Adult brain and behavioral pathological markers of prenatal immune challenge during early/middle and late fetal development in mice

Abstract

Maternal infection during pregnancy increases the risk for neurodevelopmental disorders such as schizophrenia and autism in the offspring. This association appears to be critically dependent on the precise prenatal timing. However, the extent to which distinct adult psychopathological and neuropathological traits may be sensitive to the precise times of prenatal immune activation remains to be further characterized. Here, we evaluated in a mouse model of prenatal immune challenge by the viral mimic, polyriboinosinic–polyribocytidilic acid (PolyIC), whether prenatal immune activation in early/middle and late gestation may influence the susceptibility to some of the critical cognitive, pharmacological, and neuroanatomical dysfunctions implicated in schizophrenia and autism. We revealed that PolyIC-induced prenatal immune challenge on gestation day (GD) 9 but not GD17 significantly impaired sensorimotor gating and reduced prefrontal dopamine D1 receptors in adulthood, whereas prenatal immune activation specifically in late gestation impaired working memory, potentiated the locomotor reaction to the NMDA-receptor antagonist dizocilpine, and reduced hippocampal NMDA-receptor subunit 1 expression. On the other hand, potentiation of the locomotor reaction to the dopamine-receptor agonist amphetamine and reduction in Reelin- and Parvalbumin-expressing prefrontal neurons emerged independently of the precise times of prenatal immune challenge. Our findings thus highlight that prenatal immune challenge during early/middle and late fetal development in mice leads to distinct brain and behavioral pathological symptom clusters in adulthood. Further examination and evaluation of in utero immune challenge at different times of gestation may provide important new insight into the neuroimmunological and neuropathological mechanisms underlying the segregation of different symptom clusters in heterogeneous neuropsychiatric disorders such as schizophrenia and autism.

Introduction

Maternal infection during pregnancy is an environmental risk factor for the offspring to develop severe neuropsychiatric disorders in later life, including schizophrenia (Brown and Susser, 2002), autism (Chess, 1971, Arndt et al., 2005), and cerebral palsy (Dammann et al., 2002). However, the strength of these associations appears to be dependent on the exact timing of the prenatal infectious events (Juul-Dam et al., 2001, Brown et al., 2004, Babulas et al., 2006). Recent investigations in rodents have also provided initial experimental evidence that the effects of prenatal immune challenge between early/middle and late gestation periods are dissociable in terms of the postnatal neuropathological and behavioral signs (reviewed in Meyer et al., 2007c). In the mouse species, prenatal immunological stimulation specifically in early/middle gestation leads to impaired selective associative learning (Meyer et al., 2006a) and suppressed spatial exploration (Meyer et al., 2006b) in adulthood, whereas prenatal immune challenge in late but not early/middle gestation results in the emergence of perseverative behavior (Meyer et al., 2006b). Furthermore, maternal immune activation during late but not early/middle pregnancy in rats leads to a disruption of spatial reference memory (Samuelsson et al., 2006) and sensorimotor gating (Fortier et al., 2007) in the adult offspring. These initial findings highlight that the precise times of prenatal immune challenge may determine the susceptibility to distinct behavioral and/or cognitive pathological symptoms cluster emerging in adult life.

In the present study, we sought further evidence for this hypothesis by extending the comparison of prenatal immune challenge in early/middle and late gestation in mice to other cognitive and pharmacological tests, including prepulse inhibition (PPI) of the acoustic startle reflex, spatial working memory, and sensitivity to acute treatment with the indirect dopamine-receptor agonist, amphetamine (AMPH) and the non-competitive N-methyl-d-aspartate (NMDA)-receptor antagonist, dizocilpine (MK-801). Deficits in sensorimotor gating and working memory have been noticed in a number of neuropsychiatric disorders with a putative developmental origin, including schizophrenia (Goldman-Rakic, 1994, Braff et al., 2001) and autism (Perry et al., 2007, Steele et al., 2007). Furthermore, enhanced susceptibility to dopaminergic stimulation by acute AMPH treatment has been linked especially to the positive symptoms of schizophrenia (Laruelle et al., 1996, Laruelle et al., 1999), whereas administration with NMDA-receptor antagonists may mimic both positive and negative (Krystal et al., 1994, Lahti et al., 2001), as well as the cognitive (Malhotra et al., 1996, Morgan et al., 2004) deficits of this disorder. Hence, extending the comparison of prenatal immune challenge in early/middle and late gestation to the tests of PPI, working memory and sensitivity to psychostimulant drugs may provide important insight as to whether the precise times of prenatal immune activation may influence the susceptibility to some of the critical cognitive and pharmacological dysfunctions implicated in neuropsychiatric disorders of neurodevelopmental origin, including schizophrenia and autism.

The reported dissociable behavioral and cognitive outcomes between early/middle and late prenatal immune activation (Meyer et al., 2006a, Meyer et al., 2006b, Samuelsson et al., 2006, Fortier et al., 2007) may further indicate that prenatal immune challenge at distinct prenatal times can differentially compromise the integrity of specific neural circuitry in adulthood. To test this hypothesis directly, we also investigated whether maternal immune activation in early/middle and late gestation may lead to distinct neuropathological profiles in the adult offspring. In order to capture possible effects of the prenatal immunological manipulations at multiple neurotransmitter systems, we analyzed dopaminergic, glutamatergic and γ-amino-butyric acid (GABA)ergic markers in the mesocorticolimbic structures of the adult offspring. Dopamine- and glutamate-related markers were studied by immunohistochemical (IHC) evaluations of the dopamine D1 receptor (D1R) and NMDA-receptor subunit 1 (NR1) because of their important roles in the regulation and modulation of sensorimotor gating (Ellenbroek et al., 1996, Shoemaker et al., 2005, Duncan et al., 2006a, Duncan et al., 2006b) and working memory (Abi-Dargham and Moore, 2003, Nakazawa et al., 2003, Williams and Castner, 2006, Niewoehner et al., 2007). Abnormalities in GABA-related systems were studied by IHC analyses of Reelin, Parvalbumin (PV), and GABA_A receptors containing the α2 subunit primarily because of their suggested pathophysiological contribution to working memory deficiency (Lewis et al., 2005, Hashimoto et al., in press).

Here, we used a well-established mouse model of prenatal immune challenge by the synthetic analogue of double-stranded RNA, polyriboinosinic–polyribocytidilic acid (PolyIC). PolyIC mimics the acute phase response to viral infection, which is accompanied by the presence of high levels of pro-inflammatory cytokines and other mediators of inflammation (Fortier et al., 2004, Meyer et al., 2006b, Cunningham et al., 2007). This immunological treatment also increases pro-inflammatory cytokine levels in the fetal brain (Meyer et al., 2006b, Meyer et al., in press-a), lending credence to the use of the prenatal PolyIC model in mice as an experimental tool to study the long-term consequences of fetal brain inflammation on subsequent brain and behavioral development. In order to capture early/middle and late gestation periods in mice, pregnant dams were subjected to inflammatory or control treatment on gestation day (GD) 9 or GD17. These gestational windows are identical to the ones included in our previous time window studies (Meyer et al., 2006a, Meyer et al., 2006b) and correspond roughly to the end of the first trimester and middle-to-late second trimester of human pregnancy, respectively, with respect to developmental biology and percentage of gestation from mice to human (Clancy et al., 2001, Kaufman, 2003).