Combination of gut microbiota and plasma amyloid-β as a potential index for identifying preclinical Alzheimer's disease: a cross-sectional analysis from the SILCODE study

Abstract

Background: Plasma amyloid-β (Aβ) may facilitate identification of individuals with brain amyloidosis. Gut microbial dysbiosis in Alzheimer's disease (AD) is increasingly being recognized. However, knowledge about alterations of gut microbiota in preclinical AD, as well as whether the combination of plasma Aβ and gut microbiota could identify preclinical AD, remains largely unknown.

Methods: This study recruited 34 Aβ-negative cognitively normal (CN-) participants, 32 Aβ-positive cognitively normal (CN+) participants, and 22 patients with cognitive impairment (CI), including 11 patients with mild cognitive impairment (MCI) and 11 AD dementia patients. All participants underwent neuropsychological assessments and fecal microbiota analysis through 16S ribosomal RNA (rRNA) Illumina Miseq sequencing technique. Meso Scale Discovery (MSD) kits were used to quantify the plasma Aβ40, Aβ42, and Aβ42/Aβ40 in CN- and CN+ participants. Using Spearman's correlation analysis, the associations of global standard uptake value rate (SUVR) with altered gut microbiota and plasma Aβ markers were separately evaluated. Furthermore, the discriminative power of the combination of gut microbiota and plasma Aβ markers for identifying CN+ individuals was investigated.

Results: Compared with the CN- group, the CN+ group showed significantly reduced plasma Aβ42 (p = 0.011) and Aβ42/Aβ40 (p = 0.003). The relative abundance of phylum Bacteroidetes was significantly enriched, whereas phylum Firmicutes and class Deltaproteobacteria were significantly decreased in CN+ individuals in comparison with that in CN- individuals. Particularly, the relative abundance of phylum Firmicutes and its corresponding SCFA-producing bacteria exhibited a progressive decline tendency from CN- to CN+ and CI. Besides, the global brain Aβ burden was negatively associated with the plasma Aβ42/Aβ40 (r = -0.298, p = 0.015), family Desulfovibrionaceae (r = -0.331, p = 0.007), genus Bilophila (r = -0.247, p = 0.046), and genus Faecalibacterium (r = -0.291, p = 0.018) for all CN participants. Finally, the combination of plasma Aβ markers, altered gut microbiota, and cognitive performance reached a relatively good discriminative power in identifying individuals with CN+ from CN- (AUC = 0.869, 95% CI 0.782 ~ 0.955).

Conclusions: This study provided the evidence that the gut microbial composition was altered in preclinical AD. The combination of plasma Aβ and gut microbiota may serve as a non-invasive, cost-effective diagnostic tool for early AD screening. Targeting the gut microbiota may be a novel therapeutic strategy for AD.

Trial registration: This study has been registered in ClinicalTrials.gov (NCT03370744, https://www.clinicaltrials.gov ) in November 15, 2017.

Keywords: Alzheimer’s disease; Amyloid-β; Gut microbiota; Plasma; Preclinical.

Conflict of interest statement

The authors declare that they have no competing interests.

© 2022. The Author(s).

Figures

Fig. 1

Plasma Aβ levels for CN− and CN+ participants. Scatter plot presented mean with range. No significant difference was observed in plasma Aβ40 (A), whereas there were significant differences in plasma Aβ42 (B) and Aβ42/Aβ40 (C) between CN− and CN+. *p < 0.05; **p < 0.01. Aβ amyloid-β; CN−, amyloid-β-negative cognitively normal participants; CN+, amyloid-β-positive cognitively normal participants

Fig. 2

The overall gut microbiota community for CN− and CN+ participants. A The Venn diagram illustrated the overlap of ASVs between CN− and CN+, with 1929 ASVs shared in both groups, and 147 ASVs unique for CN+. B The key ASVs in both groups. The abundance of ASV_19 and ASV_9 was significantly reduced in CN+ compared with that in CN−. C The alpha diversity of gut microbiota between CN− and CN+. Each bar graph represented the mean and standard deviation. There were no significant differences in Chao1, ACE, Simpson, and Shannon indexes between CN− and CN+. D The PCoA based on the distribution of ASVs. The structure of gut microbiota in the CN+ group was not significantly different from that in the CN− group. *p< 0.05; **p < 0.01. CN− amyloid-β-negative cognitively normal participants, CN+ amyloid-β-positive cognitively normal participants, ASV amplicon sequence variants, PCoA principal coordinates analysis, PERMANOVA permutational multivariate analysis of variance

Fig. 3

The gut microbial compositions for CN− and CN+ participants. A The bacterial community in both groups at different taxonomic levels. Bar graphs indicated the relative abundance of phylum-level, class-level, order-level, and family-level taxa. B LEfSe analysis between CN− and CN+. As shown in the histogram of LDA scores for differentially abundant taxa, positive LDA scores indicated the enrichment of taxa in the CN+ group (green), and negative LDA scores indicated the enriched taxa in the CN− group (red). The LDA scores (log10) > 2 and p < 0.05 were listed. Cladogram indicated the phylogenetic distribution of gut bacteria. Colors represented different groups (CN−, red; CN+, green). Nodes with different colors represented important taxa in different groups. Yellow nodes suggested no significantly differential taxa between the two groups. C The heatmap showing the relative abundance and distribution of differentially abundant taxa identified by the LEfSe method. CN− amyloid-β-negative cognitively normal participants, CN+ amyloid-β-positive cognitively normal participants, LEfSe linear discriminant analysis (LDA) effect size

Fig. 4

The association of plasma Aβ markers and gut microbiota with global brain Aβ burden. A There was negative correlation between global SUVR and plasma Aβ42/Aβ40. The family Desulfovibrionaceae (B), genus Bilophila (C), and genus Faecalibacterium (D) were negatively correlated with the global SUVR. Aβ amyloid-β, SUVR standard uptake value rate

Fig. 5

ROCs for CN+ participants. A The discriminative power of individual plasma Aβ markers and the combined panel in identifying CN+ from CN−; B The discriminative power of each of the gut taxa and the combined panel in identifying CN+ from CN−; C The predicted values of the combined panels in identifying CN+ from CN−. Aβ, amyloid-β; CN−, amyloid-β-negative cognitively normal participants; CN+, amyloid-β-positive cognitively normal participants; panel 1, the combined plasma Aβ40, Aβ42, and Aβ42/ Aβ40; panel 2, the combined taxa 1, taxa 2, and taxa 3; panel 3, the combined clinical cognitive tests (including MoCA-B, AVLT-long delayed recall, AVLT-R, STT-A, STT-B, AFT, BNT); panel 4, the combined plasma Aβ markers, gut taxa, and cognitive tests