Focus piece - More than just a cold: The impact of infection during early pregnancy on the developing fetal brain
By: Faraj Haddad
A brief introduction to the developing brain
86 billion. That is approximately the number of nerve cells in the human brain. These nerve cells contribute to our ability to experience the world around us through sensations, thoughts, feelings, and actions. The process by which these billions of nerve cells come about and organise into the brain is called brain development.
A lot of factors influence brain development, some of which are inherited. For instance, certain individuals may learn to walk or speak a little earlier or later because of genetic differences which control speech or movement. Similarly, the environment that an individual is raised in can influence their brain growth. Examples include how nutritious their diets were or the kind of parental care they got as a child.
Most of the time, genetic or environmental factors change brain development, but do not drastically change a person's life. Someone may have learned to walk a month later compared to their older sibling, but given a few more months, they learned to walk just as well.
Photo by Alina Grubnyak on Unsplash
Sometimes, however, certain genetic or environmental factors can steer brain development drastically off course, causing what are known as neurodevelopmental disorders. You may have heard of some of these: autism spectrum disorder, schizophrenia, and intellectual disability. Combined, these 3 disorders affect roughly 2-3% of the population.
Out of the 3 disorders listed above, autism spectrum disorder has come under the spotlight in recent years as more kids are being diagnosed with it. There are many reasons for this increase in autism diagnosis, including increased awareness and a change in diagnostic criteria that are used by physicians to classify autism. The hallmarks of autism, according to the recent diagnostic criteria set by the American Psychiatric Association, are changes in social communication and repetitive behaviour. Additionally, these new criteria also highlight the importance of changes in sensory processing in kids with autism. In support of this notion, recent studies show that autistic individuals do not experience their environment and surroundings in the same way as typically developing individuals.
More than just a cold – Infection during early pregnancy
In this research article, my work explores one environmental factor that can throw brain growth off track and increase the risk for autism: maternal infection during pregnancy.
The idea that maternal infection during pregnancy is bad for the embryo is relatively new compared to what we know about the harmful effects of smoking or drinking alcohol during pregnancy.
Research over the past two decades has highlighted the extent to which infection during pregnancy can increase the risk of autism in the offspring. In the USA and various Scandinavian countries, researchers discovered that children whose mothers had infections like the flu during pregnancy were almost twice as likely to be diagnosed with autism. To put this in context, the world health organization (WHO) estimates that 1 in 160 children worldwide develops autism. Applying the research findings mentioned above to the WHO estimate would mean that about 1 in 80 children would develop autism if they were exposed to maternal infection during pregnancy. However, not all infections are created equal, and the estimate of 1 in 80 is an unlikely worst-case scenario.
Photo by Amina Filkins on Pexels
The timing of maternal infection is crucial. Evidence suggests that the earlier the infection happens during pregnancy, the more severe the implications. Disrupting brain development early is like disrupting the foundations on which a house is built, everything that is built on top of faulty foundations is bound to be shaky and vulnerable.
The severity of maternal infection is also crucial. Most of the human studies referenced above obtained their data from hospital records, where a maternal infection was severe enough to require the mother to go to the hospital.
Unintentional damage – The downside of fighting off infection during pregnancy
The human studies summarized above also tell us that when it comes to the risk of autism, the kind of infection is not as important as its timing and severity. Whether a mother gets a viral infection like the flu, a bacterial urinary tract infection, or maybe even COVID-19, the effects on the baby could be similar.
Rather than the virus or bacteria being the problem, it is the mother's body trying to fight back against infection that unintentionally harms the developing fetal brain.
This maternal immune response (also known as maternal immune activation) is a clear way to rationalise how different infections can lead to the same outcome. After all, the body responds similarly to different kinds of infections. For instance, almost any infection can cause a fever.
My study attempted to investigate the effects of maternal immune activation on brain and behaviour. Specifically, our lab was interested in studying sensory processing which is disrupted in neurodevelopmental disorders like autism.
Tricking the maternal body to produce an immune response – goals and research techniques
To perform the study, our lab exposed pregnant rats to maternal immune activation during pregnancy. In this model, pregnant rats are administered a molecule that looks like a virus. This injection was performed around the equivalent of human 1st or 2nd-trimester pregnancy, which is when human studies show the biggest risk for the offspring to develop autism.
When the pregnant rat’s body detects this virus-like molecule, it thinks it’s facing something like the flu (or even COVID-19, as they both have similar molecular components) and develops an immune response to fight back.
In this process, the rat typically gets a mild sickness for 1-2 days, similar to how people might feel when getting a cold. Then, the rat continues with the pregnancy and gives birth approximately 12 days later. Their offspring then grow up and are tested in adolescence and adulthood.
The study sought to answer 3 main questions:
Do offspring exposed to maternal immune activation during pregnancy have altered sensory processing?
