Enhancing rigor and reproducibility in maternal immune activation models: practical considerations and predicting resilience and susceptibility using baseline immune responsiveness before pregnancy

Myka L. Estes, Kathleen Farrelly, Scott Cameron, John Paul Aboubechara, Lori Haapanen, Joseph Schauer, Aurora Horta, Kathryn Prendergast, Jeremy A. MacMahon, Christine I. Shaffer, Catherine T. Le, Gregory N. Kincheloe, Danielle John Tan, Deborah van der List, Melissa D. Bauman, Cameron S. Carter, Judy Van de Water, A. Kimberley McAllisterbioRxiv 2019 · DOI

TL;DR

This paper tackles the critical issue of reproducibility in maternal immune activation (MIA) mouse models, which are crucial for studying neurodevelopmental disorders. The authors identify key sources of variability – the poly(I:C) reagent itself, the dam's baseline immune responsiveness (BIR), and even the source of the mice – and propose practical strategies to improve rigor and predict offspring outcomes.

What Problem Does This Paper Address?

Maternal immune activation (MIA) models in mice are invaluable tools for understanding the prenatal origins of neurodevelopmental disorders like autism and schizophrenia. These models mimic the effects of maternal infection during pregnancy, which is a known risk factor for these conditions. However, the field faces a significant challenge: a lack of reproducibility. Studies often yield conflicting results, hindering progress in identifying biomarkers and developing effective treatments. This variability stems from multiple factors, making it difficult to reliably translate findings from animal models to human health. The authors aim to dissect these sources of variability and provide actionable recommendations for enhancing the reliability and predictive power of MIA models.

Methods

The researchers employed a multi-faceted approach to investigate variability in MIA models:

  • Baseline Immune Responsiveness (BIR) Measurement: Virgin female mice were injected with a low dose of poly(I:C) to assess their individual IL-6 response, categorizing them into low, medium, and high responders. This was done before pregnancy to establish a baseline.
  • Maternal Immune Activation (MIA) Induction: Pregnant mice were injected with varying doses of poly(I:C) at gestational day 12.5.
  • Outcome Measures: A range of assessments were performed on both dams and offspring, including:
  • Maternal: Sickness behavior, body temperature, weight change, IL-6 and IL-17a levels, and litter loss.
  • Offspring: Behavioral tests (self-grooming, rearing), synapse density in brain tissue (immunohistochemistry), and protein levels of key neurodevelopmental markers (STAT3, MEF2A, TH) via Western blot.
  • Microbiome Analysis: qPCR was used to detect the presence of segmented filamentous bacteria (SFB) in fecal samples from mice sourced from different vendors.
  • Statistical Analysis: Standard statistical methods including t-tests, ANOVAs, and post-hoc tests were used to analyze the data.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Poly(I:C) is Not a Monolith: The immunogenicity of poly(I:C), a common reagent used to induce MIA, varies significantly depending on the manufacturer, lot, and even bottle. This variability directly impacts the maternal IL-6 response and subsequent offspring outcomes.
  • Baseline Immune Responsiveness (BIR) is Predictive: A female mouse's immune response to a low-dose poly(I:C) challenge before pregnancy is a strong predictor of her offspring's susceptibility or resilience to MIA-induced abnormalities. Low BIR dams tend to have more resilient offspring, while medium and high BIR dams' offspring show distinct behavioral phenotypes.
  • Mouse Source Matters: Even within the same strain (C57BL/6J), the vendor from which mice are sourced can dramatically influence MIA outcomes. This is partly due to differences in gut microbiome composition, specifically the presence of segmented filamentous bacteria (SFB).
  • A Threshold for MIA: Not all doses of poly(I:C) lead to robust MIA phenotypes. The study suggests a threshold of MIA must be crossed for consistent and reproducible effects in offspring, and this threshold can be influenced by the factors mentioned above.
  • Biomarkers Identified: The study identified potential biomarkers (MEF2A, STAT3, TH) in the striatum of offspring that correlate with MIA outcomes, offering a potential avenue for future diagnostic or predictive tools.

Key Results

  • Poly(I:C) Variability: Different lots and sources of poly(I:C) elicited IL-6 responses in dams ranging from 0-2% of what was reported in earlier studies, highlighting a significant decline in potency.
  • Dose-Dependent Effects: A 30 mg/kg dose of pure poly(I:C) was identified as the most effective for inducing reproducible changes while minimizing litter loss. Doses of 20 mg/kg were insufficient to consistently elicit MIA phenotypes.
  • BIR Predicts Behavior: MIA offspring from low BIR dams showed increased grooming, while those from medium and high BIR dams exhibited altered rearing behavior (decreased in medium, increased in high BIR).
  • Vendor Differences: Mice from Taconic (TAC) dams showed increased IL-17a and TH17 cell responses to poly(I:C) compared to Charles River (CR) dams. TAC dams also did not exhibit fetal loss, unlike CR dams.
  • Synapse Density: MIA offspring from low BIR dams (especially from TAC) showed a significant decrease in synapse density.

Limitations and Future Work

  • Controlling for Variables: The authors emphasize the need to control for factors like injection timing, estrus cycle, and stressful vivarium conditions when comparing absolute IL-6 levels.
  • BIR Grouping: While useful for partitioning, the quartile-based grouping of BIR precludes continuous statistical analysis. The authors recommend using enough poly(I:C) to allow for direct comparison of absolute IL-6 levels.
  • Future Directions: The paper suggests that understanding the circuit and molecular mechanisms underlying the distinct phenotypes caused by different BIR levels could uncover factors influencing susceptibility and resilience. Further research is needed to fully characterize the differences and causes of the wide range of BIR observed. The authors also propose that BIR should be considered in human epidemiological study design to identify vulnerable populations.

Summary updated

Abstract

Abstract Despite the potential of rodent models of maternal immune activation (MIA) to identify new biomarkers and therapeutic interventions for a range of psychiatric disorders, their value is currently limited by issues of scientific rigor and reproducibility. Here, we report three sources of variability—the immunogenicity of the poly(I:C), the baseline immune responsiveness (BIR) of the females prior to pregnancy, and differences in immune responses in C57/B6 dams across vendors. Similar to the variable effects of human maternal infection, MIA in mice does not cause disease-related phenotypes in all offspring and the magnitude and type of maternal response, determined by a combination of poly(I:C) dose and BIR, predicts offspring outcome. Together, our results provide recommendations for optimization of MIA protocols to enhance rigor and reproducibility and reveal new factors that drive susceptibility of some pregnancies and resilience of others to MIA-induced abnormalities in offspring.
BibTeX
@misc{estes2019enhancing,
  author = {Myka L. Estes and Kathleen Farrelly and Scott Cameron and John Paul Aboubechara and Lori Haapanen and Joseph Schauer and Aurora Horta and Kathryn Prendergast and Jeremy A. MacMahon and Christine I. Shaffer and Catherine T. Le and Gregory N. Kincheloe and Danielle John Tan and Deborah van der List and Melissa D. Bauman and Cameron S. Carter and Judy Van de Water and A. Kimberley McAllister},
  title = {{Enhancing rigor and reproducibility in maternal immune activation models: practical considerations and predicting resilience and susceptibility using baseline immune responsiveness before pregnancy}},
  year = {2019},
  doi = {10.1101/699983},
  howpublished = {bioRxiv},
}

Loading related content...

Figure