A longitudinal study of the diabetic skin and wound microbiome

View article
PeerJ

Main article text

 

Introduction

Materials & Methods

Study design, ethics approval, and sample collection

Extraction of microbial DNA from skin and wound swabs and wound debridement tissue

Preparation of 16S rRNA gene libraries for Illumina sequencing

Illumina sequencing and data analysis

Results

Cohort characteristics

Sample processing, 16S PCR and sequencing

The microbiome of diabetic skin is less diverse than control skin

The skin microbiome is significantly different between diabetic and control subjects

Abundant taxa from skin are similar between persons with diabetes and healthy controls

The foot skin microbiome may predict diabetic status

Stability of the diabetic skin microbiome over time

Microbiota of chronic diabetic wounds overlap with skin and differ between patient

Discussion

Conclusions

Supplemental Information

Summary of samples collected from each diabetic patient enrolled in the study

Sample types are indicated in the second column: skin swab from an area adjacent to the foot wound (SA), skin swab from the contralateral foot (SC), wound swab (WS) and debrided wound tissue (WD). A tick (x) indicates that a sample was collected, a cross (x) indicates that no samples for that time point were collected because the patient could not make their scheduled appointment, and NC indicates that a sample was not collected because podiatry staff deemed there was not sufficient tissue available for a debridement sample. In the case that a wound healed (indicated by WH) no further samples were collected for that patient. In the case of P4, the contralateral foot had previously been amputated, and the foot containing the wound was amputated after time point 2. Samples for which a 16S rRNA gene PCR product was not obtained are indicated with (x).

DOI: 10.7717/peerj.3543/supp-1

Summary of samples collected from each control subject enrolled in the study

Sample types are indicated in the second column: skin swab from the base of the left foot (L), or right foot (F). A tick (x) indicates that a sample was collected, a cross (x) indicates that no samples for that time point were collected because the subject was not available. Samples for which a 16S rRNA gene PCR product was not obtained are indicated with (x). Control subject 8 (CP8) was removed from the study as not enough samples were collected.

DOI: 10.7717/peerj.3543/supp-2

Patient wound location, size and treatment over time

Wound size is shown as height × width × depth in mm. All wounds were routinely irrigated with Prontosan solution, followed by dressing with either Allevyn foam for moist wound healing, Zetuvit dressing to remove excess wound exudate, Inadine antimicrobial dressing (10% povidone-iodine), or Acticoat flex (silver coated antimicrobial dressing). Decisions on wound dressing were made by the treating podiatrist or wound care nurse.

DOI: 10.7717/peerj.3543/supp-3

Sequence coverage

Coverage before and after quality filtering for each sample is indicated, along with the sequencing run that each sample was sequenced in.

DOI: 10.7717/peerj.3543/supp-4

Results for differential abundance between diabetic and control skin

The Wald test was used as implemented in DESeq2. The model health + subject:health was used, and the OTU table was filtered to remove OTUs with a count of less than 2, and/or present in less than 10% of samples. Log2FC was calculated relative to the control skin samples.

DOI: 10.7717/peerj.3543/supp-5

Results for differential abundance between control and diabetic skin—genera level

The Wald test was used as implemented in DESeq2. The model health + subject:health was used, and OTUs were collapsed into genera level classifications. The original OTU table was filtered to remove OTUs with a count of less than 2, and/or present in less than 10% of samples. Log2FC was calculated relative to the control skin samples.

DOI: 10.7717/peerj.3543/supp-6

Important classification OTUs

OTUs important to the classification in the Random Forrest Model calculated using the GINI index.

DOI: 10.7717/peerj.3543/supp-7

Results for differential abundance between diabetic skin and wounds

The Wald test was used as implemented in DESeq2. The model sampletype + sampletype:health was used, and the OTU table was filtered to remove OTUs present in less than 20% of samples. Log2FC was calculated relative to the diabetic skin samples.

DOI: 10.7717/peerj.3543/supp-8

Box plots comparing weighted unifrac distances between diabetic skin adjacent to wounds, and diabetic skin on the contralateral foot to wounds

Comparison as adjacent diabetic skin to wound swabs (adj_V_ws), contralateral diabetic skin to wound swabs (con_V_ws), adjacent diabetic skin to wound debridement (adj_V_wd), and contralateral diabetic skin to wound debridement (con_V_ws). Small but statistically significant differences were observed, such that there was greater similarity between adjacent skin and wounds compared to contralateral skin and wounds. Significant differences are indicated with ∗∗∗(p < 0.001).

DOI: 10.7717/peerj.3543/supp-9

Inter-visit weighted unifrac distance vs degree of healing

Inter-visit weighted unifrac distances from wound swabs for individual patients were plotted against the degree of healing, where 1 equals complete healing, 0 equals no change in wound size and negative numbers indicate and increase in wound size. No significant correlation was found using a linear model.

DOI: 10.7717/peerj.3543/supp-10

R script and input data for analysis

DOI: 10.7717/peerj.3543/supp-11

Additional Information and Declarations

Competing Interests

The authors declare there are no competing interests.

Author Contributions

Melissa Gardiner performed the experiments, analyzed the data, wrote the paper, prepared figures and/or tables, reviewed drafts of the paper.

Mauro Vicaretti, Jill Sparks and Michael Liu performed the experiments, reviewed drafts of the paper.

Sunaina Bansal and Stephen Bush analyzed the data, reviewed drafts of the paper.

Aaron Darling contributed reagents/materials/analysis tools, reviewed drafts of the paper.

Elizabeth Harry conceived and designed the experiments, contributed reagents/materials/analysis tools, reviewed drafts of the paper.

Catherine M. Burke conceived and designed the experiments, performed the experiments, analyzed the data, contributed reagents/materials/analysis tools, wrote the paper, prepared figures and/or tables, reviewed drafts of the paper.

Human Ethics

The following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers):

Ethical approval for the study was obtained from both the University of Technology Sydney Human Research Ethics Committee (approval number 2013000170), and the Western Sydney Local Health District Human Research Ethics Committee (approval number HREC2013/9/5.3(3809) AU RED LNR/13/WMEAD/294). Diabetic individuals and control subjects provided written consent for sample collection and all subsequent analyses.

Data Availability

The following information was supplied regarding data availability:

Quality filtered sequence data has been deposited in the European Nucleotide Archive under study accession number PRJEB17696.

Funding

Funding for this study was provided by internal grants from the University of Technology Sydney. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

105 Citations 10,953 Views 1,454 Downloads

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more