This study aimed to investigate how dietary vitamin A affects the gut microbiota and intestinal tissue, influencing intestinal permeability, inflammatory factors, and Aβ pathology. The APP/PS1-AD mouse model was used, and mice were fed different vitamin A diets for 12 weeks. The groups included vitamin A-deficient (VAD), normal vitamin A (VAN), and vitamin A-supplemented (VAS). No significant changes in food intake and body weight were observed among the groups. However, the VAD and VAS groups showed reduced food intake compared to the VAN group at various time points. In terms of cognitive function, the VAN group performed better in the Morris Water Maze Test, indicating superior learning and memory abilities. The VAD and VAS groups exhibited impaired performance, with the VAS group performing relatively better than the VAD group. Serum vitamin A concentrations differed significantly among the groups, with the VAS group having the highest concentration. Aβ levels were significantly higher in the VAD group compared to both the VAN and VAS groups. Microbial analysis revealed that the VAS and VAN groups had higher microbial diversity than the VAD group, with specific taxa characterizing each group. The VAN group was characterized by taxa such as Actinohacteriota and Desulfovibrionaceae, while the VAD group was characterized by Parabacteroides and Tannerellaceae. The VAS group showed similarities with both VAN and VAD groups, with taxa like Desulfobacterota and Desulfovibrionaceae being present. The VAD vs. VAS, VAD vs. VAN, and VAS vs. VAN comparisons identified 571, 313, and 243 differentially expressed genes, respectively, which associated with cellular and metabolic processes, and pathway analysis revealed enrichment in pathways related to chemical carcinogenesis, drug metabolism, glutathione metabolism, and immune-related processes. The VAD group exhibited higher levels of D-lactate, diamine oxidase, and inflammatory cytokines (TNF-α, IL-1β, IL-6) compared to the VAN and VAS groups. In conclusion, dietary vitamin A modulates the gut microbiota, intestinal permeability, inflammatory factors, and Aβ protein formation, providing insights into AD and potential therapeutic avenues for further investigation.

Experimental design and collection of small intestine tissue samples：

In this study, we utilized thirty male APP/PS1 mice (4 weeks old) and randomly allocated them into three groups based on their body weight: VAD, VAN, and VAS groups. The mice were fed with AIN93G diet supplemented with varying levels of vitamin A: 0 IU/g (VAD group), 4 IU/g (VAN group), or 15 IU/g (VAS group). Following a 12-week experimental period, the mice underwent a 12-hour fasting period, followed by intraperitoneal injection of sodium pentobarbital for anesthesia, and subsequent blood collection from the eyeballs. Subsequently, the mice were euthanized, and intestinal tissues were collected. A 10-centimeter segment from the upper part of the small intestine was isolated and washed with PBS. These tissues were promptly frozen in liquid nitrogen and stored at -80°C for subsequent RNA sequencing and transcriptome analysis.

Experimental procedure for transcriptome sequencing analysis:

1.The workflow of mRNA sequencing includes sample preparation, library construction, library quality control and sequencing.

1.1 RNA Quality Assessment

Purity, concentration and integrity of RNA sample were examined by NanoDrop, Qubit 2.0, Agilent 2100, etc. Only RNA with good quality could move on to following procedures.

1.2 Library Construction

Qualified RNA were processed for library construction. The procedures are described as follow:

(1) mRNA was isolated by Oligo(dT)-attached magnetic beads.

(2) mRNA was then randomly fragmented in fragmentation buffer.

(3) First-strand cDNA was synthesized with fragmented mRNA as template and random hexamers as primers, followed by second-strand synthesis with addition of PCR buffer, dNTPs, RNase H and DNA polymerase I. Purification of cDNA was processed with AMPure XP beads.

(4) Double-strand cDNA was subjected to end repair. Adenosine was added to the end and ligated to adapters. AMPure XP beads were applied here to select fragments within size range of 300-400 bp.

(5) cDNA library was obtained by certain rounds of PCR on cDNA fragments generated from step 4.

1.3 Library Quality Control

In order to ensure the quality of library, Qubit 2.0 and Agilent 2100 were used to examine the concentration of cDNA and insert size. Q-PCR was processed to obtain a more accurate library concentration. Library with concentration larger than 2 nM is acceptable.

1.4 Sequencing

The qualified library was pooled based on pre-designed target data volume and then sequenced on Illumina sequencing platform.

2. Clean data with high quality was obtained by filtering Raw data, which removes adapter sequence and reads with low quality. These clean data were further mapped to pre-defined reference genome generating mapped data. Assessment on insert size and sequencing randomness were processed on mapped data as library quality control. Basic analysis on mapped data included gene expression quantification, alternative splicing analysis, novel genes prediction and genes structure optimization.

3. Data Quality Control

Based on sequencing-by-synthesis (Sequencing By Synthesis, SBS) technology, cDNA libraries were sequenced on Illumina high-throughput platform, generating significant amounts of high-quality data known as raw data. Raw data was saved in FASTQ format. Each sample has two FASTQ file, containing cDNA reads measured at both ends respectively.

4.Sequencing Quality Control

It is crucial to ensure the quality of the reads before moving onto following analysis. Raw data contains useless data such as primers, adapters, etc., which need to be removed before analysis. Procedures for data quality control were listed as follow:

(1) Trim adapter contaminations

(2) Remove nucleotides with low Quality-score.

Data processed by above steps is named "Clean data". Clean data was provided in FASTQ format.