Lung Inammation from Single and Repetitive Exposure to Glyphosate

Background: Glyphosate is an active ingredient in herbicides used in agriculture worldwide. Exposure to glyphosate has been associated with respiratory dysfunctions in agricultural workers. However, the ability of glyphosate to induce inammation in the lung is not well studied. Therefore, we evaluated lung inammatory response to glyphosate at agricultural relevant dose for single and repetitive exposures. Methods: Male C57BL/6 mice were intranasally exposed to glyphosate (1 μg/40 μl) for 1-day or once daily for 5-days, and 10-days. After the exposure periods, mice were euthanized to collect the bronchoalveolar lavage (BAL) uid and lung tissue. Results: Repetitive exposure to glyphosate for 5-days and 10-days showed an increase of neutrophils in BAL uid and eosinophil peroxidase levels in lungs, a marker for eosinophils. Leukocyte inltration in lungs was further conrmed through lung histology. Th2 cytokines including IL-5 and IL-13 were increased in BAL uid after 10-days of glyphosate exposure whereas IL-4 was not increased. Lung sections from all glyphosate groups showed higher expression for ICAM-1, VCAM-1, and vWF adhesion molecules. TLR-4 and TLR-2 expression was increased in lungs after repetitive exposure to glyphosate. Conclusions: We conclude that repetitive exposure to glyphosate induces migration of neutrophils and eosinophils and release of Th2 cytokines. This study, for the rst time, provides evidence for the role of ICAM-1, VCAM-1 and vWF in lungs of glyphosate-treated animals. goat anti-rabbit incubated using end, series of xed before Controls with the omission of the primary antibody or secondary occasional staining, 1: weak staining, 2: moderate staining, 3: intense staining).

cytokines. The pulmonary endothelial cells show constitutive expression of various adhesion molecules such as intercellular adhesion molecule-1 (ICAM-1), vascular adhesion molecule-1 (VCAM-1) and von Willebrand factor (vWF) (12). However, there is no data on the expression of adhesion molecules in the lungs of glyphosate-exposed animals.
The present study was designed to characterize the lung in ammation induced with exposure to agricultural relevant dose of glyphosate (1 µg) for different periods (1-day, 5-days, and 10-days) by using mice model. Our data shows that repetitive exposure to glyphosate increased the in ammatory markers and expression of adhesion molecules in lungs.

Mice exposures
The experimental protocols were approved by the Animal Ethics Research Board of the University of Saskatchewan (Protocol # 20160106). Male C57BL/6 mice (Charles River Laboratories, Montreal, QC Canada), 6-8 weeks old, were maintained at the Laboratory Animal Services Unit of the University of Saskatchewan. Mice were fed ad libitum and were acclimatized for one week after arrival.
Mice were divided into glyphosate and control treatment groups (n = 5 per group). The 1 µg dose of glyphosate treatment was selected based on glyphosate levels found in the agricultural environment and has been utilized in other studies (11). The stock solution of glyphosate (0.8 M; analytical grade PESTANAL standard, Sigma, St. Louis, MO USA) was prepared in Hank's Balanced Salt Solution (HBSS).
It was vortexed for 10 minutes and syringe ltered (0.22 µm; Fisher Scienti c). Mice received 40 µL of either glyphosate (1 µg/40 µl) or saline intranasally for 1-day or daily for 5-days or 10-days. Mice were lightly anesthetized using iso urane before treatments. There were no differences in the weight of mice in the control and the glyphosate-treated groups. After 4-hours of last treatment, mice were euthanized by CO2 inhalation, and BAL uid and lung samples were collected.
Bronchoalveolar lavage collection and processing BAL uid was collected by washing the airways three times with 0.5 ml ice-cold HBSS. The collected BAL uid was centrifuged at 1000 g for 10 minutes at 4° C, and supernatants were stored at -80° C for cytokine analysis. Cells from BAL uid was resuspended in HBSS and kept on ice until used for leukocyte counts.
Total and differential leukocyte count The total and differential leukocyte counts in BAL uid were performed using a hemocytometer and cytospin stained with Protocol Hema 3 kit (ThermoFisher Scienti c, Waltham, MA USA), respectively.

Histology and scoring for in ammation
Lung sections from all the mice were stained with hematoxylin & eosin stain. Two stained lung sections from each mouse were reviewed and scored for lung in ammation. Each of the sections on the slide was reviewed at different magni cations (x20, x40, x100). Each section's scoring was performed by multiple reviewers blinded to the exposure groups, and scores were averaged. Cellular in ltration in alveolar, perivascular, and peribronchiolar compartments of lungs were scored. Each parameter was given a score based on the intensity and was statistically analyzed (0: absent, 1: mild, 2: moderate, 3: severe).

Immunohistochemistry and analysis
Lung sections from all mice were stained with antibodies against ICAM-1, VCAM-1, and vWF markers.
Brie y, lung sections were immersed in a series of xylene baths for depara nization and different alcohol grades for rehydration. Endogenous peroxidase activity was quenched with 0.5% hydrogen peroxide in methanol for 20 minutes. Antigen unmasking and blocking were done for 30 minutes with 2 mg/ml pepsin and 1% bovine serum albumin, respectively. The lung sections were incubated overnight at 4 0 C with the following primary antibodies: ICAM-1 (dilution 1: 100; rabbit monoclonal anti-mouse ICAM-1, The expression of ICAM-1, VCAM-1, and vWF was reviewed in ve random elds of the lung sections from each mouse (N = 3). Expression of ICAM-1 was scored in the vasculature, and bronchial epithelium. Each parameter staining was given a score based on staining intensity by a reviewer blinded to exposure groups (0: no or occasional staining, 1: weak staining, 2: moderate staining, 3: intense staining).

Data analysis and statistics
Data was analyzed using GraphPad Prism 6 (GraphPad Software, San Diego, CA). Error bars represent mean +/-SD. For values outside the assay limit of detection, either the LLOD/2 (lowest limit of detection) or a minimum value below the lowest attained value was designated. Statistical signi cance was determined using one-way ANOVA with a follow-up Tukey test for multiple comparisons. If the assumption of equal variance was not met, the data was either log-transformed, followed by one-way ANOVA and multiple comparison tests, or a non-parametric Kruskal-Wallace test was conducted. A pvalue < 0.05 was considered signi cant for differences between groups. For graphing of data, "a" indicates a signi cant difference compared with the control group; "1" indicates a signi cant difference compared with the 1-day exposure group; "2" indicates a signi cant difference compared with the 5-days exposure group.

Leukocyte counts in bronchoalveolar lavage uid
Glyphosate exposure for 5-days and 10-days signi cantly increased the total leukocyte counts in BAL uid as compared to control exposures (Fig. 1A). Total leukocytes count in 10-days glyphosate exposure group was signi cantly higher than 1-day glyphosate treatment group.
Neutrophils were signi cantly higher after glyphosate exposure for 5-days and 10-days as compared to same days of control exposure (Fig. 1C, D). Macrophage and lymphocyte count in any of the glyphosate treatment groups were not signi cantly different than control exposures (Fig. 1B-D).

Eosinophil peroxidase in lungs
Eosinophil peroxidase (EPO) is released by activated eosinophils and acts as a marker for the presence of eosinophils. Glyphosate exposure signi cantly increased the EPO levels in the lungs after 1-day, 5-days, and 10-days exposures as compared to the respective control exposure (Fig. 3).

