Maternal Immune Activation during Pregnancy Alters Development of T Helper Cell Subsets of Offspring in Prenatal Models of Autism

Background: Autism spectrum disorder (ASD) is a perinatal developmental disorder characterized by behavioral, neurological, and immunological abnormalities. ASD is a multifactorial disorder, the causes of which have not been fully defined. However, clinical and experimental studies indicate that in addition to genetic predisposition, immune mechanisms, in general, and cytokine dysregulation, in particular, are contributing etiological factors in ASD. In murine models of autism, injection of pregnant dams with cytokines (IL-2, IL-6) or polyclonal activators (poly I:C, LPS) at mid-gestation causes autism-like behavioral abnormalities in their offspring in comparison to offspring of control PBS-injected dams.

Objectives: To examine murine prenatal models of autism to determine if immunostimulation during pregnancy causes alterations in the development and/or function of lymphoid and myeloid lineages in offspring, with particular emphasis on analysis of T helper subsets.

Methods: In the first model, we administered five daily i.p. injections (0.4 ug) of murine IL-2 to pregnant SJL/J mice during mid-gestation (E12-E16). In second model, C57BL/6 (B6) dams were given one i.p. injection of 20 mg/kg poly I:C (pI:C) or IL-6 (0.5 ug) at E12. Spleen cells from offspring of immunostimulated dams (IL-2, IL-6, and pI:C pups) were compared to offspring of pregnant mice injected with vehicle only (PBS pups) for mixed lymphocyte reaction (MLR) and cytotoxic T lymphocyte (CTL) function. Multi-color flow cytometry (FACS) analysis and TH cell differentiation cultures were also performed on spleen cells from pI:C and IL-6 pups, respectively. Sera and amniotic fluids from pregnant dams and supernatants from activated lymphocyte cultures of offspring were analyzed for the presence of multiple cytokines by Luminex assay.

Results: In these well-characterized prenatal mouse models of autism, we observed high levels of pro-inflammatory cytokines in maternal sera and amniotic fluids, and changes in the T cell subsets of their offspring. In the model where SJL dams were injected with IL-2 (vs. PBS), in addition to their previously shown abnormal behavior, IL-2 pups also exhibited accelerated T cell development, with a skewing toward TH1 cell differentiation, and significantly higher MLR and CTL responses to syngeneic B lymphoma cells or allogeneic spleen cells. In the second model, 24 hrs after injection of pregnant B6 dams with pI:C (vs. PBS), high levels of pro-inflammatory cytokines were detected in the sera and amniotic fluids. In addition, FACS analysis of activated T cells from 2-3 week old pI:C pups showed higher percentages of CD4+ TH cells with intracellular IL-17 (TH17 cells) than neonates from PBS-injected dams. Naïve CD4+ TH cells from IL-6 (vs. PBS) pups also showed significantly greater ability to differentiate towards TH17 cells.

Conclusions: Maternal immune activation during pregnancy caused production of cytokines that crossed the placenta and promoted the development of pro-inflammatory TH1 and TH17 cells in their offspring. Since TH1 and TH17 cells have each been shown to mediate pathogenesis of autoimmune disease, their presence in the offspring of IL-2, IL-6, and pI:C-injected pregnant dams suggests that TH1 and TH17 cells may contribute to the immunological and behavioral abnormalities observed in these rodent models of autism.