Maternal glyphosate exposure causes autism-like behaviors in offspring through increased expression of soluble epoxide hydrolase

Article abstract of DOI:10.1073/pnas.1922287117

Epidemiological studies suggest that exposure to herbicides during pregnancy might increase risk for autism spectrum disorder (ASD) in offspring. However, the precise mechanisms underlying the risk of ASD by herbicides such as glyphosate remain unclear. Soluble epoxide hydrolase (sEH) in the metabolism of polyunsaturated fatty acids is shown to play a key role in the development of ASD in offspring after maternal immune activation. Here, we found ASD-like behavioral abnormalities in juvenile offspring after maternal exposure to high levels of formulated glyphosate. Furthermore, we found higher levels of sEH in the prefrontal cortex (PFC), hippocampus, and striatum of juvenile offspring, and oxylipin analysis showed decreased levels of epoxy-fatty acids such as 8 (9)-EpETrE in the blood, PFC, hippocampus, and striatum of juvenile offspring after maternal glyphosate exposure, supporting increased activity of sEH in the offspring. Moreover, we found abnormal composition of gut microbiota and short-chain fatty acids in fecal samples of juvenile offspring after maternal glyphosate exposure. Interestingly, oral administration of TPPU (an sEH inhibitor) to pregnant mothers from E5 to P21 prevented ASD-like behaviors such as social interaction deficits and increased grooming time in the juvenile offspring after maternal glyphosate exposure. These findings suggest that maternal exposure to high levels of glyphosate causes ASD-like behavioral abnormalities and abnormal composition of gut microbiota in juvenile offspring, and that increased activity of sEH might play a role in ASD-like behaviors in offspring after maternal glyphosate exposure. Therefore, sEH may represent a target for ASD in offspring after maternal stress from occupational exposure to contaminants.

Full text of DOI:10.1073/pnas.1922287117

Maternal glyphosate exposure causes autism-like
behaviors in offspring through increased expression of
soluble epoxide hydrolase
Yaoyu Pua, Jun Yangb, Lijia Changa, Youge Qua, Siming Wanga, Kai Zhanga, Zhongwei Xionga, Jiancheng Zhanga,
Yunfei Tana, Xingming Wanga, Yuko Fujitaa, Tamaki Ishimaa, Debin Wangb, Sung Hee Hwangb, Bruce D. Hammockb,1,

and Kenji Hashimotoa,1
aDivision of Clinical Neuroscience, Chiba University Center for Forensic Mental Health, 260-8670 Chiba, Japan; and bDepartment of Entomology and
Nematology and UC Davis Comprehensive Cancer Center, University of California, Davis, CA 95616
Contributed by Bruce D. Hammock, March 30, 2020 (sent for review January 6, 2020; reviewed by Staci D. Bilbo and Mikhail M. Pletnikov)
Epidemiological studies suggest that exposure to herbicides
during pregnancy might increase risk for autism spectrum disorder
(ASD) in offspring. However, the precise mechanisms underlying
the risk of ASD by herbicides such as glyphosate remain unclear.
Soluble epoxide hydrolase (sEH) in the metabolism of polyunsat-
urated fatty acids is shown to play a key role in the development
of ASD in offspring after maternal immune activation. Here, we
found ASD-like behavioral abnormalities in juvenile offspring after
maternal exposure to high levels of formulated glyphosate. Fur-
thermore, we found higher levels of sEH in the prefrontal cortex
(PFC), hippocampus, and striatum of juvenile offspring, and oxylipin
analysis showed decreased levels of epoxy-fatty acids such as 8
(9)-EpETrE in the blood, PFC, hippocampus, and striatum of juve-
nile offspring after maternal glyphosate exposure, supporting
increased activity of sEH in the offspring. Moreover, we found
abnormal composition of gut microbiota and short-chain fatty
acids in fecal samples of juvenile offspring after maternal glyph-
osate exposure. Interestingly, oral administration of TPPU (an sEH
inhibitor) to pregnant mothers from E5 to P21 prevented ASD-like
behaviors such as social interaction deficits and increased groom-
ing time in the juvenile offspring after maternal glyphosate ex-
posure. These findings suggest that maternal exposure to high
levels of glyphosate causes ASD-like behavioral abnormalities
and abnormal composition of gut microbiota in juvenile off-
spring, and that increased activity of sEH might play a role in
ASD-like behaviors in offspring after maternal glyphosate expo-
sure. Therefore, sEH may represent a target for ASD in offspring after
maternal stress from occupational exposure to contaminants.
period as reported in the US public school system (11–13). A
recent population-based case-control study in California showed
that the risk of ASD was associated with the use of glyphosate
(odds ratio = 1.16) (14). For ASD with intellectual disability,
estimated odds ratio were higher (by about 30%) with prenatal
exposure to glyphosate (odds ratio = 1.33) (14). These reports
suggest that possible relationships between glyphosate and ASD
should be explored in animal models.
Epidemiological studies implicate prenatal environmental
factors, including maternal immune activation (MIA), playing a
key role in the etiology of developmental disorders such as ASD
(15–19). There are a number of positive associations between
maternal infections or inflammatory biomarkers and ASD (15,
16, 20). Collectively, MIA during pregnancy can increase the risk
of developmental disorders such as ASD in offspring.
Epoxy fatty acids (EpFAs) are produced from the correspond-
ing polyunsaturated fatty acids by cytochrome P450 enzymes.
