Mom's gut to baby's brain: what's the link?
I received a couple inquiries regarding the recent publication of two reports from the Choi and Huh labs of MIT, so I thought I should write about them. I had my second Microbiome Diet article slated, but thought this would be more timely. Thanks so much to those people who wrote me, I love to hear about your interests!
These two new studies shed light on one potential mechanism contributing to the development of autism-spectrum disorder (ASD). These papers linked several factors, including maternal immune activation, gut bacteria, and abnormal brain development that contribute to the behaviors observed in ASD.
Keep in mind while you read this that although these new studies offer a lot of insight, they are just a couple pieces in the complex puzzle of understanding autism.
The takeaway:
An infection triggers maternal immune activation
Maternal immune activation involves a specialized subset of immune cells, T-helper-17 (Th17) cells
Commensal gut bacteria that adhere to intestinal cells promote the growth of Th17 cells
Th17 cells produce the proinflammatory molecule interleukin-17 (IL-17)
High IL-17 levels lead to abnormal cortical patches in the somatosensory region of the brain of offspring
These patches lead to the neurodevelopmental abnormalities characteristic of ASD
For more of the story, read on!
Let’s start with a little bit of background. Both genetic and environmental factors contribute to the development of autism [1]. Epidemiologic studies identify maternal exposure to stress, infection, anti-depressant medications, thalidomide or the anticonvulsant valproic acid as risk factors for ASD in offspring [1]. A review of 10,000 medical records suggests an association between autism and maternal viral infection in the first trimester and to a lesser extent, bacterial infection in the second trimester [2]. To study the effect of virus-induced maternal infection
and ASD, the two highlighted studies used mouse models.
Now let’s review some terminology. To avoid actually infecting mice with live virus, researchers often use poly(I:C), a molecule that resembles double-stranded viral RNA. Injection of poly(I:C) into mice mimics a viral infection, subsequently triggering an immune response. The experimenters refer to this as maternal immune activation (MIA). The immune response involves a specialized subset of immune cells called T-helper-17 (Th17) cells, which are normally beneficial to host defense [3]. Th17 cells produce a pro-inflammatory effector molecule (cytokine) called interleukin 17 (IL-17). Baby mice (pups) born to moms injected with poly(I:C) are called MIA-mice. Now that’s a bowl of alphabet soup!
A 2016 study by the same group showed [4] that injection of pregnant mice with poly(I:C) on embryonic developmental day 12.5 had the following effects;
increases IL-6, which is another inflammatory molecule
IL-6 induces Th17 cells to produce IL-17, in both mother and placenta
IL-17 passes through placenta
results in patches of disorganized cortical cells in the brains of offspring, both during embryonic development and in adulthood
increases the occurrence of offspring that displayed behavioral abnormalities
These cortical patches are responsible for the abnormal behaviors characteristic of ASD. The patches observed in MIA-mice resemble those described in ASD patients [5, 6]. One of the new studies determined that these patches occur in the primary somatosensory cortex called S1DZ. The same behaviors observed in MIA-mice were observed when specific neurons in S1DZ were activated in normal mice [7]. Reduction of neural activity in this region actually reversed the abnormal behavior, and researchers could modulate specific behaviors depending on which neuron they targeted in MIA-offspring [7].
It is important to note that 12% of mice (n = 50) with neurodevelopmental abnormalities had patches outside of the S1 region, suggesting that additional regions of the brain may control the behaviors affected by maternal immune activation [7].
While the neurobiology is fascinating, I will spend the rest of the article discussing the link between maternal immune activation, inflammation and gut microbiota.
The researchers also showed that the development stage during which the maternal immune activation occurred is important: injection of poly(I:C) into pregnant mice at embryonic day 12.5, but not days 15.5 or 18.5, resulted in the cortical patches and abnormal behavior [7].
Time of injection proportion of pups with cortical patches Day 12.5 79% (n =24) Day 15.5 13% (n = 21) Day 18.5 none (n = 22)
Thus, maternal immune activation at those later stages did not generate cortical patches or lead to the behavioral abnormalities in offspring. Day 12.5 in mice resembles late first trimester in humans; a key developmental stage during which viral infections have been shown to increase frequency of ASD offspring [3].
