Low-carbohydrate-high-protein diet accelerates diabetic vascular calcification via gut microbiota-dependent TUDCA depletion
Data files
Aug 17, 2026 version files 4.86 GB
Abstract
Low-carbohydrate-high-protein diet (LC-HPD) is widely adopted for weight loss and glycaemic control, yet its long-term cardiovascular safety remains controversial. This study aimed to investigate whether LC-HPD promotes diabetic vascular calcification (VC) and to elucidate the mechanistic role of gut microbiota and its metabolites. The effect of LC-HPD on diabetic VC was examined in induced (ApoE−/−) and spontaneous (db/db) mouse models of type 2 diabetes mellitus (T2DM). To establish causality, faecal microbiota transplantation (FMT), mono-colonisation, and depletion of Faecalibaculum rodentium (F. rodentium) were performed. Integrated metabolomics identified key metabolites, with mechanistic validation in vivo and in vitro.LC-HPD exacerbated diabetic VC and promoted osteogenic transdifferentiation of vascular smooth muscle cells (VSMCs). FMT from LC-HPD-fed donors recapitulated these effects, indicating a microbiota-dependent mechanism. 16S rRNA sequencing revealed marked enrichment of F. rodentium. Depletion of F. rodentium attenuated, whereas colonisation aggravated, aortic calcification and VSMC phenotypic switching. Metabolomics demonstrated that F. rodentium possesses bile salt hydrolase activity, hydrolysing tauroursodeoxycholic acid (TUDCA) to ursodeoxycholic acid, thereby reducing circulating TUDCA. TUDCA supplementation suppressed excessive endoplasmic reticulum (ER) stress, blocked VSMC osteogenic differentiation, and alleviated VC. In vitro experiments confirmed that TUDCA dose-dependently suppressed VSMC calcification by inhibiting ER stress. LC‑HPD accelerates diabetic VC by enriching F. rodentium, which depletes TUDCA and thereby unleashes ER stress‑driven VSMC osteogenesis. These findings identify the TUDCA/ER stress axis as a potential therapeutic target for diabetic vascular complications, suggesting that related microbiota–bile acid mechanisms in humans warrant further investigation.
Dataset DOI: 10.5061/dryad.v6wwpzhc8
Description of the data and file structure
This dataset contains a 16S rRNA gene sequencing for the induced T2DM mouse model (ApoE−/− male mice), a 16S rRNA gene sequencing for the induced T2DM mouse model (db/db male mice), an untargeted metabolomics and bile acid targeted metabolomics of mouse serum in induced T2DM mouse model. This study used two T2DM mouse models, integrating in vivo LC-HPD interventions, 16S rRNA gene sequencing, and metabolomics analysis to investigate the impact of LC-HPD on diabetic VC and the mediating role of the gut microbiota and its metabolites.
Files and variables
File: Microbiome_(16S_in_induced_T2DM)_fastq.zip
Description: This dataset was generated to assess the gut microbiome of Ctrl group (ApoE−/− ), T2DM group (ApoE−/− induced T2DM model),T2DM+LC-HPD group (ApoE−/− induced T2DM model+LC-HPD intervention), and LC-HPD group (ApoE−/− +LC-HPD intervention).
File naming convention: sample name_Laboratory abbreviation_File creation date_Phase_Sample testing date_File sorting_R1/R2
sample name:
C1: faeces sample 1 of the Ctrl group
C2: faeces sample 2 of the Ctrl group
C4: faeces sample 3 of the Ctrl group
C6: faeces sample 4 of the Ctrl group
C7: faeces sample 5 of the Ctrl group
C9: faeces sample 6 of the Ctrl group
L11: faeces sample 1 of the LC-HPD group
L13: faeces sample 2 of the LC-HPD group
L14: faeces sample 3 of the LC-HPD group
L16: faeces sample 4 of the LC-HPD group
L17: faeces sample 5 of the LC-HPD group
L19: faeces sample 6 of the LC-HPD group
T21: faeces sample 1 of the T2DM group
T23: faeces sample 2 of the T2DM group
T25: faeces sample 3 of the T2DM group
T26: faeces sample 4 of the T2DM group
T27: faeces sample 5 of the T2DM group
T29: faeces sample 6 of the T2DM group
TL36: faeces sample 1 of the T2DM+LC-HPD group
TL38: faeces sample 2 of the T2DM+LC-HPD group
TL39: faeces sample 3 of the T2DM+LC-HPD group
TL40: faeces sample 4 of the T2DM+LC-HPD group
TL42: faeces sample 5 of the T2DM+LC-HPD group
TL44: faeces sample 6 of the T2DM+LC-HPD group
Laboratory abbreviation: The laboratory where the author works
File creation date：The time when the original data was generated
Phase：Experimental stage
Sample testing date：The time when the sample began to be tested
File sorting：The files are naturally sorted in the testing machine.
R1/R2: Paired-end sequencing read direction. R1 = forward read, R2 = reverse read. Each R1 and R2 file pair represents the two ends of the same sequencing library, not biological replicates.