Epoxyeicosatrienoic acids (EpETrEs) and epoxydocosapentaenoic
acids (EpDPEs) are produced from arachidonic acid and doco-
sahexaenoic acid (DHA), respectively. EpETrEs, EpDPEs, and some
other EpFAs have potent antiinflammatory properties. However,
Significance
Maternal exposure to high levels of the herbicide glyphosate
may increase the risk for autism spectrum disorder (ASD) in
offspring; however, the underlying mechanisms remain largely
unknown. Maternal glyphosate exposure during pregnancy
and lactation caused ASD-like behavioral abnormalities and
abnormal composition of gut microbiota in murine male off-
spring. Soluble epoxide hydrolase (sEH) in the brain of off-
spring after maternal glyphosate exposure was higher than
controls. Treatment with an sEH inhibitor from pregnancy to
weaning prevented the onset of ASD-like behavioral abnor-
malities in offspring after maternal glyphosate exposure. The
glyphosate exposures used here exceed any reasonable di-
etary, environmental, or occupational exposure, but they in-
dicate that increased sEH plays a role in ASD-like behaviors in
offspring.
glyphosate gut microbiota soluble epoxide hydrolase
|
|
utism spectrum disorder (ASD) is a developmental disorder
characterized by social and communication impairments,
A
combined with limited or focused interests and repetitive be-
haviors (1, 2). Although the prevalence of ASD has been rising
since the 1980s, the detailed reasons underlying this rise remain
unknown (3, 4). In addition to genetic factors, accumulating
evidence supports a significant contribution of environmental
factors in ASD etiology (1, 2, 5, 6). Environmental factors, in-
cluding exposures to synthetic chemicals during pregnancy and
lactation, are suggested to play a role in the development of
ASD. These chemicals include selective serotonin reuptake in-
hibitors (SSRIs), pesticides, phthalates, polychlorinated biphe-
nyls, solvents, air pollutants, fragrances, and heavy metals (6–8).
Glyphosate [N-(phosphonomethyl)glycine] is the active in-
gredient in Roundup and other herbicides, and, because of its
efficacy, excellent environmental profile, and low toxicity, it is
the most widely used herbicide in the world (9, 10). Interestingly,
a positive correlation was reported between the rise of glyph-
osate usage on corn and soy crops in the United States over the
years 1995 to 2010 and the increase in ASD rates over the same
Author contributions: B.D.H. and K.H. designed research; Y.P., J.Y., L.C., Y.Q., S.W., K.Z.,
Z.X., J.Z., Y.T., X.W., Y.F., T.I., D.W., and S.H.H. performed research; J.Y., D.W., S.H.H., and
B.D.H. contributed new reagents/analytic tools; Y.P. analyzed data; and Y.P., B.D.H., and
K.H. wrote the paper.
Reviewers: S.D.B., Duke University; and M.M.P., Johns Hopkins University.
The authors declare no competing interest.
Published under the PNAS license.
1To whom correspondence may be addressed. Email: bdhammock@ucdavis.edu or
hashimoto@faculty.chiba-u.jp.
This article contains supporting information online at https://www.pnas.org/lookup/suppl/
doi:10.1073/pnas.1922287117/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1922287117
PNAS Latest Articles
|
1 of 7

these lipid mediators are metabolized rapidly into their corre- and prepulse inhibition (PPI; for psychosis) were not different
sponding diols by soluble epoxide hydrolase (sEH), and inhibition between the two groups (Fig. 1 D and F). In the novel object
of sEH enhances the beneficial effects of EpFAs (21–24). Accu-
mulating evidence demonstrate a key role of sEH in multiple
animal models, including depression, ASD, schizophrenia, and
Parkinson’s disease (25–32). Recently, we reported that sEH in
the prefrontal cortex (PFC) plays a key role in the development of
ASD-like behavioral abnormalities in juvenile offspring after MIA
(30). However, there is no previous report showing the role of
sEH in the pathogenesis of ASD in offspring after maternal ex-
posure to formulated glyphosate.
The purpose of this study was to examine the role of sEH in
the pathogenesis of ASD in offspring after maternal glyphosate
exposure. First, we examined whether maternal glyphosate ex-
posure causes ASD-like behavioral abnormalities in juvenile off-
recognition test (NORT), offspring after maternal glyphosate ex-
posure showed cognitive deficits (Fig. 1E). In the three-chamber
test, juvenile offspring after maternal glyphosate exposure showed
social interaction deficits compared to the water-treated group
(Fig. 1G). The data suggest that maternal exposure to glyphosate
in a commercial formation (SI Appendix, Supplementary Material
for composition) causes ASD-like cognitive deficits and social
interaction deficits in juvenile offspring.
Increased Expression of sEH in the Brain of Juvenile Offspring after
Maternal Glyphosate Exposure. We measured the expression of
sEH in the brain since increased expression of sEH in the PFC
plays a role in the ASD-like behaviors after MIA (30). Protein
spring. Second, we examined whether expression of sEH is altered levels of sEH in the PFC and striatum, but not hippocampus, from
in the brain regions of juvenile offspring after maternal glyphosate mothers treated with glyphosate were significantly higher than
exposure. Furthermore, we performed oxylipin analysis of blood those of water-treated mice (Fig. 1H). Protein levels of sEH in the
and brain regions from juvenile offspring. Moreover, we measured PFC, hippocampus, and striatum from juvenile offspring (P28)
levels of N-methyl-D-aspartate receptor (NMDAR)-related amino
acids in the blood and brain from juvenile offspring since
NMDAR-related amino acids were altered in patients with
ASD (33–36). Third, we performed 16S rRNA analysis and
measurement of short-chain fatty acids of fecal samples in juvenile
off\spring after maternal glyphosate exposure since abnormal
composition of gut microbiota is shown in patients with ASD
(36–40). Finally, we examined whether treatment with TPPU
[1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl)urea] (40–42),
a potent sEH inhibitor, in pregnant mice from pregnancy to
weaning could prevent behavioral abnormalities in juvenile off-
spring after maternal glyphosate exposure.
after maternal glyphosate exposure were significantly higher than
those of water-treated mice (Fig. 1I). Furthermore, gene expres-
sion of sEH (or Ephx2) mRNA in the PFC, hippocampus, and
striatum from juvenile offspring (P28) after maternal glyphosate
exposure was significantly higher than those of water-treated mice
(Fig. 1J).
Next, we performed parvalbumin (PV) immunohistochemistry
in the brain from juvenile mice (Fig. 1K). PV immunoreactivity
in the prelimbic (PrL), but not IL (infralimbic), of medial PFC in
the offspring of maternal glyphosate exposure was significantly
lower than that of the water-treated group (Fig. 1K).
Oxylipin Analysis of Blood and Brain Regions. Using oxylipin anal-
ysis, we measured the levels of eicosanoid metabolites in the
blood, PFC, hippocampus, and striatum from juvenile offspring
(P28) after maternal glyphosate exposure (SI Appendix, Fig. S2
and Tables S2–S5). Blood levels of many epoxides were signifi-
cantly lower in juvenile offspring after maternal glyphosate ex-
posure (SI Appendix, Table S2). We found higher levels of 8
(9)-EpETrE [8,9-epoxy-5Z,11Z,14Z-eicosatrienoic acid] compared
to other EpFAs in the mouse brain. Levels of 8 (9)-EpETrE in the
PFC, hippocampus, and striatum were significantly lower in juvenile
offspring (P28) after maternal glyphosate exposure (Fig. 2 and SI
Appendix, Tables S3–S5). Lower levels of 8 (9)-EpETrE in the brain
regions from juvenile offspring after maternal glyphosate exposure
support the increased expression of sEH in these regions. In con-
trast, tissue levels of other EpFAs in the PFC, hippocampus, and
striatum from juvenile offspring after maternal glyphosate expo-
sure were significantly higher than those of control mice (SI Appendix,
Tables S3–S5).
Results
General and Behavioral Data of Mother and Juvenile Offspring after
Maternal Glyphosate Exposure. First, we examined whether ma-
ternal glyphosate exposure could affect the general and behav-
ioral outcomes in male offspring (SI Appendix, Fig. S1A).
Previous studies used drinking water containing 1% Roundup
[0.38% (wt/vol) glyphosate] during pregnancy and lactation (43, 44).
For this study, water or formulated glyphosate [0.039% (wt/vol)
glyphosate] were given to pregnant mice from E5 to P21 (weaning).
The mortality of pregnant mice at the highest concentration
(0.39%) in our hands was 100%, although the mortality of lower
concentrations (0.039% and 0.098%) was 0% (SI Appendix, Table
S1), even though the 0.39% concentration has been previously used
in other studies (43, 44). This discrepancy may have been due to
different formulations of glyphosate or mouse strain differences
(ddy vs. ICR) between the studies. Body weight of pregnant mice
was increased gradually after maternal glyphosate (0.039 to 0.293%)
exposure, whereas body weight of pregnant mice treated with the
high concentration (0.39%) did not increase (SI Appendix, Fig.
S1B). The mortality of offspring in the 0.039% glyphosate group
was 0%, and juvenile offspring after maternal 0.039% glyphosate
exposure did not show any behavioral abnormality such as loco-
motion, social interaction deficits in a three-chamber test, and
depression-like phenotype in the forced swimming test (SI Appen-
dix, Fig. S1 C–E and Table S1). In contrast, we found social in-
teraction deficits in juvenile offspring after maternal 0.098%
glyphosate exposure. Therefore, we used 0.098% glyphosate in
the subsequent experiments. This concentration corresponded with
1/80th of the glyphosate no-observed-adverse-effect level, as
reported previously (45).
Measurement of Amino Acids in the Blood and Brain. Next, we
measured levels of NMDAR-related amino acids (i.e., glutamate,
glutamine, glycine, D-serine, L-serine, GABA) in the plasma and
brains of juvenile offspring (P28) after maternal glyphosate ex-
posure. Maternal glyphosate exposure caused significant reduc-
tions of glutamate in the plasma and brain regions. In addition,
maternal glyphosate exposure caused significant reductions of
other amino acids (i.e., glycine, L-serine, GABA) in the PFC (SI
Appendix, Table S6). The data suggest abnormalities in NMDAR-
related neurotransmission in the PFC of juvenile offspring after
maternal glyphosate exposure.
16S rRNA Analysis and Measurement of Short-Chain Fatty Acids of
Fecal Samples of Juvenile Offspring after Maternal Glyphosate
Exposure. We performed 16S rRNA analysis of fecal samples of
offspring (P28). Maternal glyphosate exposure caused abnormal
composition of gut microbiota in juvenile offspring (Fig. 3). At the
Body weight of glyphosate-exposed mothers was significantly
lower than that of water-exposed mothers at E17 (Fig. 1 A and
B). On P21 (weaning), we could detect blood levels of glyphosate
in the mothers treated with 0.098% glyphosate and their off-
spring, although glyphosate was not detected in the dams and
offspring of the water-treated group (Fig. 1 A and C). Locomotion species level, the relative abundance of Eubacterium plexicaudatum,
2 of 7
|
www.pnas.org/cgi/doi/10.1073/pnas.1922287117
Pu et al.

Mother
A
C
B
80
60
40
20
0
Water
Glyphosate
**
E5
E12
E17
NORT
24 Hrs
Locomotion
N.S.
PPI
Blood
D
Glyphosate
E
F
60
50
40
30
20
10
0
80
60
40
20
0
600
400
200
0
6000
4000
2000
0
Water
*
N.S.
Water
N.S.
Glyphosate
N.D.
N.D.
Mother
Offspring
Water Glyphosate
PP69
PP73
PP77
PP81
Training
Test
G
300
200
100
0
**
300
**
Stranger 1
Stranger 2
Empty
Stranger 1
200
100
0
N.S.
N.S.
Water
Glyphosate
Water
Glyphosate
Offspring
Mother
Offspring
***
H
I
J
150
100
50
200
150
Water
*
***
*
Glyphosate
*
*
N.S.
***
*
150
100
50
100
50
0
0
0
PFC
Hippocampus Striatum
PFC
Hippocampus Striatum
PFC
Hippocampus Striatum
sEH
β-actin
Water
Glyphosate
K
PrL
IL
IL
N.S.
PrL
*
25
20
15
10
5
40
30
20
10
0
Water
Glyphosate
0
Fig. 1. Social interaction deficits, increased expression of sEH, and decreased PV immunoreactivity in the brain from juvenile offspring after maternal
glyphosate exposure. (A) Schedule of treatment, behavioral tests, and sample collection. (B) Change of body weight of pregnant mothers (n = 6). (C) Blood
levels of glyphosate in the mothers and offspring at P21. Data are shown as mean SEM (mother n = 7, offspring n = 10). (D) Locomotion. Data are shown as
mean SEM (n = 7 or 8). (E) Novel object recognition test (NORT). Data are shown as mean SEM (n = 8). (F) Prepulse inhibition (PPI) test. Data are shown as
mean SEM (n = 8). (G) Three-chamber social interaction test. (Left) Two-way ANOVA (glyphosate, F1,22 = 4.747, P = 0.040; stranger, F1,22 = 141.2, P < 0.001;
interaction, F1,22 = 76.77, P < 0.001). (Right) Two-way ANOVA (glyphosate, F1,22 = 9.760, P = 0.005; stranger, F1,22 = 26.75, P < 0.001; interaction, F1,22 = 33.38,
P < 0.001). Data are shown as mean SEM (n = 6 or 7). (H) Protein expression of sEH in the PFC, hippocampus, and striatum of mothers. Data are shown as
mean SEM (n = 4 or 5). (I) Protein expression of sEH in the PFC, hippocampus, and striatum from juvenile offspring (P28). Data are shown as mean SEM (n =
10). (J) Gene expression of Ephx2 mRNA in the mouse brain regions from juvenile offspring (P28). Data are shown as mean SEM (n = 8). (K) PV immu-
noreactivity in the prelimbic area (PrL) and infralimbic (IL) of mPFC. The values represent the mean SEM (n = 8; *P < 0.05, **P < 0.01, ***P < 0.001 compared
to control group by Student t test). N.S., not significant.
Pu et al.
PNAS Latest Articles
|
3 of 7

Fig. 2. Oxylipin analysis of blood and brain regions. (A) Arachidonic acid is metabolized to 8,9-EpETrE by several cytochrome P450 enzymes. Subsequently,
8,9-EpETrE is metabolized to 8,9-DiHETrE by sEH. (B) Levels of 8,9-EpETrE in the plasma, PFC, hippocampus, and striatum from juvenile offspring (P28). The
values represent the mean SEM (n = 8 to 10; *P < 0.05, **P < 0.01 compared to control group by Student t test).
Lachnospiraceae bacterium 538, and Clostridium tertium was sig- striatum from juvenile offspring after maternal glyphosate ex-
nificantly lower in the juvenile offspring after maternal glyphosate posure was higher than that of the control group. Oxylipin
exposure compared to the water-treated group (Fig. 3). In con- analysis showed a marked reduction of 8 (9)-EpETrE in the
trast, the relative abundance of Clostridium sp. Clone-1, Enter- plasma, PFC, hippocampus, and striatum from juvenile offspring
orhabdus muris, Clostridium sp. Clone-46, and Butyricimonas virosa after maternal glyphosate exposure, supporting higher levels of
was significantly higher in juvenile offspring after maternal sEH in these regions. Third, maternal glyphosate exposure caused
reduced PV immunoreactivity in the prelimbic of medial PFC in
the offspring compared to the water-treated group. Furthermore,
maternal glyphosate exposure caused significant alterations of
NMDAR-related amino acids in the blood and brain of offspring.
Fourth, maternal glyphosate exposure caused significant abnormal
composition of gut microbiota and increased levels of acetic acid
in the fecal samples from juvenile offspring. Finally, repeated
treatment with TPPU in glyphosate-treated pregnant mice from
pregnancy (E5) to weaning (P21) prevented the onset of ASD-like
behaviors (i.e., increased grooming time and social interaction
deficits) in juvenile offspring after maternal glyphosate exposure.
Collectively, these findings suggest that the sEH enzyme plays a
key role in the development of ASD-like behavioral abnormalities
in offspring after maternal glyphosate exposure, and that sEH
inhibitors may prove to be promising prophylactic or therapeutic
drugs for ASD.
In this study, we found increased expression of sEH protein in
the PFC of juvenile offspring after maternal glyphosate exposure,
consistent with our report using MIA (30). Thus, it seems that
increases in the sEH in the PFC and other regions (hippocampus
and striatum) might play a role in the behavioral and biochemical
abnormalities seen in juvenile offspring after maternal glyphosate
exposure. Previously, we reported higher levels of EPHX2 mRNA
in the postmortem brain samples from ASD patients (30). These
findings suggest that increased metabolism of EpFAs to the cor-
responding diols by increased sEH may play a role in the patho-
genesis of ASD, although further detailed studies on how
glyphosate exposure compared to the water-treated group (Fig. 3).
Furthermore, levels of acetic acid in the fecal samples of the
offspring were significantly increased after maternal glyphosate
exposure (Fig. 3). Other short-chain fatty acids including pro-
pionic acid, butyric acid, and valeric acid were not different. The
data suggest that maternal exposure to formulated glyphosate
causes abnormal composition of gut microbiota in juvenile
offspring.
Effects of TPPU on ASD-Like Behaviors in Juvenile Offspring of
Maternal Glyphosate Exposure. Water or glyphosate was given to
pregnant mice from E5 to P21. In addition, the pregnant mice
were orally administered vehicle (5 mL/kg/d) or vehicle and TPPU
(3 mg/kg/d) from E5 to P21. Behavioral tests such as grooming
test and three-chamber social interaction test were performed
from P28 to P35 (Fig. 4A). Body weight was significantly in-
creased in TPPU-treated glyphosate-exposed mothers com-
pared to vehicle-treated glyphosate exposure mothers (Fig. 4B).
Treatment with TPPU significantly ameliorated the increased
grooming time of juvenile offspring after maternal glyphosate
exposure (Fig. 4C). In the three-chamber social interaction test,
treatment with TPPU significantly improved social interaction
deficits in juvenile offspring after maternal glyphosate exposure
(Fig. 4D).
Discussion
The present results demonstrate a role of sEH in the onset of
ASD-like behaviors in murine offspring after maternal glyphosate maternal glyphosate exposure induces abnormalities in the eicos-
exposure. The major findings of the present study are as follows. anoid metabolism by sEH and behavioral abnormalities in off-
First, exposure to high levels (0.098%) of glyphosate during preg-
nancy and lactation caused ASD-like behaviors in juvenile offspring.
Second, expression of sEH protein in the PFC, hippocampus, and
spring are needed.
We found decreased levels of many EpFAs including 8 (9)-
EpETrE in the blood of juvenile offspring after maternal
4 of 7
|
www.pnas.org/cgi/doi/10.1073/pnas.1922287117
Pu et al.

Fig. 3. Composition of gut microbiota in fecal samples of juvenile offspring. (A) Histogram of microbiota at species level of offspring (P28). (B) Several
bacteria were significantly altered in the offspring after maternal glyphosate exposure. Data are shown as mean SEM (n = 10; *P < 0.05, **P < 0.01
compared to control group by Student t test).
glyphosate exposure compared to the water-treated group. Inter-
It is recognized that mechanism of action of glyphosate is to
estingly, tissue levels of 8 (9)-EpETrE, the abundant EpFA in the disrupt the shikimate pathway, which is absent from human cells.
brain, were significantly lower in the PFC, hippocampus, and However, human gut microbiomes contain the shikimate pathway,
striatum from juvenile offspring after maternal glyphosate expo-
which plays a key role in the synthesis of aromatic amino acids in
sure than those of control mice, supporting an increased activity of both plants and microbiomes (11, 46–48). Therefore, it is sug-
sEH in these brain regions. The data on 8 (9)-EpETrE are con-
sistent with our previous report using MIA model of ASD (30).
gested that exposure to glyphosate can affect gut microbiota in
humans (6, 49). In this study, we found abnormal composition of
Although the precise mechanisms underlying the relationship be- gut microbiota such as Clostridium in juvenile offspring after
tween 8 (9)-EpETrE and sEH in the brain from juvenile offspring maternal glyphosate exposure. A recent review pointed an in-
after maternal glyphosate exposure are currently unclear, it seems
teraction between Clostridium bacteria and ASD (50). In addition,
that low levels of 8 (9)-EpETrE by increased levels of sEH in the we found higher levels of acetic acid in fecal samples of juvenile
brain may be involved in behavioral abnormalities of offspring offspring after maternal glyphosate exposure. It is reported that
after maternal glyphosate exposure. By contrast, other EpFAs
were significantly higher in the brain regions of juvenile offspring those in controls (51). It seems that increased intestinal perme-
after maternal glyphosate exposure than those of the water-treated ability by acetic acid might play a role in fecal production of acetic
fecal levels of acetic acid in children with ASD were higher than
group, although tissue levels of sEH in the brain regions were acid since acetic acid plays a role in gut epithelial barrier func-
increased after maternal glyphosate exposure. Although the rea- tion (51). Given the crucial role of gut microbiota in ASD
sons underlying this discrepancy are currently unknown, it seems pathogenesis (39, 52, 53), abnormal composition of gut micro-
that multiple pathways may contribute to formation and degra- biota may be, in part, involved in the ASD-like behaviors in
dation of EpFAs in the brain regions.
offspring after maternal glyphosate exposure. At present, specific
Pu et al.