Do offspring exposed to maternal immune activation during early pregnancy show different sensory processing changes compared to those exposed to it during mid-pregnancy?
How can we enhance the reliability of results from maternal immune activation studies?
A startling method to study sensory processing
Sensory processing is the ability to take in the world around us in sounds, sights, scents, and touch, and then put it all together to experience the world as a whole. Sensory processing changes are common in individuals with autism. For example, kids with autism find background sounds like the noise of a fan more annoying, and they tend to react strongly to sounds that others may find quiet or easy to ignore.
To study sensory processing, our lab uses a test called the acoustic startle response. This is a response that mammals exhibit, where a loud sound causes them to startle – think of the last time somebody slammed the door behind you and you instinctively jumped!
By studying the acoustic startle response and how this response changes, we can gauge auditory sensory processing. For example, an individual may startle to a relatively quiet sound if their brain over processes this sound.
One of the strengths of using the startle response to study sensory processing is that practically the same tests can be performed in rats and in humans. In fact, students in our lab work with autistic individuals using the same sound levels we use in animal models to help translate our findings to clinical contexts.
Photo by Meru Bi on Pexels
Early exposure, late manifestation, selective changes
The main finding of the study was that early but not mid-pregnancy exposure to the virus-like molecule altered the offspring’s sensory processing. Rats born from these immune-activated mothers had a higher acoustic startle response, similar to individuals with autism, who also exhibit increased startle, a symptom typically termed as increased sensory reactivity.
Despite this similarity, the increased startle response only showed up when the rats were adults, but not when they were younger. This is quite different from individuals with autism who begin to experience this increased startle in childhood.
Beyond reactivity, rats exposed to maternal immune activation showed normal behaviour in other tests of auditory sensory processing. For example, when repeatedly exposed to the loud startling sound, typically developing individuals show a reduced startle response over time, a phenomenon known as habituation. Think about watching a horror movie, the more frequent the jump scares the less startling they become!
Despite reacting more to startle sounds, autistic individuals habituate to these sounds with repeated exposure just like typically developing individuals. Similarly, rats exposed to maternal immune activation showed similar habituation to control rats.
Another test that we performed was the pre-pulse inhibition test, which measures the brain’s ability to avoid sensory overstimulation. As the brain receives a piece of sensory information to process, it puts on the breaks, preventing more sensory information from overwhelming it until it finishes processing. Think about trying to hold a conversation while texting on your phone. As you process the texts, your ability to tend to the conversation is diminished.
The study found that rats exposed to maternal immune activation during pregnancy showed no changes in pre-pulse inhibition compared to the control group. This finding echoes some results from human studies showing no change in pre-pulse inhibition in autistic individuals.
Informing the reliability of maternal immune activation models
The final section of the study took a bit of a twist. Rather than running different kinds of experiments, we decided to focus on improving the techniques used in the maternal immune activation field, as they pertain to acoustic startle reflex testing.
Just as not all humans who get the flu during pregnancy give birth to individuals who develop autism, not all rats who are exposed to maternal immune activation exhibit sensory processing changes. Pregnant rats give birth to an average of 12 offspring per pregnancy, and each of the offspring may be affected to a different extent. In the field of maternal immune activation, researchers tend to only test 3-4 rats from each pregnancy.
Our study showed that to obtain reliable results, at least half of the rats from an average pregnancy should be tested (6-8 from each pregnancy). This ensures that enough offspring are tested to detect changes that may only affect a couple of offspring.
An imperfect model that paves the way forward
Our study builds on previous research that highlights how a maternal immune response during early pregnancy, even in the absence of infection, can lead to sensory processing changes in the offspring.
Although the nature of our findings matches results from human studies, the timing does not. Maternal immune activation rats show increased sensory reactivity in adulthood, which is different from what is seen in autistic individuals, where increased reactivity is seen early on in life.
There are many potential explanations to this discrepancy, including the idea that it takes more than exposure to just one risk factor to cause autism in humans. Instead, it could be an interplay between multiple environmental exposures and genetic predispositions. Since we only manipulated one factor in this study, the findings may be less severe or take longer to manifest than in humans.
While this study tells us a lot about sensory processing changes, it does not shed light on the brain structures that cause these changes. As a follow up study, our lab is currently looking at the brains of maternal immune activation rats, both in the womb and in adulthood. If our findings are similar to what is seen in the brains of humans with autism, we can start thinking about testing sensory processing treatments in our rat model.
“There are no autistic rats”, my research supervisor Dr. Susanne Schmid often emphasizes. One cannot expect a rat to show the complex social behaviour exhibited by humans, which is altered in autism. However, basic brain functions like sensory processing are highly similar in rats and humans. Because our experience of the world is so dependent on sensory processing, even slight changes could have major outcomes, such as the altered social behaviour seen in autism.