U.S. flag

An official website of the United States government

Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Comparative investigation of the potential of glyphosate and glyphosate-based formulations (GBFs) to cause oxidative stress and DNA damage in human skin and liver cell systems

Stephanie L. Smith-Roe1, Michael J. DeVito1, Caroll Co2, Sreenivasa C. Ramaiahgari1, Michael Easterling2, Julie R. Rice1, Paul E. Dunlap1, David M. Crizer1, Zhifeng Zhou1, B. Alex Merrick1, Guanhua Xie2, Shawn F. Harris2, Keith R. Shockley3, Arpit Tandon4, Ayse Oktay4, Deepak Mav4, Ruchir Shah4, Alexandre Borrel4, Vijay Gombar4, Scott A. Masten1, Richard S. Paules1, Stephen S. Ferguson1*

Affiliations
1Division of Translational Toxicology, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA.
2DLH, LLC, Bethesda, Maryland, USA.
3Biostatistics and Computational Biology Branch, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA.
4Sciome, Durham, North Carolina, USA.

*Corresponding Author
Stephen S. Ferguson, Ph.D.
Mechanistic Toxicology Branch
Division of Translational Toxicology
National Institute of Environmental Health Sciences
111 TW Alexander Dr.
Building 101, Room E140B
P.O. Box 12233
Research Triangle Park, NC 27709
Email: stephen.ferguson@nih.gov; (984)287-3128

DOI: https://doi.org/10.22427/NTP-DATA-500-106-001-000-3

Publication

Abstract

Glyphosate is an herbicide found worldwide in glyphosate-based formulations (GBFs). Although glyphosate appears to have a low toxicity profile for humans and mammals, conflicting reports exist regarding the risk for cancer in humans. US-EPA and European regulatory agencies have described glyphosate as unlikely to pose a carcinogenic hazard to humans. However, the International Agency for Research on Cancer (IARC) classified glyphosate as “probably carcinogenic to humans (Group 2A)”, citing “mechanistic data provide strong evidence for genotoxicity and oxidative stress”. Given these discrepancies, the Division of Translational Toxicology at NIEHS designed an experimental strategy to expand mechanistic evidence and address critical gaps within existing literature (e.g., mechanistic evaluations of glyphosate alongside GBFs, inclusion of context-defining positive controls). Cell morphology, cell viability, H2O2 and H2AX formation were assayed in human keratinocytes (HaCaT), previously cited by IARC, and human hepatocytes (HepaRG™) to derive benchmark concentrations and fold-change metrics. Our findings revealed glyphosate alone was weakly and inconsistently bioactive for oxidative stress and DNA damage when compared to positive controls. In contrast, most of the 13 GBFs evaluated were more clearly bioactive with no apparent correlation to their varied glyphosate concentrations. Hierarchical clustering of biological responses revealed some bioactive GBFs to cluster near well-characterized positive controls for oxidative stress, while four GBFs clustered more similar to negative controls and glyphosate. Collectively, this study provides a robust dataset with context-defining results that advance our understanding of the hazard potential of GBFs.

Figures

Figure 1

Concentration response curves for loss of cell viability using the ATP depletion assays (CellTiter-Glo®) for the median BMC replicate of positive controls (menadione, TBHP; red), 13 GBFs (A to M; blue), glyphosate forms (glyphosate, G-IPA, AMPA; black) in HepaRG (HPRG2DC; A & C) and HaCaT (B & D) at 1 hour (A & B) and 24-hour (C & D) exposure times. GBFs are labeled with individual letters in their respective BMC orders. Dashed lines indicate fits for data filtered as non-responsive.

Figure 2

Relationship between GBF-induced benchmark concentrations (BMCs) for cell viability loss (i.e., ATP depletion) and percent glyphosate content for the 13 GBFs evaluated (i.e., A to M), glyphosate actives (i.e., glyphosate, G-IPA, and AMPA) in HepaRG (top; HPRG2DC) and HaCaT (bottom) following 24-hour exposures. Concentration response curves for inactive responses are reported as having BMCs greater than the maximum concentration tested (>max) for each test article. Glyphosate formulation maximum concentration ranged from 2,535 – 4,050 µg/mL, with a maximum of 8,454 µg/mL for glyphosate, 11,409 µg/mL for G-IPA, and 5,552 µg/mL for AMPA. One spurious replicate of GBF-K in HepaRG had an estimated BMC less than the minimum concentration tested.

Figure 3

Median BMC concentration response curves for H2O2 formation, a biomarker of ROS formation and oxidative stress for positive controls (menadione, TBHP; red), 13 glyphosate formulations (A to M; blue), and glyphosate forms (glyphosate, G-IPA, AMPA; black) in HepaRG (HPRG2DC; A & C) and HaCaT (B & D) at 1-hour (A & B) and 24-hour (C & D) exposure times. Concentration response curves for the positive controls (i.e., menadione and TBHP), glyphosate, and glyphosate forms G-IPA and AMPA are shown in red, black and green lines, respectively. GBFs shown in blue lines. GBFs (light blue) are labeled with individual letters in their respective BMC orders. Dashed lines indicate fits for data filtered as non-responsive.