File: Microbiome_(16S_in_spontaneous_T2DM)_fastq.zip
Description: This dataset was generated to assess the gut microbiome of Ctrl group (WT), T2DM group (db/db spontaneous T2DM model),T2DM+LC-HPD group (db/db spontaneous T2DM model+LC-HPD intervention), and LC-HPD group (WT+LC-HPD intervention).
File naming convention: sample name_null_Sample testing date_File sorting_R1/R2
sample name:
C46: faeces sample 1 of the Ctrl group
C47: faeces sample 2 of the Ctrl group
C48: faeces sample 3 of the Ctrl group
C50: faeces sample 4 of the Ctrl group
C51: faeces sample 5 of the Ctrl group
C53: faeces sample 6 of the Ctrl group
L70: faeces sample 1 of the LC-HPD group
L71: faeces sample 2 of the LC-HPD group
L72: faeces sample 3 of the LC-HPD group
L73: faeces sample 4 of the LC-HPD group
L74: faeces sample 5 of the LC-HPD group
L76: faeces sample 6 of the LC-HPD group
T2: faeces sample 1 of the T2DM group
T3: faeces sample 2 of the T2DM group
T4: faeces sample 3 of the T2DM group
T6: faeces sample 4 of the T2DM group
T8: faeces sample 5 of the T2DM group
T9: faeces sample 6 of the T2DM group
TL27: faeces sample 1 of the T2DM+LC-HPD group
TL30: faeces sample 2 of the T2DM+LC-HPD group
TL31: faeces sample 3 of the T2DM+LC-HPD group
TL33: faeces sample 4 of the T2DM+LC-HPD group
TL35: faeces sample 5 of the T2DM+LC-HPD group
TL36: faeces sample 6 of the T2DM+LC-HPD group
null: The omission due to the failure to abbreviate the laboratory institution name
Sample testing date：The time when the sample began to be tested
File sorting：The files are naturally sorted in the testing machine.
R1/R2: Paired-end sequencing read direction. R1 = forward read, R2 = reverse read. Each R1 and R2 file pair represents the two ends of the same sequencing library, not biological replicates.
File: Serum_bile_acid_targeted_metabolomics_in_induced_T2DM.zip
Description: This dataset was generated to assess the serum bile acid metabolites of T2DM group (ApoE−/− induced T2DM model),T2DM+F. rodentium group (ApoE−/− induced T2DM model+F. rodentium ).
File naming convention: sample name
sample name:
T1: Serum sample 1 of the T2DM group
T2: Serum sample 2 of the T2DM group
T3: Serum sample 3 of the T2DM group
T4: Serum sample 4 of the T2DM group
T5: Serum sample 5 of the T2DM group
T6: Serum sample 6 of the T2DM group
TF1: Serum sample 1 of the T2DM+F. rodentium group
TF2: Serum sample 2 of the T2DM+F. rodentium group
TF3: Serum sample 3 of the T2DM+F. rodentium group
TF4: Serum sample 4 of the T2DM+F. rodentium group
TF5: Serum sample 5 of the T2DM+F. rodentium group
TF6: Serum sample 6 of the T2DM+F. rodentium group
File naming convention: serial number _1_Quality control sample name_number
Serial number: The serial number of the testing machine
1: The plan for the first sequencing on that day
Quality control sample name_number：
QC-1
QC-2
QC-3
File: Serum_non-target_metabolomics_in_induced_T2DM.zip
Description: This dataset was generated to assess the serum metabolites of T2DM group (ApoE−/− induced T2DM model),T2DM+F. rodentium group (ApoE−/− induced T2DM model+F. rodentium ).
File naming convention: sample name_NEG/POS
sample name:
T1: Serum sample 1 of the T2DM group
T2: Serum sample 2 of the T2DM group
T3: Serum sample 3 of the T2DM group
T4: Serum sample 4 of the T2DM group
T5: Serum sample 5 of the T2DM group
T6: Serum sample 6 of the T2DM group
TF1 Serum sample 1 of the T2DM+F. rodentium group
TF2: Serum sample 2 of the T2DM+F. rodentium group
TF3: Serum sample 3 of the T2DM+F. rodentium group
TF4 Serum sample 4 of the T2DM+F. rodentium group
TF5: Serum sample 5 of the T2DM+F. rodentium group
TF6 Serum sample 6 of the T2DM+F. rodentium group
NEG/POS：They respectively represent the negative ion mode (Negative mode) and the positive ion mode (Positive mode) during the mass spectrometry data acquisition.
File naming convention: Quality control sample name_NEG/POS
Quality control sample name：
QC-1
QC-2
QC-3
QC-4
NEG/POS：They respectively represent the negative ion mode (Negative mode) and the positive ion mode (Positive mode) during the mass spectrometry data acquisition.
Code/Software
R (version 4.2.2) was used for all primary analyses. The .mzml files also can be opened using ProteoWizard (with SeeMS).