It is important to note an incidence of ASD in 79% of offspring is indicative of a very controlled experiment, rather than of reality. Clearly, 79% of pregnant women who get viral infections in the first trimester do not have children with ASD. Epidemiological studies show that only a very small fraction of mothers who experience infection while pregnant will go on to have a child with ASD [8, 9]. Other factors are clearly at play, which these studies tease apart.
One factor is gut microbiota, which are involved in inflammation. Specific gut bacteria are required for the (poly(I:C)-induced maternal immune response that led to the cortical patches and abnormal behaviors. The bacteria were initially identified as segmented filamentous bacteria (SFB), which are part of the normal gut flora (i.e., commensal) in mice. Additionally, some human commensal bacteria also promoted the Th17 response [10]. Pregnant mice colonized with mouse commensal SFB or human commensal bacteria that induce intestinal Th17 cells were more likely to produce offspring with MIA-associated abnormalities [10].
In contrast, mice lacking SFB did not have immune activation in response to injection with poly(I:C) and gave birth to normal pups. But when SFB was introduced into these mice, injection with poly(I:C) triggered immune activation and resulted in offspring with abnormalities.
To further confirm the necessity of SFB in the inflammatory response, poly(I:C)-injected pregnant mice were treated with vancomycin, an antibiotic that kills SFB. Vancomycin treatment reduced the proportion of intestinal Th17 cells, decreased the levels of IL-17 in pregnant poly(I:C)-injected mice and prevented development of behavioral abnormalities and cortical patches in their offspring [10].
Let's take a moment and appreciate the many factors going on here. Everyone has commensal gut bacteria that activate Th17 cells to some extent. Many women will have an infection or some kind of inflammatory response during pregnancy. Thus, it seems like every mother is set up to give birth to an autistic child, but we know that isn't the case. Adherent commensal bacteria and infection appear to play a role in some cases of ASD and these studies define those roles further.
What is the role of SFB and commensal bacteria? Previously, two groups independently identified SFB as sufficient to activate the Th17 response in mice [11,12]. SFB induces the proliferation of Th17 cells in the small intestine [12]. Th17 cells are highly abundant in the lamina propria of the small intestine, and the new study confirmed that in response to injection with poly(I:C), Th17 cells in the lamina propria (as well as the ileum) produced high levels of IL-17 [10].
As mentioned previously, SFB is not a pathogen; rather it is part of the normal gut flora. Commensal gut bacteria normally promote proper functioning of Th17 cells [12] by "priming" or "educating" the immune system. So, some immune response to commensal bacteria is normal. However, under some conditions, the interactions between commensal bacteria and the immune system can have detrimental effects.
SFB and other commensal bacteria can promote Th17 proliferation. Fecal samples obtained from both healthy human volunteers or patients with ulcerative colitis were tested in mice. All samples contained mixtures of bacteria that triggered proliferation of intestinal Th17 cells in mice, but none of them contained SFB (which is usually found in mice and rats) [13]. Thus, there are other bacteria that act similarly to SFB.
To understand which bacteria could be causing the Th17 response, researchers treated mice with different antibiotics that target different bacteria. They observed varied Th17 responses; in some cases the response was inhibited and in other cases it was amplified [13]. Thus, different species or subsets of gut bacteria are responsible for Th17 induction, and the mixture of bacteria may influence the intensity of the response.
It was also shown that bacterial colonization alone didn’t trigger Th17; the SFB or other commensal bacteria actually had to adhere to intestinal epithelial cells to elicit the Th17 response [13]. Furthermore, other commensal bacteria such as Bifidobacerium adolescentis or an adherent strain of Escherichia coli, and mucosal pathogens like E. coli O157:H7 and the yeast Candida albicans, all adhere to intestinal epithelial cells and all trigger a similar Th17 response [13]. Removing the ability of these bacteria to adhere to intestinal cells decreases the Th17 response. Bacteria that don't adhere don't induce Th17 responses.
