Sequence analysis of microbiota in clinical human cases with diabetic foot ulcers from China

Abstract

Background: Diabetic foot ulcers (DFU) seriously threaten the health and quality of life of patients. The microbiota is the primary reason for the refractory and high recurrence of DFU. This study aimed to determine the wound microbiota at different DFU stages.

Methods: Wound samples were collected from 48 patients with DFU and divided into three phases: inflammatory (I, n = 49), proliferation (P, n = 22), and remodeling (R, n = 19). The wound samples obtained at different stages were then subjected to 16S rRNA gene sequencing. The number of operational taxonomic units (OTUs) in the different groups was calculated according to the criterion of 97 % sequence similarity. The diversity of the microbiota differentially presented bacterial taxa at the phylum and genus levels, and important phyla and genera in the different groups were further explored.

Results: After sequencing, 3351, 925, and 777 OTUs were observed in groups I, P, and R, respectively, and 175 OTUs overlapped. Compared with the inflammatory stage, the α-diversity of wound microbiota at proliferation and remodeling stages was significantly decreased (P < 0.05). At the phylum level, Firmicutes, Proteobacteria, Actinobacteriota, and Bacteroidota were the dominant phyla, accounting for more than 90 % of all the phyla. At the genus level, Random Forest and linear discriminant analysis effect size analyses showed that Peptoniphilus, Lactobacillus, Prevotella, Veillonella, Dialister, Streptococcus, and Ruminococcus were the signature wound microbiota for the inflammatory stage; Anaerococcus, Ralstonia, Actinomyces, and Akkermansia were important species for the proliferation stage; and the crucial genera for the remodeling stage were Enterobacter, Pseudomonas, Sondgrassella, Bifidobacterium, and Faecalibacterium.

Conclusions: There were significant differences in the composition and structure of the wound microbiota in patients with DFU at different stages, which may lay a foundation for effectively promoting wound healing in DFU.

Keywords: Diabetic foot ulcers; Inflammation stage; Proliferation stage; Remodeling stage; Wound microbiota.

Fig. 1

Overall structure of the microbiota in the wounds of diabetic foot ulcers (DFU) at different stages. (A) Venn diagram of annotated wound microbiota (OTUs) in different groups. (B) The α-diversity of the wound microbiota including Chao1, Good's coverage, Simpson, Pielou's evenness, Faith's PD, Shannon, and observed species in the different groups. I: inflammatory stage; P: proliferation stage; R: remodeling stage.

Fig. 2

Specific microbiota in DFU wounds at different stages at the phylum level. (A) Distribution of the top 10 dominant phyla of the wound microbiota in different groups. (B) Clustering heatmap of the top 10 phyla in different groups. (C) Differences in the top 10 phyla among the different groups using Random Forest analysis. The heatmap shows the abundance distribution of the wound microbiota phyla in different samples, and the importance of species to the model decreases from top to bottom. I: inflammatory stage; P: proliferation stage; R: remodeling stage.

Fig. 3

Specific microbiota in DFU wounds at different stages at the genus level. (A) Distribution of the top 30 dominant genera of the wound microbiota in the different groups. (B) Cluster heatmap of the top 30 genera in the different groups. (C) Differences in the top 30 genera in the different groups using Random Forest analysis. The heatmap shows the abundance distribution of the genera of the wound microbiota in the different samples, and the importance of species in the model decreases from top to bottom. I: inflammatory stage; P: proliferation stage; R: remodeling stage.

Fig. 4

Linear discriminant analysis (LDA) effect size (LEfSe) was employed to identify biomarkers of bacterial communities among different groups at different levels. The current low-density LDA threshold is 2. I: inflammatory stage; P: proliferation stage; R: remodeling stage.

Fig. 5

Functional analysis of the annotated microbiota in the wounds of DFU using the MetaCyc database. The horizontal coordinate represents the abundance of functional pathways, the ordinate represents the item of the functional pathway of the second MetaCyc classification level, and the far right represents the first-level classification to which the pathway belongs.

Fig. 6

Significantly different metabolic pathways enriched by bacterial communities among different groups. Significantly different metabolic pathways were enriched in bacterial communities between the inflammatory and proliferation stages (A) and between the inflammatory and remodeling stages (B). I: inflammatory stage; P: proliferation stage; R: remodeling stage.

Fig. 7

Correlation analysis between the top 50 genus at different stages and clinical indices in DFU. Correlation between genus and clinical indices at the inflammatory (A), proliferation (B), and remodeling (C) stages. BMI: body mass index; FBG: fasting blood glucose; HbA1c: glycated hemoglobin; TC: total cholesterol; TG: triglyceride; HDL: high-density lipoprotein; LDL: low-density lipoprotein; RBC: red blood cell; Hb: hemoglobin; WBC: white blood cells; N: nitrogen; CRP: C-reactive protein; PCT: procalcitonin; ESR: erythrocyte sedimentation rate.