PNAS Latest Articles
|
5 of 7

A
B
70
C
Water+TPPU
300
200
100
0
0.098%Glyphosate+TPPU
Water+Vehicle
Water
***
***
0.098%Glyphosate+Vehicle
Glyphosate
60
50
40
30
*
*
N.S.
**
**
*
20
0
Verhicle
TPPU
E5 E7 E9 E11 E13 E15 E17
D
300
300
200
100
0
Stranger 1
Stranger 2
Stranger 1
Empty
***
***
***
***
N.S.
N.S.
200
100
0
***
***
Vehicle
TPPU
Vehicle
TPPU
Glyphosate
Vehicle
TPPU
Water
Vehicle
TPPU
Water
Glyphosate
Fig. 4. Effects of TPPU on ASD-like behavioral abnormalities in juvenile offspring after maternal glyphosate exposure. (A) Schedule of treatment and be-
havioral tests. Water or glyphosate (0.098%) was given to pregnant mice. Vehicle (5 mL/kg/d) or TPPU (3 mg/kg/d) was administered orally to pregnant mice
from E5 to P21. Subsequently, all mice received normal water. Grooming test and three-chamber social interaction test were performed from P28 to P35. (B)
Change of body weight of mothers (n = 5 or 6). Two-way ANOVA (glyphosate, F1,17 = 7.66, P = 0.013; TPPU, F1,17 = 9.14, P = 0.008; interaction, F1,17 = 1.59, P =
0.225). (C) Grooming test. Treatment with TPPU significantly attenuated the increased grooming time in juvenile offspring after maternal glyphosate ex-
posure. Two-way ANOVA (glyphosate, F1,36 = 14.19, P = 0.001; TPPU, F1,36 = 25.34, P < 0.001; interaction, F1,36 = 11.31, P = 0.002). Data are shown as mean
SEM (n = 10; ***P < 0.01 compared to glyphosate + vehicle group). (D) Three-chamber social interaction test. (Left) Three-way ANOVA (glyphosate, F1,56
=
9.948, P = 0.003; TPPU, F1,56 = 0.301, P = 0.585; stranger, F1,56 = 135.27, P < 0.001; interaction [glyphosate × TPPU], F1,56 = 0.845, P = 0.362; interaction
[glyphosate × stranger], F1,56 = 38.61, P < 0.001; interaction [TPPU × stranger], F1,56 = 3.593, P = 0.063; interaction [glyphosate × TPPU × stranger], F1,56 = 11.06,
P = 0.002). (Right) Three-way ANOVA (glyphosate, F1,56 = 4.168, P = 0.046; TPPU, F1,56 = 21.96, P < 0.001; stranger, F1,56 = 104.28, P < 0.001; interaction
[glyphosate × TPPU], F1,56 = 1.902, P = 0.173; interaction [TPPU × stranger], F1,56 = 3.870, P = 0.054; interaction [glyphosate × stranger], F1,56 = 13.10, P = 0.001;
interaction [glyphosate × TPPU × stranger], F1,56 = 19.69, P < 0.001). Data are shown as mean SEM (n = 8; ***P < 0.01). N.S., not significant.
bacteria that can cause ASD were not yet identified. Therefore,
Although the current animal data do not necessarily translate to
further study on the role of gut microbiota on glyphosate-induced humans, further study connecting animal data with the find-
ASD is needed.
ings from epidemiological studies is needed to identify the de-
tailed mechanisms of action of glyphosate exposure for ASD
pathogenesis.
In conclusion, this study suggests that maternal exposure to
high levels of formulated glyphosate might play a role in the
etiology of ASD-like behaviors in murine offspring through in-
creased activity of sEH in the brain, and sEH inhibitors could
be a useful tool to dissect the mechanism. However, a recent
comprehensive review on human exposure to glyphosate indicates
Maternal exposure to 0.098% glyphosate causes ASD-like
behaviors and abnormal composition of gut microbiota in juve-
nile offspring. Although it is exceptionally unlikely that such
exposure could be reached during human pregnancy, mater-
nal exposure to high levels of technical glyphosate could have
detrimental side effects in offspring. A cohort study on mea-
surement of blood (or urine) levels of glyphosate in pregnant
mothers who have offspring with or without ASD is of interest.
6 of 7
|
www.pnas.org/cgi/doi/10.1073/pnas.1922287117
Pu et al.

Data Availability. All data in the paper are included in the dataset of the
SI Appendix.
that human exposure approaching these levels is exceptionally
unlikely (54).
Materials and Methods
ACKNOWLEDGMENTS. This study was supported by the Japan Society for
the Promotion of Science (JSPS) (to K.H., 17H04243), Japan Agency for
Medical Research and Development (to K.H., JP19dm0107119), the National
Institute of Environmental Health Sciences (NIEHS) River Award R35
ES030443-01 (to B.D.H.), and NIEHS Superfund Program P42 ES004699
(to B.D.H.).
Details of the experimental protocols, including animals, maternal glyph-
osate exposure, measurement of glyphosate in the blood, oxylipin analysis,
Western blot analysis, RT-PCR, behavioral tests, treatment of TPPU, immuno-
histochemistry, measurement of amino acids, gut microbiota analysis, and sta-
tistical analysis, are given in the SI Appendix.
