Prenatal Infections and Autism Risk: The Biology Underpinnings Explained

How infections can rewire the developing fetal brain in ways that may lead to autism.

Image adapted from Freepik.com

In my previous article, “How Prenatal Infections Can Amplify Autism Risk (and Help Explain 1 in 7 Cases),” I explored the epidemiological evidence that infections during pregnancy, especially those accompanied by fever, can modestly increase the risk of autism in children, with a 12-17% population attributable risk (i.e., accounting for about 1 in 7 autism cases).

In this follow-up, I’ll look into the biology: How exactly might prenatal infections (i.e., occurring during pregnancy) rewire the fetal brain in ways that contribute to autism risk? What are the mechanisms involved?

Does it involve a Specific Pathogen?

First, it should be clarified that some pathogens exhibit natural tropism (the predilection to infect) the developing fetal nervous system due to their ability to cross the placenta. This group of pathogens is called TORCH, which stands for toxoplasmosis, other (syphilis and hepatitis B), rubella, cytomegalovirus (CMV), and herpes simplex virus (HSV). Human immunodeficiency virus (HIV) and Zika virus are sometimes included in this group as well.

But when it comes to autism, the classical TORCH infections are not solely responsible. As I covered in the prequel, the risk of autism following prenatal infections was not influenced by either the type (bacterial, viral, or fungal) or the site (respiratory, gastrointestinal, or urinary) of infections. In fact, the only significant predictor of autism risk was the severity of infection, as indicated by the presence of fever or hospitalization status.

This means that infections occurring during pregnancy in general confer an increased risk of autism, favouring the maternal immune activation hypothesis of autism. In short, it’s not the pathogen itself that matters most but the mother’s immune response to it.

Maternal Immune Activation in Autism

Basically, it refers to the immune activation of the mother (maternal) in response to infections during pregnancy, where inflammation is initiated to control the infection. But such inflammation, especially when excessive, can reach the placenta and affect the neurodevelopment of the fetus.

But what’s the evidence supporting this hypothesis?

It started with observations that children with autism often showed higher levels of inflammatory cytokines in the blood circulation and brain. For example, a 2014 meta-analysis of 17 studies reported that 6 out of 19 cytokines—i.e., interleukin (IL)-1β, IL-6, IL-8, interferon-gamma (IFN-ɣ), eotaxin, and monocyte chemotactic protein-1 (MCP-1)—were consistently higher in those with autism compared to those without. Such elevated cytokine levels were also more pronounced in children with more severe autism.

When children with autism pass away from various causes, their families may choose to donate their brain tissues for research. These samples are typically preserved in specialized brain banks and made available to scientists, who have since identified elevated levels of inflammatory cytokines, such as IL-6, IL-8, and MCP-1, particularly in the cortex and cerebellum (Figure 1).

Figure 1. Elevated inflammation in the brain of children with autism. (A) Expression of immune-related proteins in the brain cerebellum of children with (left) and without (right) autism. Each dot represents a specific protein, with larger and darker dots indicating higher levels. As shown, several inflammation-related molecules, including IL-6 and MCP-1, are more abundant in the autism brain. (B) Immunostaining for IL-6 in the cerebellum from children with and without (control) autism. The darker brown patches (arrows) indicate higher expression of IL-6. Sources: Wei et al. (2011), Journal of Neuroinflammation; Vargas et al. (2004), American Neurological Association.

So we have correlational evidence that pro-inflammatory cytokines are linked to autism. To demonstrate causation—that such cytokines were responsible for the pathogenesis of autism—we have to turn to animal models. After all, we can’t just artificially induce or administer these cytokines to mothers and see if their child will develop autism. But we can work with animals as long as certain ethical standards are adhered to.