r, Cusabio, Wuhan, China) and rat TNF-α (cat no. CSB-E11987r, Cusabio, Wuhan, China) ELISA kits.

The quantities of MMP-9 in mammary gland homogenates were measured using the MMP-9 ELISA kit (cat no. ABIN6730943, antibodies-online GmbH, Germany).

Each mammary gland sample was weighed, homogenized on ice (100mg tissue to 1mL PBS), and stored at-20°C overnight.

Following two rounds of freezing and thawing to rupture cell membranes, the homogenates were centrifuged for 5 minutes (5000 x g at 2-5°C) and tissue homogenate supernatant frozen.

Statistical analysis

SPSS 20 (SPSS Incorporated, Chicago, USA) was used to statistically analyze the data.

For continuous data, we calculated the mean±standard deviation of the mean.

Dunnett's test was used in combination with a one-way ANOVA test for multiple comparisons.

A p value of <0.05 was taken as statistical significance.

Results

Determination of serum sex hormone levels

The exposure to EU ADI doses of dicamba, 2,4-D and glyphosate resulted in an increase of progesterone levels at 73.8±2.59ng/mL compared to animals from the un-treated group, 68.00±3.08ng/mL (p<0.05) (Table 1).

Table 1.

Serum estradiol, progesterone, and testosterone levels in dams at the end of the exposure period

Parameter	Untreated group	EU glyphosate ADI dose group	EU glyphosate NOAEL dose group	Mixture group
Progesterone (ng/mL)	68.00±3.08	70.60±1.95	69.60±2.41	73.80±2.59*
Estradiol (pg/mL)	37.60±1.95	38.60±1.52	47.20±2.59**	43.40±3.65
Testosterone (ng/mL)	0.53±0.07	0.44±0.08	0.64±0.01	0.49±0.04

Note: *p<0.05 vs. untreated group, **p<0.01 vs. untreated group

In addition, animals exposed to the glyphosate NOAEL dose, estradiol levels were increased at 47.20±2.59pg/mLcompared to untreated rats, 37.60±1.95pg/mL (p<0.01) (Table 1).

In contrast, no change in testosterone levels were observed in any treatment group (Table 1.

Histopathological evaluation of the ovary, salpinx, uterus and mammary gland

Histopathological analysis of the ovary was performed for all groups of animals with normal architecture observed including the presence of ovarian follicles in different stages of maturation-early primary, primary, secondary and dominant follicles in all the groups (Figure 1).

Figure 1.

Ovary, HE staining, X200. A-Control group: ovarian follicles in different stages of evolution-blue arrow; B-Adi group: corpus luteus-blue arrow; C-Noael group: ovarian follicles in different stages of evolution-blue arrow; D-Mixt group: ovarian follicles in different stages of evolution-blue arrow, vascular hyperemia-green arrow

We noticed marked vascular hyperemia, with vessels dilated (Figure 1).

Another normal aspect such as the presence of corpus luteus was present in the non-treated and the herbicide mixture groups (Figure 1A and 1D).

Attached to the ovary were fallopian tubes with pathological aspects observed being the presence of hyperplastic mucosa, especially in the untreated group (Figure 2A), or aspects of atrophy and hypertrophy involving the mucosa, especially in the herbicide mixture group (Figure 2D).

Figure 2.

Salpinx, HE staining, X200. A-Control group; B-Adi group; C-Noael group: hyperplastic mucosa-blue arrow; D-Mixt group: atrophy and hypertrophy involving the mucosa-blue arrow

In the muscularis, we identified foci of vascular hyperemia.

The described aspects are found in the specimens from all the groups (Figure 2).

Histopathology highlighted various structures in the uterine body, namely the endometrium (Figure 3).

Figure 3.

Uterine body/endometrium, myometrium, HE staining, X200. A-Control group: secretory-type mucosa-blue arrow, poorly represented vascular differentiation-green arrow; B-Adi group: simple cuboidal-cylindrical epithelium-blue arrow, rare endometrial glands-green arrow; C-Noael group: single layered cylindrical epithelium-blue arrow, proliferative phase glands-green arrow; D-Mixt group: vascular differentiation-blue arrow.

