A new study reveals that maternal immune activation promotes sex-biased activation of the integrated stress response in the developing mouse brain and that this mechanistically contributes to the onset of autism-related behaviors uniquely in male offspring.
Autism spectrum disorder (ASD) is one of the most prevalent neurodevelopmental disorders, with current estimates identifying a diagnosis of one child out of every 54 in the US1. Epidemiological reports and recent experimental evidence together suggest that activation of a mother’s immune response during pregnancy may contribute to the etiology of neurodevelopmental disorders2,3,4. Indeed, various triggers of maternal immune activation (MIA)—including multiple pathogen-derived stimuli, environmental irritants and autoinflammatory conditions—have been reported to induce cortical abnormalities, circuit dysfunction and behavioral changes in animal models5,6,7,8,9,10,11.
While there is emerging evidence that gestational inflammation can perturb fetal brain development, the cellular and molecular events that promote neural dysfunction in MIA offspring remain incompletely understood. In the current issue of Nature Neuroscience, Kalish et al. find, in mice, that sex-dependent induction of the integrated stress response in developing cortical cells contributes to altered neurodevelopment and precipitates autistic-related behaviors in offspring exposed to maternal inflammation. These findings expand our understanding of the sex bias and etiology of autism-related neurodevelopmental disorders12.
To interrogate a connection between maternal inflammation and altered neurodevelopment, the authors employed a mouse MIA model whereby pregnant mothers were injected with polyinosinic:polycytidylic acid (polyI:C; a mimic of viral infection) or saline as a control. The authors first conducted single-cell RNA sequencing (scRNA-seq) on cortices of embryos from polyI:C- or saline-injected mothers to gain an unbiased and comprehensive picture of changes to the fetal brain that occur shortly after the onset of maternal inflammation. Analysis of differential gene expression 2 and 6 days after the induction of maternal inflammation revealed substantial changes in the transcriptional states of a variety of cortical cell types following MIA.
Unlike the roles of MIA in neurodevelopmental disorders, which have only recently emerged, the sex bias in ASD has long been appreciated, with ASD cases in males outnumbering those in females by approximately 4:1 (ref. 13). This sex bias in the human condition is also mirrored in the MIA model used in this study, since only male offspring developed behavioral abnormalities (Fig. 1). To gain new insights into potential factors that underlie male susceptibility and/or female resilience in MIA-induced neurodevelopmental disorders, Kalish et al. strategically took sex into account in their experimental design and analysis. Notably, many of the transcriptional changes induced by MIA in their scRNA-seq analysis were found to be sex-dependent, with cell-specific gene expression differing substantially between male and female MIA-exposed offspring across many cell types.
The authors discovered profound changes in translation initiation factors and other regulators of protein synthesis, both when comparing differentially expressed genes between MIA-exposed and saline male offspring and when comparing MIA-exposed female to MIA-exposed male offspring. These data lead the authors to speculate that maternal inflammation induces altered proteostasis in the fetal brain and that this impacts males to a more significant degree. To broadly address potential alterations in protein synthesis, they collected brains from embryos 6 days after polyI:C or saline injection and then assessed the total level of active protein synthesis using chemistry that specifically labels nascent peptides. In line with their scRNA-seq data demonstrating that ribosome subunit synthesis was downregulated in MIA-exposed males, Kalish et al. found that the total level of protein synthesis was also decreased in the cortex of MIA-exposed males compared to all other groups.
The question then arises as to what instigates this MIA-triggered arrest in protein synthesis in the fetal brain. The authors turned their attention to the integrated stress response (ISR), which is known to be a major adaptive process that obstructs translation to conserve cellular resources in dire situations. The eukaryotic translation initiation factor 2α (eIF2α) serves as a master regulator of the ISR, where phosphorylation of eIF2α coordinates an abrupt halt in protein synthesis. Consistent with a role for the ISR in their model, they observed elevated levels of phosphorylated eIF2α in the brains of male MIA-exposed mice (Fig. 1).
The ISR may be triggered by an array of pathological states to aid in adaptation to a changed environment. While a maternal inflammatory response certainly constitutes a divergence from homeostasis, the authors wondered what might be the specific molecular inducers that activate the ISR in the MIA-exposed male fetal brain. Again blotting cortical lysates, they found that the kinase PERK, which senses endoplasmic reticulum (ER) stress and signals directly to eIF2α, was specifically activated in the brains of MIA-exposed males and not in MIA-exposed females or saline controls. Collectively, these findings suggest that MIA-driven ER stress in the developing brain can disrupt protein synthesis as part of the ISR and that this is orchestrated by engagement of PERK and eIF2α signaling.
