Diabetic foot wounds are commonly colonised by taxonomically diverse microbial communities and may additionally be infected with specific pathogens. Since biofilms are demonstrably less susceptible to antimicrobial agents than are planktonic bacteria, and may be present in chronic wounds, there is increasing interest in their aetiological role. In the current investigation, the presence of structured microbial assemblages in chronic diabetic foot wounds is demonstrated using several visualization methods. Debridement samples, collected from the foot wounds of diabetic patients, were histologically sectioned and examined using bright-field, fluorescence, and environmental scanning electron microscopy and assessed by quantitative differential viable counting. All samples (
The aetiological role of biofilms in diabetic wounds remains poorly understood but their formation is increasingly recognised as a potential barrier to healing [
Whilst various definitions for biofilms have been adopted, it is generally accepted that they are structured bacterial communities that are often but not always attached to surfaces and which are encased in a self-produced exopolymer matrix [
The current pilot study was conducted as part of a larger study into the presence of unculturable bacteria in diabetic foot wounds [
Unless otherwise stated chemicals used were supplied by Sigma (Poole, Dorset, UK). Dehydrated bacteriological media were obtained from Oxoid (Basingstoke, Hampshire, UK) and prepared according to instructions supplied by the manufacturer.
This study was reviewed by the North Manchester Research Ethics Committee and the Central Manchester University Hospital Research and Development Department. