The Maternal Immune System and Autism
The maternal immune system plays a pivotal role during fetal development. In addition to protecting the mother and fetus from infectious agents, maternal immunoglobulin G (IgG) passes via the placenta from the mother to the fetal compartment, providing the infant with early protection against pathogens they may encounter postnatally. Perturbations of this fine balance, including maternal autoimmunity and inflammation, has been linked to neurodevelopmental disorders in offspring.
Autism spectrum disorder (ASD) is a highly heterogenous neurodevelopmental disorder (NDD) with affected individuals displaying altered social skills and communication in addition to repetitive and stereotypical behaviors. A subtype of autism, known as maternal autoantibody- related (MAR) ASD, is associated with the maternal production of specific autoantibodies that target proteins that are expressed in the developing brain- approximately 20% of mothers of children with ASD produce such autoantibodies. To better understand the effects of MAR-ASD specific autoantibodies on offspring neurodevelopment, behavior, and neuroanatomy, we utilize clinically relevant rodent models.
The Rat Model of MAR-ASD
While mouse models are commonly used and widely accepted to model many human disorders, we chose to use the rat model for our study due to their more complex neuroanatomy, behavior, and enhanced cognitive skills. To generate a proper model of MAR-ASD, we have developed a method in which we immunize rat dams with our MAR-ASD specific proteins (lactate dehydrogenase A and B, collapsin response mediator protein 1, and stress-induced phosphoprotein 1), such that the animals continuously produce their own antibodies to the peptide targets of interest (Figure 1). As such, our model allows for gestating offspring to be continuously exposed to the maternal autoantibodies as would occur in humans. Using the offspring from immunized rat dams, we were able to assess how their behavior and neuroanatomy changed across their lifespan as a result of autoantibody exposure.
Figure 1: Generation of MAR-ASD Rats: Schematic depicting the general workflow of MAR-ASD rat dam generation.
First and foremost, we confirmed that immunized dams generate a sustained antibody response lasting well beyond conception and parturition. We also confirmed that treated MAR-ASD dams had immune profiles (cytokines and chemokines) similar to control dams, confirming an absence of maternal inflammation and delineating our model from the maternal immune activation (MIA) model of NDDs. We next established that antigen-specific IgG is capable of entering the fetal compartment, and can be detected in the sera of immunized offspring. As some of the targets of the MAR-ASD autoantibodies are in the brain, we then used immunohistochemistry (IHC) to assess if IgG can be detected in the offspring brain. At postnatal day 2 (PND2) we were able to observe IgG deposited in multiple regions of the brain, which is abnormal under healthy conditions. This suggests that the MAR-ASD specific autoantibodies can potentially access the fetal brain and reach their protein targets, although there were not differences in the levels of these protein targets in the frontal cortex or cerebellum when measured by western blot.
The Effects of Autoantibody Exposure
Utilizing an array of behavioral testing, we found that offspring exposed to MAR-ASD maternal autoantibodies have abnormal behavioral profiles, including a lack of social behaviors when allowed to interact with a novel partner and decreased time spent in play behavior, which is common for juvenile rats. Additionally, early postnatal MAR-ASD offspring displayed decreases in ultrasonic vocalizations (USVs) when separated from their mothers. Together these findings represent species-specific changes that are relevant to the clinical ASD phenotype. When assessing neuroanatomical changes, we found that sex-specific changes in total brain volume (TBV) with a decrease in TBV in male MAR offspring while an increase in TBV was observed in female MAR offspring. Related to the structural differences observed in MAR-ASD offspring, we also identified differences in the levels of the brain metabolites taurine, choline, and glutathione when compared to controls. These metabolite differences were associated with changes in regional brain volume (Figure 2).
Figure 2: Summary of the effects of gestational MAR-ASD autoantibody exposure: a.) At PND 4, 8 and 12 offspring underwent testing of USVs, a measure of early communication. b.) On PND36, 55, and 103 time engaged in social behavior was measured via the social dyad task. c.) Using IHC imaging, we assessed IgG deposition in the brain. d.) Measurements of brain volume by MRI were taken at PND30 and PND70.
The full details of our results are published in the article entitled “Altered behavior, brain structure, and neurometabolites in a rat model of autism-specific maternal autoantibody exposure.”
Heritable, But Not Genetic
MAR-ASD is a subtype of autism that is highly correlated with specific patters of maternal autoantibodies. These MAR-ASD autoantibodies are passed from mother to offspring, suggesting the potential for intervention. In developing our novel rat model, we have created a translational tool to assess how gestational exposure to specific maternal autoantibodies affects offspring behavior and neurodevelopment. Several questions we can address with this tool include: how do the antibodies interact with their target proteins and what neural cells does this involve? How do they alter the trajectory of neurodevelopment and what does this mean for offspring behavior? Are there methods we can employ to mitigate the effects of the antibodies?
The work described herein provides some of the first insights into the neuroanatomical effects of specific maternal autoantibody exposure using a highly translational animal model system. As ASD increases in incidence, understanding how the disease develops to implement potential mitigation strategies, is essential. Future studies will build upon this work, expanding to include novel rat models of the additional patterns of autoantibody reactivity identified in clinical MAR-ASD cases. In addition, we will continue to aim to understand how these autoantibodies interact with their target cells using in vitro culturing techniques. Finally, we are working to expand our clinical studies to examine how maternal autoantibody levels change during gestation and how this may correlate with periods of fetal vulnerability.