Effects Of Exposure To Lipopolysaccharides During Pregnancy On Age-Related Changes Of Behaviors And Hippocampal H4K8 Acetylation In The CD-1 Mice

Background: Along with aging, there are obviously changes in many aspects of behaviors, including emotion, cognition, sensory-motor and social activities, in humans and rodent. A lot of factors(such as gender, pregnancy infection, stress, environmental toxins, etc.) may influence age-related behavior change. Emerging data indicate that the normal reproductive experience or motherhood can mitigates age-related learning and memory impairment. Lipopolysaccharides(LPS) are the cell wall components of Gram-negative bacteria, and can be recognized by activating immune system and induce proinflammatory reactions. So, the LPS are commonly employed to mimic bacterial infections in animal experiments. Studies also demonstrate that the pregnant rats exhibit much more intense and persistent inflammation than the non-pregnant rats,when exposed to even a very low-dose LPS. A set of previous studies demonstrated that exposure to adverse factors during pregnancy can significantly affect adults offspring learning and memory ability, anxiety level, or even accelerated offspring aging process.However, our knowledge about whether gestational inflammations affect the mother-self brain aging is limited. A number of studies indicate that age-related learning and memory impairment correlates with remodeling of chromatin via histone acetylation. In particular, H4 acetylation on K8 site(H4K8ac) has recently been implicated with the formation of long-term memoriesObjective: To explore whether exposure to LPS during late pregnancy can affect the age-related behavioral changes and H4K8 ac level in the older hippocampus.Methods: The pregnant mice were randomly divided into three groups, consisting of the high-dose LPS group(HD-LPS), the low-dose LPS group(LD-LPS) and the saline group(O-con). The HD-LPS or LD-LPS pregnant mice respectively received an intraperitoneal injection of 50 or 25 μg/kg LPS daily from gd 15 to gd 17. The O-con mice were intraperitoneally injected the same volume of normal saline at the same period. After pregnancy, female mice were separated from their offspring, with water and food were available ad libitum. In the 15 months of age, the mice went into the behavioral tests, and a group of 6-month-old mice was added as the young control group(Y-con). An immunohistochemical method was utilized to detect H4K8 ac level in the hippocampus.Results: 1) Open flied: The O-con exhibited significantly longer latency and peripheral time and less squares crossed than the Y-con(Ps < 0.05). The LPS-treatment effects on the peripheral time and squares crossed were significant. The HD-LPS had longer peripheral time and less squares crossed, and the LD-LPS had less squares crossed, than the O-con(Ps < 0.05). Black-white alley: The O-con showed significantly longer latency and black-alley time than the Y-con(Ps < 0.05). The HD-LPS had longer latency and black-alley time than the O-con(Ps < 0.05). Elevated plus maze: The O-con showed significantly longer latency and less number of entries to the open arm(s)than the Y-con(Ps < 0.05). There were insignificant LPS-treatment effects on the latency, number of entries to the open arm(s) and time spent on the open arm(s). 2)RAWM: In the learning phase, the latency and number of errors progressively declined over days for control mice combined, indicating these mice were able to learn the task.The rm-ANOVA results showed that the O-con had significantly longer latency and more errors than the Y-con(Ps < 0.05). The latency and errors also progressively declined daily for all 15-month-old mice combined. The LPS treatment significantly affected the latency and errors. The post hoc analysis showed that only the HD-LPS had significantly longer latency and more errors than the O-con(Ps < 0.05). In memory phase, the latency and number of errors were progressively declined with days for all control mice combined. The O-con mice had significantly longer latency and more errors than the Y-con mice(Ps < 0.05). The latency was also progressively declined daily for all older mice combined. There were significant differences in the latency and number of errors among three older groups(Ps < 0.05). The post-hoc analysis showed the significantly longer latency and more errors only occurred in the HD-LPS,compared to the O-con(Ps < 0.05). 3) H4K8 ac in different hippocampal layers.Compared to the Y-con, the O-con had significant lower levels of H4K8 ac in the DG,CA1 and CA3(Ps < 0.05). The LPS-treatment significantly affected the H4K8 ac levels in the DG, CA1 and CA3. Further, the HD-LPS had lower levels of H4K8 ac in the DG,CA1 and CA3 subregions than the O-con(Ps < 0.05). So did the LD-LPS in the CA1 and CA3 subareas(Ps < 0.05).Conclusions: LPS-exposure during pregnancy intensified the age-related changes of behaviors in the CD-1 mice and the reduced H4K8 ac in aging hippocampus in a dose-dependent manner.