Previous work by this group demonstrated that IL-17a, whose production is ramped up by maternal T cells following polyI:C exposure, is required in this MIA model to disrupt neurodevelopment and promote the onset of autism-related behaviors in the offspring. In line with this integral role for the IL-17a signaling axis, Kalish et al. now report that antibody blockade of IL-17a signaling during gestation returns eIF2α phosphorylation and PERK activation to control levels in the brains of MIA-exposed male offspring, which implicates a novel link between maternal IL-17a and the fetal brain stress response.
Given the observed alterations in the protein synthesis landscape, the question arises: which specific proteins have altered levels of production in the MIA-exposed brain? The identity of these proteins could lend insight into the cellular components underlying alterations in brain development. To tackle this question, the authors employed another state-of-the-art sequencing technique known as Ribo-seq, in which ribosome-protected, actively translated transcripts are sequenced. Both male and female MIA-exposed offspring had significantly reduced synthesis of transcription and translation regulatory factors. Interestingly, however, MIA-exposed female brains were also uniquely characterized by increased translation of multiple factors that may promote adaptation and resiliency to MIA. For instance, MIA-exposed females upregulated the synthesis of chromatin remodeling factors and alternative translation initiation factors. Altogether, these findings support a hypothesis whereby a maternal immune response during pregnancy triggers the ISR particularly in male fetuses, resulting in disrupted protein synthesis that potentially perturbs neurodevelopment, whereas female offspring may have protective regulatory mechanisms that counteract both the induction of the ISR and its effects on protein synthesis.
This newly uncovered role for the ISR following gestational inflammation raises the intriguing possibility that blocking ISR induction may prevent some of the neurologic changes in MIA-exposed offspring. To test this idea, the authors took advantage of a mouse genetically modified to express a version of eIF2α that cannot be phosphorylated, thus blocking ISR activation. The offspring of wild-type mothers injected with saline or polyI:C and eIF2a mutant fathers were then evaluated to assess whether blocking eIF2α-mediated changes in proteostasis is sufficiently capable of hindering MIA-induced neurodevelopmental consequences. The presentation of human ASD differs tremendously, but classically recognized alterations in behavior for individuals on the autism spectrum include impaired sociability and repetitive or stereotyped behaviors. These core behavioral differences are also commonly seen in the male offspring of polyI:C-treated mothers, yet male MIA-exposed offspring expressing the mutant form of eIF2α did not exhibit such behavioral changes. Excitingly, the authors showed that treatment of pregnant mice with the ISR inhibitor ISRIB also rescued MIA-exposed male offspring from developing autism-related behaviors (Fig. 1).
Overall, these findings provide important new insights into both the sex differences and the role of the ISR in neurodevelopmental disorders. Moreover, further mining of the treasure trove of scRNA-seq and Ribo-seq data generated during this study by this group and others will likely help to uncover other novel mechanistic players that contribute to autism-related behaviors in the MIA model. It will be of particular interest in future studies to identify the key factors that afford resilience to females in this model. Conversely, the pathophysiology and expression of neurodevelopmental disorders such as autism in females is severely understudied, and greater efforts are needed to reveal the biological factors that can lead to behavioral abnormalities and altered brain maturation in females. Importantly, the work by Kalish et al. in this issue will provide a valuable blueprint to pursue similar studies in models of autism that affect females.
Maenner, M. J. MMWR Surveill. Summ. 69, 1–23 (2020).
Atladóttir, H. O. et al. J. Autism Dev. Disord. 40, 1423–1430 (2010).
Zerbo, O. et al. J. Autism Dev. Disord. 45, 4015–4025 (2015).
Estes, M. L. & McAllister, A. K. Science 353, 772–777 (2016).
Choi, G. B. et al. Science 351, 933–939 (2016).
Smith, S. E. P., Li, J., Garbett, K., Mirnics, K. & Patterson, P. H. J. Neurosci. 27, 10695–10702 (2007).
Shin Yim, Y. et al. Nature 549, 482–487 (2017).
Bolton, J. L. et al. Front. Synaptic Neurosci. 9, 10 (2017).
Giovanoli, S. et al. Science 339, 1095–1099 (2013).
Schwartzer, J. J. et al. Brain Behav. Immun. 63, 99–107 (2017).
Lammert, C. R. et al. J. Immunol. 201, 845–850 (2018).
Kalish, B. T. et al. Nat. Neurosci. https://doi.org/10.1038/s41593-020-00762-9 (2020).
Werling, D. M. & Geschwind, D. H. Curr. Opin. Neurol. 26, 146–153 (2013).
The authors declare no competing interests.
About this article
Cite this article
Zengeler, K.E., Lukens, J.R. Inflammation stresses out brain development. Nat Neurosci (2020). https://doi.org/10.1038/s41593-020-00775-4