Lung histology
Lungs of control mice showed normal architecture without any leukocyte in ltration in 1-day (Fig. 4A), 5days (Fig. 4B), and 10-days (Fig. 4C) exposure groups. Lungs of glyphosate treated mice in the 1-day exposure group showed occasional sloughing and increased thickness of bronchial epithelium, and slight leukocyte in ltration (Fig. 4D, G). Glyphosate exposure in the 5-days (Fig. 4E, H) and 10-days (Fig. 4F, I) exposure groups showed a greater leukocyte in ltration in the alveolar, perivascular, and peribronchiolar regions of the lungs. Moreover, both group's lung sections, 5-days, and 10-days glyphosate exposure groups showed cellular binding to the vascular endothelium, occasional sloughing, and increased thickness of the bronchial epithelium.
Semi-quanti cation data showed that glyphosate exposure in the 1-day group had signi cant leukocyte in ltration in the perivascular region as compared to 1-day control exposure (Fig. 4J). Glyphosate exposure for both the 5-days and 10-days signi cantly increased the leukocyte in ltration in the perivascular, peribronchiolar (Fig. 4K), and alveolar (Fig. 4L) regions as compared to the respective control exposure.
Glyphosate exposure for 1-day (Fig. 5D), 5-days (Fig. 5E), and 10-days (Fig. 5F) increased staining for ICAM-1 in alveolar septa and endothelium in large blood vessels. ICAM-1 staining was also induced in bronchial epithelium of lungs from all periods of glyphosate groups. Semi-quanti cation data showed that ICAM-1 blood vessel staining was signi cantly increased after glyphosate exposure for 1-day, 5-days, and 10-days compared to the respective control exposures (Fig. 5G). ICAM-1 bronchial epithelium staining signi cantly increased in the glyphosate 5-days and 10-days exposure groups compared to control groups (Fig. 5H).
Similarly, the staining for VCAM-1 (Fig. 6D-F) and vWF (Fig. 7D-I) was also increased in bronchial epithelium, alveolar septa, and endothelium of large blood vessels in lung sections of glyphosate groups for 1-day, 5-days, and 10-days.
Glyphosate exposures for 5-days and 10-days showed a signi cant increase in TLR-4 as compared to the control exposures (Fig. 8B). TLR-4 mRNA in the 5-days glyphosate group was signi cantly higher compared to the 1-day glyphosate exposure group.
TLR-2 mRNA expression in the glyphosate group was signi cantly higher after 5-days as compared to both the 5-days control and 1-day glyphosate exposure groups (Fig. 8C).
Hsp72 mRNA expression was signi cantly higher after glyphosate exposure for 5-days than 5-days control and 1-day glyphosate exposures (Fig. 8D). Hsp72 was signi cantly lower in the glyphosate 10days exposure group as compared to the 5-days glyphosate exposure group.
Expression of A20 was not signi cantly different between 1-day, 5-days, or 10-days glyphosate exposures compared to control exposure (Fig. 8E).

Cytotoxicity of A549 cells
Glyphosate treatment at different concentrations showed dose-dependent toxicity on A549 cells (Fig. 9). The reduction in viable cell percentage was observed with the increase in glyphosate treatment concentration (1 nM, 10 nM, 100 nM, 1 µM, 10 µM, 100 µM, 1 mM, and 10 mM). Glyphosate treatment at 1 mM and 10 mM showed a signi cant cell death compared to the untreated cells. The dotted line in the graph represents LD 50 after glyphosate treatment.