Figure 4

Relationship between GBF-induced benchmark concentrations (BMC) for ROS formation (i.e., H2O2, ROS assay) in relation to percent glyphosate content for the 13 GBFs evaluated (i.e., A to M), glyphosate actives (i.e., glyphosate, G-IPA, and AMPA) in HepaRG (top; HPRG2DC) and HaCaT (bottom) following 24-hour exposures. Concentration response curves for inactive responses are reported as having BMCs greater than the maximum concentration tested (>max) for each test article. Glyphosate formulation maximum concentration ranged from 2,535 – 4,050 µg/mL, 8,454 µg/mL for glyphosate, 11,409 µg/mL for G-IPA, and 5,552µg/mL for AMPA.

Figure 5

Benchmark concentrations (µg/mL) for H2AX formation as a biomarker of DNA damage through double-stranded DNA breaks are summarized for human liver cells (HepaRG, HRG2DC in panel A) and human keratinocytes (HaCaT in panel B) following 24-hour exposures. Actives are shown as red triangles with inactive replicates shown as gray triangles as described in Materials and Methods. Gray bars represent range of concentrations tested for each compound and replicate. Black shading represents concentrations at which cytotoxicity (> 80% cell death) was observed.

Figure 6

Dendrogram (left) and constellation plot (right) from hierarchical clustering (Ward linkage) of 24h ROS (HepaRG, HaCaT), ATP (HepaRG, HaCaT) and gH2AX (HaCaT) assay BMC and fold-change data. Colors and markers represent a 4-cluster solution.

Tables

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

Table 11

Supplemental Figures

Supplemental Figure 1

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the CTG_1Hr_HaCaT assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 2

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the CTG_24Hr_HaCaT assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 3

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the CTG_1Hr_HRG2DC assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 4

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the CTG_24Hr_HRG2DC assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 5

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the ROS_1Hr_HaCaT assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 6

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the ROS_24Hr_HaCaT assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 7

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the ROS_1Hr_HRG2DC assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 8

Distribution of External Administered Dose (EAD) estimates (mg/kg/dose) for the ROS_24Hr_HRG2DC assay. Data points represent the minimum (light blue), median (green), and maximum (dark blue). Gray bars indicate the full range and red stars denote individual EAD replicates. EAD ranges were calculated from Benchmark Concentration (BMC) ranges. Analysis was restricted to filtered BMCs that met concentration and hitcall criteria (> 0.9); specific filtering conditions are detailed in Table 3.

Supplemental Figure 9

Benchmark concentrations for Cell Titer Glo as a biomarker of cytotoxicity are summarized for human liver cells (HepaRG, HRG2DC) and human keratinocytes (HaCaT) following 1-hour and 24-hour exposures. Actives are shown as blue triangles with inactive replicates shown as gray triangles as described in Materials and Methods. Orange triangles represent point of departure where an increase in ATP was observed. Gray bars represent range of concentrations tested for each compound and replicate.

Supplemental Figure 10

Benchmark concentrations for ROS-Glo™ as a biomarker of oxidative stress are summarized for human liver cells (HepaRG, HRG2DC in panel A) and human keratinocytes (HaCaT in panel B) following 1-hour and 24-hour exposures. Actives are shown as orange triangles with inactive replicates shown as gray triangles as described in Materials and Methods. Gray bars represent range of concentrations tested for each compound and replicate.

Supplemental Figure 11

Benchmark concentrations for Mito-SOX as a biomarker of mitochondrial superoxide are summarized for human keratinocytes (HaCaT) following 1-hour exposure. Actives are shown as blue triangles with inactive replicates shown as gray triangles as described in Materials and Methods. Gray bars represent range of concentrations tested for each compound.

Supplemental Figure 12

Dendrogram (left) and constellation plot (right) from hierarchical clustering (Ward linkage) of 24h ROS (HepaRG & HaCaT) assay BMC and fold-change data.

Supplemental Figure 13

Dendrogram (left) and constellation plot (right) from hierarchical clustering (Ward linkage) of 24h CTG (HepaRG & HaCaT) assay BMC and fold-change data.

Supplemental Figure 14

Dendrogram (left) and constellation plot (right) from hierarchical clustering (Ward linkage) of 24h ROS (HepaRG & HaCaT), and CTG (HepaRG & HaCaT) assay BMC and fold-change data.

Supplemental Figure 15

Dendrogram (left) and constellation plot (right) from hierarchical clustering (Ward linkage) of 24h H2AX HaCaT assay BMC and fold-change data.

Supplemental Tables

Supplemental Table 1

Supplemental Table 2

Supplemental Table 3

Supplemental Table 4

Supplementary Files

Imaging Data

Supplemental Data Files

Processed Data Files with Metadata