1. M. C. Lai, M. V. Lombardo, S. Baron-Cohen, Autism. Lancet 383, 896–910 (2014).
2. C. Lord, M. Elsabbagh, G. Baird, J. Veenstra-Vanderweele, Autism spectrum disorder.
Lancet 392, 508–520 (2018).
3. S. N. Hansen, D. E. Schendel, E. T. Parner, Explaining the increase in the prevalence of
autism spectrum disorders: The proportion attributable to changes in reporting
practices. JAMA Pediatr. 169, 56–62 (2015).
4. J. Baio et al., Prevalence of autism spectrum disorder among children aged 8 years–
autism and developmental disabilities monitoring network, 11 sites, United States,
2014. MMWR Surveill. Summ. 67, 1–23 (2018).
5. J. Hallmayer et al., Genetic heritability and shared environmental factors among twin
pairs with autism. Arch. Gen. Psychiatry 68, 1095–1102 (2011).
6. L. A. Sealey et al., Environmental factors in the development of autism spectrum
disorders. Environ. Int. 88, 288–298 (2016).
7. S. K. Sagiv et al., Prenatal organophosphate pesticide exposure and traits related to
autism spectrum disorders in a population living in proximity to agriculture. Environ.
Health Perspect. 126, 047012 (2018).
8. J. Y. Kim et al., Environmental risk factors and biomarkers for autism spectrum dis-
order: An umbrella review of the evidence. Lancet Psychiatry 6, 590–600 (2019).
9. S. M. Bradberry, A. T. Proudfoot, J. A. Vale, Glyphosate poisoning. Toxicol. Rev. 23,
159–167 (2004).
10. L. D. Kier, D. J. Kirkland, Review of genotoxicity studies of glyphosate and glyphosate-
based formulations. Crit. Rev. Toxicol. 43, 283–315 (2013).
11. A. Samsel, S. Seneff, Glyphosate, pathways to modern diseases II: Celiac sprue and
gluten intolerance. Interdiscip. Toxicol. 6, 159–184 (2013).
12. N. L. Swanson, A. Leu, J. Abrahamson, B. Wallet, Genetically engineered crops,
glyphosate and the deterioration of health in the United States of America. J. Org.
Syst. 9, 6–37 (2014).
13. A. Samsel, S. Seneff, Glyphosate, pathways to modern diseases III: Manganese, neu-
rological diseases, and associated pathologies. Surg. Neurol. Int. 6, 45 (2015).
14. O. S. von Ehrenstein et al., Prenatal and infant exposure to ambient pesticides and
autism spectrum disorder in children: Population based case-control study. BMJ 364,
l962 (2019).
15. M. L. Estes, A. K. McAllister, Maternal immune activation: Implications for neuro-
psychiatric disorders. Science 353, 772–777 (2016).
16. H. Y. Jiang et al., Maternal infection during pregnancy and risk of autism spectrum
disorders: A systematic review and meta-analysis. Brain Behav. Immun. 58, 165–172
(2016).
17. M. Careaga, T. Murai, M. D. Bauman, Maternal immune activation and autism spec-
trum disorder: From rodents to nonhuman and human primates. Biol. Psychiatry 81,
391–401 (2017).
18. O. Zerbo et al., Association between influenza infection and vaccination during
pregnancy and risk of autism spectrum disorder. JAMA Pediatr. 171, e163609 (2017).
19. K. Hashimoto, Recent advances in the early intervention in schizophrenia: Future
direction from preclinical findings. Curr. Psychiatry Rep. 21, 75 (2019).
20. A. S. Brown, U. Meyer, Maternal immune activation and neuropsychiatric illness: A
translational research perspective. Am. J. Psychiatry 175, 1073–1083 (2018).
21. C. Morisseau, B. D. Hammock, Epoxide hydrolases: Mechanisms, inhibitor designs, and
biological roles. Annu. Rev. Pharmacol. Toxicol. 45, 311–333 (2005).
22. J. D. Imig, B. D. Hammock, Soluble epoxide hydrolase as a therapeutic target for
cardiovascular diseases. Nat. Rev. Drug Discov. 8, 794–805 (2009).
23. C. Morisseau, B. D. Hammock, Impact of soluble epoxide hydrolase and epox-
yeicosanoids on human health. Annu. Rev. Pharmacol. Toxicol. 53, 37–58 (2013).
24. K. Hashimoto, Role of soluble epoxide hydrolase in metabolism of PUFAs in psychi-
atric and neurological disorders. Front. Pharmacol. 10, 36 (2019).
25. Q. Ren et al., Gene deficiency and pharmacological inhibition of soluble epoxide
hydrolase confers resilience to repeated social defeat stress. Proc. Natl. Acad. Sci.
U.S.A. 113, E1944–E1952 (2016).
29. Q. Ren et al., Soluble epoxide hydrolase plays a key role in the pathogenesis of
Parkinson’s disease. Proc. Natl. Acad. Sci. U.S.A. 115, E5815–E5823 (2018).
30. M. Ma et al., Key role of soluble epoxide hydrolase in the neurodevelopmental dis-
orders of offspring after maternal immune activation. Proc. Natl. Acad. Sci. U.S.A. 116,
7083–7088 (2019).
31. K. Hashimoto, Understanding the link between maternal infections and neuro-
developmental disorders in offspring: The role of abnormalities in metabolism of
polyunsaturated fatty acids. Brain Behav. Immun. 81, 4–5 (2019).