Secretory-type mucosa, with vacuolar and weakly eosinophilic cytoplasm, tendency to stratification, focally with micropolyps appearance, with rare glands of secretory appearance, poorly represented vascular differentiation that alternates with areas with glands of proliferation with marked vascular differentiation, were observed in the untreated control group (Figure 3A, Table 2).

Table 2.

Grade and frequency of histological abnormalities in uterus/endometrium, and myometrium in the glyphosate and herbicide mixture treated groups according to histological evaluation

Parameters	Experimental groups
	Untreated group		EU glyphosate-ADI dose group		EU glyphosate-NOAEL dose group		Mixture group
	Intensity	Number of specimens	Intensity	Number of specimens	Intensity	Number of specimens	Intensity	Number of specimens
Proliferative/ secretory endometrium	++	4/5	+/-	1/5	++	4/5	++	4/5
Vascular differentiation (endometrium/ myometrium)	+++	5/5	++	4/5	++	5/5	+++	5/5

: Note: (-) non-existent, (+) mild, (++) moderate, (+++) intense

For the EU glyphosate ADI dose group, we noticed the presence of simple cuboidal-cylindrical epithelium and rare endometrial glands but with vascular differentiation present, especially in the myometrium (Figure 3B, Table 2).

A similar morphology with single layered cylindrical epithelium, focally with signs of micropolyps were identified within the EU gluphosate NOAEL dose group (Figure 3C, Table 2).

The proliferative phase was also noticed for the endometrial glands, and regarding vascular differentiation was marked, both in the endometrium and myometrium (Figure 3C, Table 2).

In the case of the herbicide mixture group, we observed hypertrophy within the mucosa (Figure 3D, Table 2).

We also found two other alterations in the endometrial glands, namely a proliferative influence that alternated with glands with secretory changes, and with vacuolar cytoplasm.

Vascular differentiation was also found in this group and was intense in both the endometrium and the myometrium (Figure 3D, Table 2).

Based on the severity of the alterations in histopathology and the quantity of impacted samples, a score was assigned.

The intensity of the changes was graded from non-existent to intense (Table 2).

As a hormone responsive organ, the mammary gland was also investigated.

For the untreated group, we identified specific lactation changes that alternate with normal-looking breast parenchyma, periductal fibrosis and intraluminal eosinophilic secretion (Figure 4A, Table 3).

Figure 4.

Mammary gland, HE staining, X200. A-Control group: glandular lactation changes-blue arrow, intralumenal eosinophilic secretion-green arrow; B-Adi group: atrophic acini-blue arrow; C-Noael group: fibrous stroma-blue arrow, mast cells-green arrow; D-Mixt group: lipomatous infiltration in the breast parenchyma-blue arrow, acinar hyperplasia with dilated acini-green arrow

Table 3.

Grade and frequency of histological abnormalities in mammary gland in the glyphosate and herbicide mixture treated groups according to histological evaluation

Parameters	Experimental groups
	Untreated group		EU glyphosate ADI dose group		EU glyphosate NOAEL dose group		Mixture group
	Intensity	Number of specimens	Intensity	Number of specimens	Intensity	Number of specimens	Intensity	Number of specimens
Lactation changes	+++	5/5	-	0/5	-	0/5	-	0/5
Fibrosis	+/-	2/5	+	2/5	++	3/5	+/-	1/5
Intralumenal eosinophilic secretion	+	3/5	-	0/5	-	0/5	++	3/5
Acute inflammatory infiltrate	-	0/5	-/+	1/5	++	4/5	-/+	1/5

: Note: (-) non-existent, (+) mild, (++) moderate, (+++) intense

For the EU glyphosate ADI dose group, we observed breast parenchyma with fibro-collagenous stroma and focally atrophic acini and rare eosinophilic mast cells (Figure 4B, Table 3).

Similar histopathological changes with predominantly fibrous peri-and intraductal stroma and eosinophilic mast cells, were also detected for the EU glyphosate NOAEL dose group (Figure 4C, Table 3).

Also, in this group, we detected a focally acute inflammatory infiltrate (Figure 4C, Table 3).

In the group of animals exposed to the herbicide mixture, a breast parenchyma with lipomatous infiltration, acinar hyperplasia with dilated acini and intracytoplasmic vacuolization, reduced intraluminal eosinophil secretion, and rare stromal mast cells with reduced periductal fibrosis was observed (Figure 4D, Table 3).

