Dietary intake of glucoraphanin during pregnancy and lactation prevents the behavioral abnormalities in the offspring after maternal immune activation

Abstract Aim Epidemiological data suggest that maternal immune activation (MIA) plays a role in the etiology of neuropsychiatric disorders including autism spectrum disorder (ASD) and schizophrenia. However, there is no prophylactic nutrition that can prevent the onset of neurodevelopmental disorders in offspring after MIA. The aim of this study was undertaken to examine whether dietary intake of glucoraphanin (GF: the precursor of a natural anti‐inflammatory compound sulforaphane) can prevent the onset of behavioral abnormalities in offspring after MIA. Methods One percent of GF food pellet or normal food pellet was given into female mice during pregnancy and lactation (from E5 to P21). Saline (5 mL/kg/d) or poly(I:C) (5 mg/kg/d) was injected into pregnant mice from E12 to E17. Behavioral tests and immunohistochemistry of parvalbumin (PV) were performed in male offspring. Results Dietary intake of GF during pregnancy and lactation prevented cognitive deficits and social interaction deficits in the juvenile offspring after MIA. Furthermore, dietary intake of GF during pregnancy and lactation prevented cognitive deficits in the adult offspring after MIA. Moreover, dietary intake of GF prevented the reduction of PV immunoreactivity in the medial prefrontal cortex of adult offspring after MIA. Conclusion These data suggest that dietary intake of GF during pregnancy and lactation could prevent behavioral abnormalities in offspring after MIA.


| 269
FUJITA eT Al. maternal infection during pregnancy increases the risk of these disorders in offspring. 3,4 Importantly, there are accumulating interests in the early prevention by anti-inflammatory compounds. [5][6][7] However, there are no anti-inflammatory compounds that can be used in the early intervention for pregnant women with MIA.
Polyriboinosinic-polyribocytidilic acid (poly[I:C]), a Toll-like receptor 3 agonist, is widely used as an animal model of MIA. 2,[8][9][10][11][12][13][14] The Nuclear factor erythroid 2-related 2 (Nrf2) is a transcription factor which plays a crucial role in attenuating oxidative stress and inflammation. 15,16 Sulforaphane (SFN) is a naturally occurring compound with potent anti-inflammatory effects. In addition, glucoraphanin (GF), glucosinolate precursor of SFN, is found in cruciferous vegetables. 17 SFN attenuated abnormal behaviors in rodents after the administration of phencyclidine (PCP). 18 Furthermore, the supplementation of GF during juvenile and adolescent stages prevented the behavioral abnormalities in adult mice after repeated PCP administration 19 or MIA. 12 These findings suggest that supplementation of GF may have prophylactic effects for neuropsychiatric disorders such as schizophrenia. 7  This study was undertaken to investigate whether dietary intake of GF food pellets during pregnancy and lactation could attenuate the development of abnormal behaviors in juvenile and adult offspring after MIA. Furthermore, we performed parvalbumin (PV)-immunohistochemistry since the reduction of PV immunoreactivity in the mPFC is associated with neuropsychiatric disorders. 14,20,21

| Animals
Pregnant ddY mice (embryo at the 5th day [E5], 9-10 weeks old) were obtained from Japan SLC Inc. Pregnant mice were caged into individually clear polycarbonate cage (22.5 × 33.8 × 14.0 cm) under a controlled 12/12h light-dark cycle (lights on from 07:00 am to 07:00 pm), with room temperature at 23 ± 1°C and humidity at 55 ± 5%. All mice had ad libitum access to water and food pellets. The experimental procedure using animals was approved by the Chiba University Laboratory Animal Care and Use Committee (permission number: 28-272).
The schedule of prenatal poly(I:C) treatment was performed as reported previously. [8][9][10][11][12] The pregnant mice were injected intraperitoneally (i.p.) for six consecutive days from E12 to E17 with poly(I:C) (5.0 mg/kg/d, Sigma-Aldrich Co. Ltd.) or an equivalent saline (5 mL/kg). The male offspring were separated from their mothers at P21, and mice were caged each three to five in the groups.

| Statistical analysis
All data are shown as mean ± standard error of the mean (SEM). The data were analyzed by two-way analysis of variance (ANOVA), followed post hoc Bonferroni test. Significance for results was set at P < .05.

