The principal finding of our study was that intrauterine inflammation at term results in fetal brain injury as demonstrated by increase in proinflammatory cytokines in fetal brain and neuronal damage with delayed neurotoxicity. In these studies, intrauterine inflammation did not result in changes associated with WMD such as astrogliosis (increase in GFAP marker) or decrease in pro-oligodendrocyte marker.
Our study underlines the importance of neuronal injury in response to intrauterine inflammation as a unifying mechanism, leading to neurodegenerative and neurobehavioral disorders from prenatal exposure to inflammation. Neuronal injury, in this study, is characterized by an abnormal cytoskeletal formation, based on the decreased MAP2 mRNA levels and protein levels, and a decrease in dendritic arbor. Abnormalities in cytoskeletal structure and neuronal arborization have been shown to impact synaptic connectivity and result in neurodegenerative and neurobehavioral disorders.28–32
Specifically, dendritic arborization has been found to play a role in neurobehavioral disorders such as the Rett syndrome and autism.33–35
The cumulative evidence suggests that neurodegenerative and neurobehavioral disorders are associated with cytoskeletal alterations in neurons that, in turn, have an altered synaptic connectivity. The response of the fetal brain to intrauterine inflammation may have a similar implication.
In addition to the observed neuronal injury, our results suggest a delayed neurotoxicity as LPS-exposed fetal neurons exhibited delayed changes in neuronal viability. The presence of delayed neurotoxicity indicates that neurons are continuously affected by the in vivo insult and the manifestations of such injury may not be acute. We hypothesize that the neuronal changes and delayed neurotoxicity observed in this model may be mechanistically responsible for neurobehavioral deficits, such as CP, present in offspring exposed to chorioamnionitis at term. As there are no proposed mechanisms of neuronal injury in the setting of chorioamnionitis at term, these data are crucial for directing new research and therapeutic directions for neuroprophylaxis in a setting of chorioamnionitis at term.
While animal models of fetal and neonatal brain injury following an inflammation have inherent limitations, such studies are not possible in humans. Animal models have been designed to elucidate the effect of inflammation on fetal and neonatal brain development. 17–19,36–38
However, these are either systemic models of inflammation with intraperitoneal infusion of LPS36,38
or these models concentrate on preterm brain development.18,19,37
Chorioamnionitis represents a local, intrauterine infection and is unique from the clinical scenarios observed with systemic infections, such as pyelonephritis and pneumonia. Therefore, the use of an appropriate model which mimics intrauterine inflammation is of paramount importance in order to investigate the effects of in utero inflammation on the fetal brain. Hence, the utilization of such animal model represents a strength of our study. While neuronal development of rodents is similar to humans, 1 of the possible limitations is that at this gestational age, we could only evaluate pro-oligodendrocyte markers and not the direct changes in oligodendrocytes since the maturation of the oligodendrocytes in mice is not complete until postnatal day 21.
It is noted that this model is not meant to be a specific model of CP. The model is used to accurately mimic fetal exposure to inflammation as occurs in most human cases of term chorioamnionitis. As such, we can explore the effect of intrauterine inflammation on brain development. Future work will need to determine whether the exposed offspring will have motor deficits that are similar to CP. Recent reports using a rabbit model of intrauterine inflammation provide biological plausibility that inflammatory exposure to a rodent fetus will elicit motor deficits.39
In this rabbit model, the prenatal administration of endotoxin results in behavioral abnormalities that are similar to movement disorders observed in CP.
White matter damage, including astrogliosis, and the changes in the pro-oligodendrocyte precursors have been studied extensively in other models.17–19,36,38,40
However, while WMD may be sufficient for adverse neurobehavioral outcomes, it may not be necessary as CP and other adverse neurobehavioral outcomes are known to occur in the absence of WMD.41–45
Our study is the first, to date, demonstrating an effect of intrauterine inflammation at term on neuronal morphology and viability without an overt changes in glial markers (GFAP and PLP/DM20) indicating WMD. It is possible that changes in white mater occur later in the pathogenesis of term brain injury and neurons are the primary site of initial injury.
These studies may provide a possible explanation for the observed long-term adverse neurobehavioral outcomes after exposure to inflammation at term. Further investigation of mechanisms involved in neuronal injury and viability is critical in order to recognize the possibilities for novel therapeutic strategies.