One of the specific conditions that influences the interaction between commensal gut bacteria and the immune system is genetics. Of five types of mice that were colonized with SFB, four types had a similar Th17 response, but one type had a much weaker response [13]. This may be because that particular type of mice had lower levels of Th17 cells. An important potential caveat is that some of these experiments used mice with apparent predispositions for immune-related brain abnormalities and metabolic disorders, which may skew the results [9].
Interestingly, the high levels of IL-17 were only observed in pregnant mice. After exposure to poly(I:C), non-pregnant mice did not have systemic increases in IL-17 or other inflammatory molecules (IL-1β, IL-23 and IL-6) that stimulate Th17 to produce IL-17 [10]. The reason for the differences between pregnant and non-pregnant mice isn’t fully understood.
What exactly is IL-17 doing? That question remains largely unanswered. But researchers have shown that IL-17 crosses the placenta where it signals production of more IL-17 in the fetal brain [4]. Furthermore, a functional IL-17 receptor in the fetal brain is necessary to induce the cortical and developmental abnormalities in MIA-mice [4]. IL-17 can substantially affect cellular survival, differentiation, and signalling pathways, all of which are important for cortical development and neural function [3].
Overall, these studies suggest that a number of conditions must be met for the development of ASD behavioral abnormalities by the pathway described in these new studies:
Timing and type of maternal immune activation during pregnancy
Composition of maternal gut bacteria (bacteria that adhere to intestinal epithelial cells and induce Th17 response)
Mother’s genetic makeup, which may influences the intestinal and systemic Th17 response
The authors conclude with: “Women with gut microbial communities that promote excessive Th17 cell differentiation may therefore be more likely to bear children with autistic spectrum disorder in the event of pathological inflammation during pregnancy.”
So what can you take away from this study? One thing is that sometimes, certain conditions, known or unknown, align and cause disorders and diseases. Furthermore, even knowing all or most of the factors that influence a certain disease or disorder doesn’t necessarily mean we know how to properly intervene. A genetic predisposition often underlies disease states. But based on what these and other studies indicate, there are some factors we can influence. Keeping gut and systemic inflammation low, avoiding infections while pregnant, and supporting a diverse and healthy gut microbiota are key.
References
Malkova, N.V., et al., Maternal immune activation yields offspring displaying mouse versions of the three core symptoms of autism. Brain Behav Immun, 2012. 26(4): p. 607-16.
Atladottir, H.O., et al., Association of hospitalization for infection in childhood with diagnosis of autism spectrum disorders: a Danish cohort study. Arch Pediatr Adolesc Med, 2010. 164(5): p. 470-7.
Wong, H. and C. Hoeffer, Maternal IL-17A in autism. Exp Neurol, 2017.
Choi, G.B., et al., The maternal interleukin-17a pathway in mice promotes autism-like phenotypes in offspring. Science, 2016. 351(6276): p. 933-9.
Casanova, M.F., et al., Focal cortical dysplasias in autism spectrum disorders. Acta Neuropathol Commun, 2013. 1: p. 67.
Stoner, R., et al., Patches of disorganization in the neocortex of children with autism. N Engl J Med, 2014. 370(13): p. 1209-1219.
Shin Yim, Y., et al., Reversing behavioural abnormalities in mice exposed to maternal inflammation. Nature, 2017. 549(7673): p. 482-487.
Atladottir, H.O., et al., Autism after infection, febrile episodes, and antibiotic use during pregnancy: an exploratory study. Pediatrics, 2012. 130(6): p. e1447-54.
Estes, M.L. and A.K. McAllister, Maternal immune activation: Implications for neuropsychiatric disorders. Science, 2016. 353(6301): p. 772-7.
Kim, S., et al., Maternal gut bacteria promote neurodevelopmental abnormalities in mouse offspring. Nature, 2017. 549(7673): p. 528-532.
Gaboriau-Routhiau, V., et al., The key role of segmented filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity, 2009. 31(4): p. 677-89.
Ivanov, II, et al., Induction of intestinal Th17 cells by segmented filamentous bacteria. Cell, 2009. 139(3): p. 485-98.
Atarashi, K., et al., Th17 Cell Induction by Adhesion of Microbes to Intestinal Epithelial Cells. Cell, 2015. 163(2): p. 367-80.