Similar to what was undertaken in the case of the uterus (Table 2), the histopathological changes in the mammary gland were scored based on their intensity and the number of affected samples being graded from non-existent to intense (Table 3).

MMP-2, MMP-9, and TNF-α determination in mammary gland homogenates

Exposure to the EU glyphosate NOAEL dose revealed an increase in TNF-α, MMP-2, and MMP-9 levels in mammary gland homogenates compared to untreated animals (p<0.01) (Table 4).

Table 4.

MMP-2, MMP-9, and TNF-? levels in mammary gland homogenates in the glyphosate and herbicide mixture treated groups after exposure

Parameter	Untreated group	EU glyphosate dose ADI dose group	EU glyphosate NOAEL dose group	Mixture group
TNF-α (pg/mL)	11.80±1.30	12.20±0.84	19.20±1.79**	14.80±1.30
MMP-2 (ng/mL)	11.05±1.39	11.50±0.98	16.27±0.89**	12.89±0.61
MMP-9 (ng/mL)	396.04±7.63	402.76±5.26	431.01±4.88**	409.05±6.39

: Note: **p<0.01 compared with the untreated group

In the EU glyphosate ADI dose and herbicide mixture groups even if these biomarkers did somewhat rise, the difference was not statistically significant.

Discussions

In our study, the ovaries showed normal morphology with follicles of various sizes in different stages of maturation including the presence of a collapsed follicle that suffers reorganization.

Similarly, the fallopian tubes presented normal structural aspects [32].

For the upper portion of the uterus, the body, we noticed the presence of normal changes during the oestrous cycle in both endometrium and myometrium.

The only change was focal micropolyposis that may be attributable to a proliferative stimulus [32, 33].

Regarding breast tissue, most animals presented a normal histology, with the ductal-lobular system embedded in a fibrous and adipose stroma [34], except for the EU glyphosate NOAEL dose group where clear histopathological signs of inflammation were observed.

The family of Zn2+-dependent endopeptidases called matrix metalloproteinases (MMPs) are in charge of breaking down the components of the extracellular matrix [35].

Their structural distinctions lead to their division into many families [35].

MMP-2 and -9 may degrade denatured collagen (gelatin) and collagens IV and V [36].

They facilitate the remodelling of the extracellular matrix by breaking down other extracellular matrix constituents [37], and they influence physiological processes involved in wound healing, differentiation of the cells and tissue repair [38].

Moreover, gelatinases may be involved in carcinogenic processes such as angiogenesis, tumor metastasis, and cell proliferation through their proteolytic activity [39].

Evidence from the literature points to their participation in several pathologic processes, such as angiogenesis, inflammation, cell proliferation, and tumor growth, that are essential for the development of cancer [40].

The invasive and metastatic phases of breast cancer have been specifically associated with the biological activities of MMP-2 and -9 proteins [41], while it is unknown how these proteins contribute to the development of breast cancer.

Previous studies have shown that a chronic state of inflammation may represent a functional tumour microenvironment in breast cancer that is present prior to the development of the malignant state [42].

TNFα, which has several effects on cell malignancy and metabolism, is continuously elevated in the microenvironment of breast tumors [43].

Newer research has shown that TNFα controls breast cell survival and proliferation via altering respiratory chain super complex structure and function, as well as mitochondrial metabolism [44].

In our investigation, we found that mammary gland homogenates had significantly higher levels of MMP-2, MMP-9, and TNF-α in the EU glyphosate NOAEL dose group associated with clear histopathological signs of inflammation that in time could influence the tissue microenvironment and possibly lead to degeneration into pathological tissue.

The role of glyphosate and commercial GBH exposure in breast cancer development uncertain with Glyphosate ability to act as estrogen agonist being controversial [45].

In addition, a greater than 3-fold increase in mammary tumor incidence was observed in rats after being exposed for two years at a very low dosage to a commercial GBH [46].

It’s also critical to remember that commercial GBH formulations are mixtures of GLY with toxic co-formulants [47, 48], with enhanced carcinogenic potential compared to glyphosate alone [49].

There are significant concerns about the long-term effects of chemical pollutants on the emergence of chronic illnesses given the recognition that they may have an impact on the epigenome and consequent global patterns of gene expression.

Since cancer never results from a single risk factor, a synergistic approach might help better understand a possible risk impact for the disease.