| Effects of dietary intake of 0.1% GF food pellets during pregnancy and lactation on behavioral abnormalities in the juvenile offspring after MIA
We performed two behavioral tests (NORT, and 3-chamber social interaction test) in juvenile offspring after MIA. Behavioral tests of juvenile offspring were performed at P28-P35 after prenatal poly(I:C) injections ( Figure 1A). In NORT, there was no difference between the four groups during the training session ( Figure 1B). However, during the retention session, there was significant change among the four groups ( Figure 1B). The exploratory preference of the poly(I:C) + normal food pellet group was significantly lower than that of the F I G U R E 1 Effects of 0.1% GF food pellet on cognitive deficits and social interaction deficits in the juvenile offspring after prenatal poly(I:C) exposure. A, Schedule of treatment and behavioral tests. Saline (5.0 mL/kg/d) or poly(I:C) (5.0 mg/kg/day from E12 to E17) was injected into pregnant mice. Normal food pellets or 0.1% GF food pellets were given to pregnant mice from E5 to P21. Subsequently, normal food pellets were given to all mice from P21. Behavioral tests such as novel object recognition test (NORT) and 3-chamber social interaction test were performed from P28 to P35. (B): NORT: There was no difference (two-way ANOVA: poly(I:C): F 1,39 = 0.122, P = .729, GF: F 1,39 = 0.073, P = .789, interaction: F 1,39 = 0.003, P = .954) between the four groups in the training session. In the retention session, twoway ANOVA showed the results (poly(I:C): F 1,39 = 37.73, P < .001, GF: F 1,39 = 2.039, P = .161, interaction: F 1,39 = 7.310, P = .010) between the four groups. In the retention test, the exploratory preference of poly(I:C) + GF food group was significantly higher than poly(I:C) + normal food group. **P < .05, ***P < .001 compared with poly(I:C) + normal food group. The value is expressed as the mean ± SEM (n = 10 or 11). saline + normal food pellet group. Furthermore, the exploratory preference of the GF food pellet + poly(I:C) group was significantly higher than that of the normal food pellet + poly(I:C) group ( Figure 1B).
In the three-chamber test, juvenile offspring after MIA showed social interaction deficits compared to the control group ( Figure 1C).
Dietary intake of 0.1% GF food pellet significantly improved social interaction deficits in juvenile offspring after MIA ( Figure 1C). The data suggest that MIA causes ASD-like cognitive and social interaction deficits in juvenile offspring, and that dietary intake of 0.1% GF during pregnancy and lactation could prevent the onset of ASD-like behavioral abnormalities in juvenile offspring after MIA.

PrL of mPFC IL of mPFC
higher than that in the poly(I:C) + normal food pellet group ( Figure 3B,C).
These findings suggest that supplementation of 0.1% GF food pellets during pregnancy and lactation prevented the cognitive deficits and the reduction of PV immunoreactivity in the PrL of the mPFC in adult offspring after MIA.

| D ISCUSS I ON
Here, we found that dietary intake of 0.1% GF food pellets during pregnancy and lactation prevented ASD-and schizophrenia-like behavioral abnormalities and reduction of PV immunoreactivity in the PrL of the mPFC in offspring after MIA. Therefore, it is likely that supplementation with GF-rich food in pregnant women with MIA (ie, higher inflammation) could have prophylactic effects on the development of neurodevelopmental disorders in offspring.
We found cognitive deficits of juvenile offspring after MIA, consistent with previous reports. [9][10][11][12] Given the role of cognitive impairment in ASD patients and subjects with a high risk for psychosis, 25 it is likely that cognitive deficits may be a core behavioral deficit in juvenile offspring after MIA. Interestingly, dietary intake of 0.1% GF food pellet during pregnancy and lactation could block cognitive and social interaction deficits in juvenile offspring after MIA.
In this study, we also found reduction of PV immunoreactivity in the PrL, but not IL, of mPFC at adult offspring after MIA, consistent with the previous findings. Previously, we reported that dietary intake of 0.1% GF food pellet during juvenile and adolescence blocked the onset of cognitive deficits and reduction of PV immunoreactivity in the mPFC after repeated PCP administration. 19 In addition, dietary intake of 0.1% GF food pellet during juvenile and adolescence blocked the onset of cognitive deficits and reduction of PV immunoreactivity in the mPFC of adult offspring after MIA. 12  This manuscript has limitation. In this study, we did not investigate the tissue levels of GF and its metabolite SFN in the fetal brain. Therefore, it is unknown whether GF or SFN can affect directly altered cortical development of fetal brain after MIA. Further detailed study is needed.
In conclusion, the present data suggest that dietary intake of 0.1% GF during pregnancy and lactation could prevent the behavioral abnormalities in offspring after MIA. Finally, supplementation of GF (or SFN)-rich vegetables in pregnant women with MIA or pregnant women at high risk for psychosis might reduce the risk of onset of neurodevelopmental disorders in offspring. Research Foundation, Japan (to KH).

Dr Hashimoto received speaker's honoraria from Murakami Farm
(Tokyo, Japan) which sells sulforaphane-rich vegetable. Drs. Ayumi Hirai, Shigenori Suzuki, and Hiroyuki Suganuma are employee of KAGOME which sells glucoraphanin-related products as the supplement. The other authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N S
KH is responsible for the design of the research and experiment and supervised the experimental analyses. KH wrote the paper. YF, AF, and TI performed behavioral experiments and immunohistochemistry. YF analyzed the data. AH, SS, and HS provided 0.1% GF food pellet. All authors read and approved this paper.

A N I M A L S TU D I E S
All animal experiments were approved by the Animal Care and Use Committee of Chiba University.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available in the Figure S1-S3 of this article.