Maternal Microbiota Modulate a Fragile X-like Syndrome in Offspring Mice

Abstract

Maternal microbial dysbiosis has been implicated in adverse postnatal health conditions in offspring, such as obesity, cancer, and neurological disorders. We observed that the progeny of mice fed a Westernized diet (WD) with low fiber and extra fat exhibited higher frequencies of stereotypy, hyperactivity, cranial features and lower FMRP protein expression, similar to what is typically observed in Fragile X Syndrome (FXS) in humans. We hypothesized that gut dysbiosis and inflammation during pregnancy influenced the prenatal uterine environment, leading to abnormal phenotypes in offspring. We found that oral in utero supplementation with a beneficial anti-inflammatory probiotic microbe, Lactobacillus reuteri, was sufficient to inhibit FXS-like phenotypes in offspring mice. Cytokine profiles in the pregnant WD females showed that their circulating levels of pro-inflammatory cytokine interleukin (Il)-17 were increased relative to matched gravid mice and to those given supplementary L. reuteri probiotic. To test our hypothesis of prenatal contributions to this neurodevelopmental phenotype, we performed Caesarian (C-section) births using dissimilar foster mothers to eliminate effects of maternal microbiota transferred during vaginal delivery or nursing after birth. We found that foster-reared offspring still displayed a high frequency of these FXS-like features, indicating significant in utero contributions. In contrast, matched foster-reared progeny of L. reuteri-treated mothers did not exhibit the FXS-like typical features, supporting a key role for microbiota during pregnancy. Our findings suggest that diet-induced dysbiosis in the prenatal uterine environment is strongly associated with the incidence of this neurological phenotype in progeny but can be alleviated by addressing gut dysbiosis through probiotic supplementation.

Keywords: FMRP; FXS; Lactobacillus reuteri; microbiome; probiotic.

Conflict of interest statement

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Figures

Figure 1

Experimental overview. The present study focused upon offspring from an earlier multigenerational study that spontaneously exhibited a syndrome of behavioral atypia and features resembling phenotypic features of FXS in humans. These mice were subsequently characterized using morphometrics, videotape analyses of behaviors, and post-mortem evaluation of tissues, displaying a neurodevelopmental phenotype with Fragile X-like features. To test a microbe-driven hypothesis, pregnant mice with a history of (h/o) WD (n = 12) were then randomly subdivided with half receiving a probiotic L. reuteri (LR) ATCC-PTA-6475 in their drinking water. These animals underwent testing of inflammatory cytokines, and their progeny was examined for FXS-like phenotypes and FMRP expression levels. Finally, pregnant mice underwent Caesarian (C-section) rederivation to test our hypothesis that in utero microbial events rather than post-partum microbes were leading aberrant behavioral phenotypes in offspring.

Figure 2

Measuring dysmorphia of head and ears. Misshaped head and ears are a characteristic phenotype of FXS in humans. We tested ear pinnae morphology by measuring the height and the width of each ear and estimating the effective diameter to examine the presence of this phenotype in our mice. Analysis of the control group (n = 7), the history of (h/o) WD group (n = 5), and the h/o WD + LR group (n = 8) showed significant differences in the ear size and the skull width (p < 0.05).

Figure 3

Hyperactivity (a) and stereotypic head bobbing (b) in mice with FXS-like phenotype. To examine other classic features of FXS including hyperactivity and head-bobbing stereotypy, we examined video footage of sham control animals (n = 6), animals with a history of (h/o) WD (n = 6), and animals with h/o WD + LR (n = 6). The animals were measured at 30 s intervals in home cages under standardized conditions. Significant differences were found between treatment groups (* p < 0.05 and *** p < 0.001).

Figure 4

In utero probiotic L. reuteri effects on FXS-like phenotypes in progeny mice. To test our microbe-driven hypothesis, pregnant mothers with a history of (h/o) WD were randomly subdivided with half receiving probiotic L. reuteri in their drinking water (n = 6) and half receiving regular drinking water (n = 6). The frequency of FXS-like features was measured in each treatment group. Significant differences (*** p < 0.001) were found after in utero dosing with L. reuteri, and the benefits of in utero L. reuteri were preserved after C-section rederivation. There were no significant (NS) differences between groups in females.

Figure 5

Expression of pro-inflammatory cytokine Il-17A in pregnant mother mice. To test systemic levels of inflammatory cytokines under different dietary and microbial conditions, we used whole blood from control mothers (n = 8), the mothers with a history of (h/o) WD (n = 8), and mothers with h/o WD + LR (n = 8).To test systemic levels of Il-17A in offspring mice, we used whole blood collected by terminal cardiac puncture and diluted 1:1. Circulating Il-17A levels were determined using ELISA at Eve Technologies (Calgary, AB, Canada). We found significant differences among treatment groups (* p < 0.05 and ** p < 0.001).

Figure 6

Brain expression levels of FMRP in offspring mice. FMRP expression level was measured using Western Blot analysis in male mice with a history of (h/o) WD + LR (n = 19) and male mice with h/o WD only (n = 8). Significant differences between the groups (* p < 0.05) were observed.