Discussion
We report the rst study where mice were treated for single and multiple days with glyphosate. The data reported in this paper show that multiple exposure to glyphosate induced migration of neutrophils and eosinophils and release of IL-5 and IL-13 cytokines. Staining for ICAM-1, VCAM-1, and vWF was increased in the alveolar septa and endothelium of large blood vessels in all glyphosate treated lungs, suggesting their role in leukocyte migration after glyphosate exposure. These data add to the evidence on the ability of glyphosate to induce lung in ammation and therefore it may play a role in human respiratory problems.
Lung in ammation is characterized by recruitment of in ammatory cells such as neutrophils and eosinophils in response to in ammatory agent. Glyphosate herbicides are used in agriculture worldwide and glyphosate exposure has been associated with respiratory dysfunction in agricultural workers (1, 7). Glyphosate's ability to induce in ammation in the lungs is not well understood. Therefore, we used mice to evaluate lung in ammation after single (1-day) and repetitive exposure (5-days and 10-days) to glyphosate at agricultural relevant dose. Our results show that repetitive exposure to glyphosate compared to single exposure induced greater cellular in ltration in lungs. The induction of in ammation was further underscored by neutrophils in BAL uid and levels of eosinophil marker in lungs after repetitive exposure to glyphosate. The data from an earlier study (11) showed similar ndings after 7days of glyphosate exposure. While Kumar and colleagues examined lung effects at only one time-point (7 days) (11) we did so at 1-day, 5-days, and 10-days of glyphosate exposure. The lung in ammation caused by chronic exposure to glyphosate may be associated with lung function impairments observed in pesticide applicators as high prevalence of chronic respiratory symptoms and decline in lung function were reported (13)(14)(15).
The recruitment of in ammatory cells is a multi-step process and is facilitated through the release of cytokines and upregulation of endothelial adhesion molecules. Repetitive exposure to glyphosate for 10days signi cantly increased Th2 cytokines IL-5 and IL-13 whereas there was no increase in IL-4. These Th2 cytokine results are similar to the ndings of others reported after 7-days of glyphosate exposure (11). IL-13 de cient mice exposed to glyphosate for 7-days and 21 days (3 times exposure for 3 weeks) showed less cellular in ltration and diminished production of IL-5 in lungs (11). These results suggests that IL-13 signaling in lungs may be critical in observed in ammation after glyphosate exposure. Both IL-5 and IL-13 have important roles in lung in ammation associated with asthma or COPD (16)(17)(18). IL-5 is involved in the recruitment of eosinophils in the asthmatic lung; Interestingly we did notice an increase in eosinophils in the lung tissues from glyphosate-treated mice.
Kumar and colleagues showed an increase of IL-33, and IL-10 after 7-days of glyphosate exposure (11). Our same dose of glyphosate and longer exposure periods showed no differences in these cytokines. However, glyphosate-treated lungs showed damage to lung epithelium which is one of the potential sources of these cytokines (19,20). These differences in cytokine results could be possible that we missed the time point for their increase in 5-days gap between our exposure periods. Secondly, the two studies used different gender of mice for glyphosate exposure. There is a known difference in in ammatory response between the different genders (21,22). Gender response may be an important factor for differences in cytokine responses after glyphosate exposure which needs further studies.
The activation of pulmonary endothelium by cytokines leads to the expression of adhesion molecules which are important for the tissue recruitment of in ammatory cells such as neutrophils and eosinophils (12). There is no data on expression of adhesion molecules in lungs of animals treated with glyphosate. Our study provides rst immunohistochemical data showing increased pulmonary expression of ICAM-1, VCAM-1 and vWF proteins in glyphosate treated mice. It is well known that ICAM-1 and VCAM-1 engage selectins to slow-down rolling neutrophils and vWF secreted from Weibel-Palade bodies in endothelial cells facilitates recruitment of platelets (23,24) and is a marker of in ammation (25)(26)(27). The higher levels of cytokines such as IL-13 and IL-5 likely caused the observed increase in the expression of adhesion molecules in lungs of animals treated with glyphosate. Nevertheless, the increased endothelial expression of ICAM-1, VCAM-1 and vWF provides molecular evidence of vascular in ammation in the lungs of glyphosate exposed mice.
For agricultural crop production workers, glyphosate is often co-exposed with other well-known lung in ammation stimulants, most notably endotoxin (LPS) as a component of organic dust. Grain dust inhalation has been shown to induce neutrophilic in ammation in exposed individuals as well as in animal models, and endotoxin (LPS) in the grain dust is often associated with respiratory dysfunction among grain farmers (28)(29)(30). Because bacterial molecules activate cells upon binding with TLRs (31), we examined the expression of TLR4 and TLR2 in lungs from mice in our experiments and found an increase in their expression following multiple exposure to glyphosate. While it is not known whether glyphosate directly activates TLR-mediated cell activation, the increased expression of TLRs will enhance sensitivity of the lung to bacterial challenges. Activation of the TLR receptors is critical for in ammatory signaling resulting in the release of cytokines and upregulation of endothelial adhesion molecules, which are necessary for leukocyte migration (32). These results suggest the need to study glyphosate exposure by using TLR receptor knock-out mice which would provide a better understanding on mechanistic role of TLRs in glyphosate induced lung in ammation.
Finally, our in vitro data show cytotoxicity of glyphosate (1 nM to 10 mM) on A549 cells. The dosedependent effect of glyphosate on A549 cellular toxicity suggests that lower doses of glyphosate are less likely to have a toxic effect on lung epithelial cells. Therefore, these cellular results support our lung in ammatory ndings of mice experiment after exposure to a lower dose of glyphosate. Recent data show that isopropylamine salt of glyphosate did not contribute to apoptotic and genotoxic effects on A549 cells (33). Therefore, our results need to be interpreted with caution as they are based on pure glyphosate. Further studies are needed to examine not only the mechanism of glyphosate-induced cytotoxicity but also the cytotoxicity in vivo. For example, we would like to examine apoptosis and cell death in the lungs of glyphosate-treated animals.

Conclusions
Repetitive exposure to glyphosate induced a migration of neutrophils and eosinophils and release of IL-5 and IL-13. It was concomitantly associated with increased pulmonary expression of ICAM, VCAM-1, and vWF adhesion molecules and TLR-4 and TLR-2 receptors. Exposure to glyphosate could be a risk factor for respiratory dysfunctions observed among agricultural workers. Future studies are needed to investigate the chronic effects of glyphosate on markers of airway in ammation and lung pathology.