32. J. Atone, K. Wagner, K. Hashimoto, B. D. Hammock, Cytochrome P450 derived ep-
oxidized fatty acids as a therapeutic tool against neuroinflammatory diseases. Pros-
taglandins Other Lipid Mediat. 147, 106385 (2020).
33. A. Shinohe et al., Increased serum levels of glutamate in adult patients with autism.
Prog. Neuropsychopharmacol. Biol. Psychiatry 30, 1472–1477 (2006).
34. C. Shimmura et al., Alteration of plasma glutamate and glutamine levels in children
with high-functioning autism. PLoS One 6, e25340 (2011).
35. Z. Zheng, T. Zhu, Y. Qu, D. Mu, Blood glutamate levels in autism spectrum disorder: A
systematic review and meta-analysis. PLoS One 11, e0158688 (2016).
36. M. Wang et al., Alterations in gut glutamate metabolism associated with changes in
gut microbiota composition in children with autism spectrum disorder. mSystems 4,
e00321-18 (2019).
37. A. Tomova et al., Gastrointestinal microbiota in children with autism in Slovakia.
Physiol. Behav. 138, 179–187 (2015).
38. H. E. Vuong, E. Y. Hsiao, Emerging roles for the gut microbiome in autism spectrum
disorder. Biol. Psychiatry 81, 411–423 (2017).
39. F. Liu et al., Altered composition and function of intestinal microbiota in autism
spectrum disorders: A systematic review. Transl. Psychiatry 9, 43 (2019).
40. M. Xu, X. Xu, J. Li, F. Li, Association between gut microbiota and autism spectrum
disorder: A systematic review and meta-analysis. Front. Psychiatry 10, 473 (2019).
41. T. E. Rose et al., 1-Aryl-3-(1-acylpiperidin-4-yl)urea inhibitors of human and murine
soluble epoxide hydrolase: Structure-activity relationships, pharmacokinetics, and
reduction of inflammatory pain. J. Med. Chem. 53, 7067–7075 (2010).
42. A. I. Ostermann et al., Oral treatment of rodents with soluble epoxide hydrolase in-
hibitor 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl)urea (TPPU): Bioavailability,
resulting drug levels and modulation of oxylipin pattern. Prostaglandins Other Lipid
Mediat. 121, 131–137 (2015).
43. H. Ji, L. Xu, Z. Wang, X. Fan, L. Wu, Differential microRNA expression in the prefrontal
cortex of mouse offspring induced by glyphosate exposure during pregnancy and
lactation. Exp. Ther. Med. 15, 2457–2467 (2018).
44. N. Yu et al., Circular RNA expression profiles in hippocampus from mice with perinatal
glyphosate exposure. Biochem. Biophys. Res. Commun. 501, 838–845 (2018).
45. G. M. Williams, R. Kroes, I. C. Munro, Safety evaluation and risk assessment of the
herbicide Roundup and its active ingredient, glyphosate, for humans. Regul. Toxicol.
Pharmacol. 31, 117–165 (2000).
46. K. M. Herrmann, L. M. Weaver, The shikimate pathway. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 50, 473–503 (1999).
47. H. Maeda, N. Dudareva, The shikimate pathway and aromatic amino acid biosynthesis
in plants. Annu. Rev. Plant Biol. 63, 73–105 (2012).
48. A. Samsel, S. Seneff, Glyphosate’s suppression of cytochrome P450 enzymes and
amino acid biosynthesis by the gut microbiome: Pathways to modern diseases. En-
tropy 15, 1416–1463 (2013).
49. L. Rueda-Ruzafa, F. Cruz, P. Roman, D. Cardona, Gut microbiota and neurological
effects of glyphosate. Neurotoxicology 75, 1–8 (2019).
50. I. Argou-Cardozo, F. Zeidán-Chuliá, Clostridium bacteria and autism spectrum con-
ditions: A systematic review and hypothetical contribution of environmental glyph-
osate levels. Med. Sci. 6, 29 (2018).
51. L. Wang et al., Elevated fecal short chain fatty acid and ammonia concentrations in
children with autism spectrum disorder. Dig. Dis. Sci. 57, 2096–2102 (2012).
52. J. F. Cryan et al., The micribiota-gut-brain axis. Physiol. Rev. 99, 1877–2013 (2019).
53. E. Sherwin, S. R. Bordenstein, J. L. Quinn, T. G. Dinan, J. F. Cryan, Microbiota and the
social brain. Science 366, eaar2016 (2019).
26. K. Hashimoto, Soluble epoxide hydrolase: A new therapeutic target for depression.
Expert Opin. Ther. Targets 20, 1149–1151 (2016).
27. K. M. Wagner, C. B. McReynolds, W. K. Schmidt, B. D. Hammock, Soluble epoxide
hydrolase as a therapeutic target for pain, inflammatory and neurodegenerative
diseases. Pharmacol. Ther. 180, 62–76 (2017).
54. K. R. Solomon, Estimated exposure to glyphosate in humans via environmental, oc-
cupational, and dietary pathways: An updated review of the scientific literature. Pest
Manag. Sci., 10.1002/ps.5717 (2019).
28. W. Swardfager et al., Metabolic/inflammatory/vascular comorbidity in psychiatric
disorders; soluble epoxide hydrolase (sEH) as a possible new target. Neurosci. Biobehav.
Rev. 87, 56–66 (2018).
Pu et al.
PNAS Latest Articles
|
7 of 7