Sex disparate gut microbiome and metabolome perturbations precede disease progression in a mouse model of Rett syndrome - PubMed (original) (raw)

doi: 10.1038/s42003-021-02915-3.

Tianna E Grant 1, Rebecca L Palmer 1, Demario Chappell 1, Sophia M Hakam 1, Kendra M Yasui 2, Matt Rolston 1, Matthew L Settles 3, Samuel S Hunter 3, Abdullah Madany 1, Paul Ashwood 1, Blythe Durbin-Johnson 3 4, Janine M LaSalle # 5 6, Dag H Yasui # 1

Affiliations

Sex disparate gut microbiome and metabolome perturbations precede disease progression in a mouse model of Rett syndrome

Kari Neier et al. Commun Biol. 2021.

Abstract

Rett syndrome (RTT) is a regressive neurodevelopmental disorder in girls, characterized by multisystem complications including gut dysbiosis and altered metabolism. While RTT is known to be caused by mutations in the X-linked gene MECP2, the intermediate molecular pathways of progressive disease phenotypes are unknown. Mecp2 deficient rodents used to model RTT pathophysiology in most prior studies have been male. Thus, we utilized a patient-relevant mouse model of RTT to longitudinally profile the gut microbiome and metabolome across disease progression in both sexes. Fecal metabolites were altered in Mecp2e1 mutant females before onset of neuromotor phenotypes and correlated with lipid deficiencies in brain, results not observed in males. Females also displayed altered gut microbial communities and an inflammatory profile that were more consistent with RTT patients than males. These findings identify new molecular pathways of RTT disease progression and demonstrate the relevance of further study in female Mecp2 animal models.

© 2021. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1

Fig. 1. Longitudinal phenotypes in Mecp2-e1 mutant and wild-type mice.

a Comparison of neurophenotyping score (range 1–7, 1 being least severe and 7 being most severe) in Mecp2-e1 mutant vs. wild-type (wt) mice longitudinally and cross-sectionally across disease course (between 6 and 19 weeks of age for females and 6 and 16 weeks of age for males), stratified by sex and controlling for within-litter effects. b Comparison of the most sensitive measure of gait (overlap distance in centimeters between hind and front legs, with a greater distance indicating a more impaired gait) in mutant vs. wild-type mice longitudinally and cross-sectionally between 6 and 17 weeks of age in females and 6 and 15 weeks of age for males, stratified by sex and controlling for within-litter effects. c Comparison of body weight of mutant vs. wild-type mice longitudinally and cross-sectionally between 5 and 19 weeks of age for females and 5 and 19 weeks of age for males, stratified by sex and controlling for within-litter effects. Cross-sectional analyses were controlled for multiple comparisons at each time point. N = 11–19/genotype/sex. Each solid dot represents the mean of each group at each time point and error bars are standard error of the mean (SEM). *FDR < 0.05, **FDR < 0.01, ***FDR < 0.001 in mutant vs. control, cross-sectionally. Longitudinal _p_-values represent the overall association between mutant and wild-type mice across disease course using linear mixed effects models.

Fig. 2

Fig. 2. Longitudinal differences in the fecal microbiome in mutant vs. wild-type Mecp2-e1 mice.

a Venn diagram comparing ASVs that were significantly associated (FDR < 0.05) with genotype longitudinally across disease course in females vs. males, while controlling for within-litter effects. b Number of differentially abundant taxa (ASVs) by genotype in females and males cross-sectionally at each time point between 5 and 19 weeks of age in females and 5 and 16 weeks of age in males, colored by phylum (FDR < 0.05). c the top 25 ASVs with the most statistically significant genotype by age interaction plotted by log(fold change) in mutants vs. wild-type (wt) mice at each time point between 5 and 19 weeks of age in females and 5 and 16 weeks of age in males. Each line represents a single ASV and lines are colored by Family. N = 11–19/genotype/sex.

Fig. 3

Fig. 3. Differences in genotype- and phenotype-associated ASVs in mutant vs. wild-type Mecp2-e1 mice across disease course.

Log(fold change) in normalized read counts of mutant vs. wild-type (wt) mice at each time point in ASVs that were significantly associated with phenotype and genotype at a minimum of one-quarter of the time points in both females and males. Each line represents an individual ASV and lines are colored by Genus. Heatmaps at the bottom of each graph are colored red with increasing intensity based on the difference in phenotype measure by genotype at each time point as reported in Fig. 1. a ASVs associated with both body weight and genotype in females, b ASVs associated with both body weight and genotype in males, c ASVs associated with both neurophenotyping score and genotype in females, d ASVs associated with both neurophenotyping score and genotype in males. N = 11–19/genotype/sex. *FDR < 0.05, **FDR < 0.01, ***FDR < 0.001 in Mecp2-e1 mutant vs. wild-type mice cross-sectionally at each time point.

Fig. 4

Fig. 4. Fecal levels of cytokines in Mecp2-e1 mutant vs. wild-type mice across disease course.

Comparison of a fecal interferon-gamma (IFNγ), b interleukin-4 (IL-4), and c interleukin-1alpha (IL-1α) levels between Mecp2-e1 mutant and wild-type (wt) mice longitudinally and cross-sectionally across disease course (at 6, 10, and 18 weeks in females and 6, 10, and 14 weeks in males). Analyses were stratified by sex and controlled for within-litter effects. N = 6-8/genotype/sex. Detailed Ns for each time point are included in Supplementary Data 12. Each solid dot represents the mean of each group at each time point and error bars are standard error of the mean (SEM). ^FDR < 0.10, *FDR < 0.05 in mutant vs. wild-type controls cross-sectionally.