Duforestel and colleagues showed that 50% of the mice developed tumors when glyphosate and the breast cancer-associated microRNA (miR) 182-5p were combined.

However, tumor development could be avoided by either blocking miR 182-5p specifically or treating the glyphosate-miR 182-5p-cells with dimethyloxallyl glycine, an inhibitor of the TET pathway.

These results demonstrate that, corroborated with an additional possible risk factor, low-pressure and long-term TET-mediated DNA hypomethylation prepares cells for an oncogenic response [50].

Breast cancer-promoting cytokines (IL-6, IL-4, TNF, and IL-17A) in the mammary gland are all increased by estrogens, which also impair the ability of CD8+T cells and NK cells to fight cancer [51, 52].

The influence of hormonal exposure on the incidence of breast cancer may be partially explained by its immune-modulatory action, since estrogen exposure decreases anti-tumor cytotoxic T cells and boosts suppressive myeloid cells.

Signals from chemokines and mammary epithelial cells exert specific effects on the innate immune cells.

During development, puberty, and the phases of pregnancy, breastfeeding, and involution, the normal mammary gland enlists and stimulates immune cells.

Normal breast development is impacted if certain immune cell populations are restricted [53, 54].

Breast cancer incidence is significantly associated with mammographic breast density and increases following estrogen exposure, which has been linked to direct effects on estrogen receptor (ER)+ mammary epithelium [55].

According to earlier pre-clinical preventive studies conducted on mice, oophorectomy and estrogen receptor antagonists (fulvestrant) both prevent the development of breast cancer [56, 57], while estrogen can stimulate tumorigenesis [58].

Exposure to estrogen reduced the number of anti-tumor cytotoxic T cells and boosted immune cell-suppressive myeloid cells. EDCs may affect hormonal balance through a variety of mechanisms, including agonism, antagonism, and interference with hormonal synthesis, transport, metabolism and excretion.

Therefore, exposure to EDCs may lead to various effects mainly due to the different behavior of each specific receptor [59].

In our study, we observed a significant increase in estradiol levels in the EU glyphosate NOAEL dose group and this increase was also observed in the GLY, 2,4-D and dicamba herbicide mixture group, but without reaching the statistical significance suggesting an additive effect of 2,4-D and dicamba.

Interestingly, these effects were observed after short-term exposure during certain particular physiological periods with intense hormonal activity as pregnancy and lactation.

The ability of glyphosate, 2,4 -D, or dicamba to modulate the activity of estrogen receptors (ER) has been already well studied.

Glyphosate is considered to be an endocrine disruptor because it produced alterations in the function of reproductive organs in several studies [60, 61, 62].

Mechanistically, there is also evidence that glyphosate has 8 of the 10 characteristics defining an EDC capability [63].

Whether it activates directly estrogen receptors in mammary cells is still a matter of debate, since studies suggested both ligand-independent activation [64] and ligand-dependant activation [65] as a possible mechanism of glyphosate-mediated ER activation.

For 2,4-D, studies suggest that there is no direct activation of estrogen or androgen pathways but that the metabolism of other hormones like thyroid hormones can be disrupted albeit at doses that are significantly higher than the ones tested in our study [66].

Little evidence is available on dicamba except for a study suggesting alterations in sex hormone metabolism in a fish model [67].

Altogether, it is likely that the effects of the herbicide mixture observed in this study could be driven by the effects of glyphosate.

Conclusions

Our study has demonstrated that short-term exposure to glyphosate even at doses that according to regulators are considered without risk corroborated with a particular physiological state as gestation and lactation, can determine inflammatory changes in breast tissue in rats.

These findings support the need for further evaluation of glyphosate and its mixture with other pesticides for public health protection, especially for those categories vulnerable to the potential endocrine disruptor properties of these pesticides as pregnant women, newborns, and children.

Source of funding

This study was supported by internal grant no. 26/53/2/31.05.2022 of the University of Medicine and Pharmacy of Craiova, Craiova, Romania.

The work done by Robin Mesnage was funded in part by the Sustainable Food Alliance (USA) whose support is gratefully acknowledged.

Partially the work was also supported by Decree No 220 by the Government of the Russian Federation, Mega-Grant No 075-15-2022-1138.

Conflict of interests

None to declare.