Fig. 5

Fig. 5. Differentially abundant fecal metabolites by Mecp2-e1 genotype and age in female and male mice.

a Difference in butyrate concentrations in fecal samples in Mecp2-e1 mutant vs. wild-type (wt) mice longitudinally and cross-sectionally across disease course (at 5, 9, and 19 weeks of age in females and 5, 9, and 16 weeks of age in males), stratified by sex and controlling for within-litter effects. Each solid dot represents the mean of each group at each time point and error bars are standard error of the mean (SEM). N = 8/genotype/sex. *FDR < 0.05, **FDR < 0.01 in Mecp2-e1 mutant vs. wild-type cross-sectionally. Longitudinal _p_-values represent the overall association between mutant and wild-type mice across disease course using linear mixed effects models. b Venn diagrams comparing fecal biogenic amines and c comparing fecal lipids that were longitudinally (at 5, 9, and 19 weeks for females and 5, 9, and 16 weeks for males) associated with age, genotype, and a genotype by age interaction (FDR < 0.05) in both females and males. N = 8/genotype/sex.

Fig. 6

Fig. 6. Longitudinal fecal metabolome profiles in mutant vs. wild-type Mecp2-e1 mice.

Principal component analysis (PCA) plots of the top two principal components (PC1 and PC2) for fecal biogenic amines (a) and fecal lipids (b) with each point representing a single fecal sample and genotype represented by color and time point represented by shape. The log2(fold change) between mutant and wild-type (wt) for the top 10 biogenic amines (c) and lipids (d) with the most significant genotype by age interactions in females and males, plotted longitudinally across disease course (at 5, 9, and 19 weeks for females and 5, 9, and 16 weeks for males). Each line represents a single biogenic amine or lipid and is colored as such. FA = fatty acid, FAHFA = fatty acid esters of hydroxy fatty acid, TAG = triglyceride, PG = phosphatidylglycerol, SM = sphingomyelin. N = 8/genotype/sex.

Fig. 7

Fig. 7. Brain lipidome profiles in Mecp2-e1 mutant and wild-type mice.

a Principal component analysis (PCA) plot of female and male mutant and wild-type cortical lipids measured at the end of disease course (19 weeks for females and 16 weeks for males). The two principal components (PCs) that explain the most variation in samples are plotted with PC 1 on the x-axis and PC 2 on the y-axis. Each dot represents one sample, with colors representing genotype and shapes representing sex. b Heatmap depicting the relationship between the 10 fecal lipids at 19 weeks and the 10 brain lipids at 9 weeks with the most significant association with genotype in females. Associations were strongest for fecal lipids at 9 weeks of age, and thus those are depicted here. Blocks are colored based on correlation coefficient, with red representing positive correlations and blue representing negative correlations. FA = fatty acid, FAHFA = fatty acid esters of hydroxy fatty acid, TAG = triglyceride, PG = phosphatidylglycerol, SM = sphingomyelin. N = 6–8/genotype/sex. ^FDR < 0.10.

Fig. 8

Fig. 8. Female and male Mecp2-e1 mutants exhibit divergent phenotypic and molecular RTT disease progression.

In early adulthood at 5–7 weeks of age, Mecp2-e1 mutant males begin to show declines in motor and neurological function and changes in their gut microbiome. On the other hand, female Mecp2-e1 mutants at 5–8 weeks of age do not yet display motor and neurological symptoms, but begin to show increased body weight, altered gut microbiota and altered gut metabolites, such as short chain fatty acids (SCFAs). Females begin to display progressively increased body weight and drastic shifts in the gut metabolome at 9 weeks of age, prior to persistent declines in neurological and motor function which occurred at 10–11 weeks of age. Peak differences in the gut microbiome occurred at 12 weeks of age in females. Males, on the other hand, display progressive worsening of neuromotor symptoms and progressive changes in the gut microbiome in the absence of drastic shifts in the gut metabolome between 9 and 15 weeks of age. In late-stage RTT disease between 16 and 19 months of age, males exhibit severe morbidity and early mortality, whereas females continue to display progressively increased body weight and neuromotor function, continued evidence of a disrupted gut microbiome and metabolome, a Th2 inflammatory response in the gut, and altered brain lipid profiles. Figure created with BioRender.com.

References

    1. Neul JeffreyL, et al. Rett syndrome: revised diagnostic criteria and nomenclature. Ann. Neurol. 2010;68:944–950. -PMC -PubMed
    1. Amir RE, et al. Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl- CpG-binding protein 2. Nat. Genet. 1999;23:185–188. -PubMed
    1. Meehan R, Lewis JD, Bird AP. Characterization of MECP2, a vertebrate DNA binding protein with affinity for methylated DNA. Nucleic Acids Res. 1992;20:5085–5092. -PMC -PubMed
    1. Mellén M, Ayata P, Dewell S, Kriaucionis S, Heintz N. MeCP2 binds to 5hmC enriched within active genes and accessible chromatin in the nervous system. Cell. 2012;151:1417–1430. -PMC -PubMed
    1. Guo JU, et al. Distribution, recognition and regulation of non-CpG methylation in the adult mammalian brain. Nat. Neurosci. 2014;17:215–222. -PMC -PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources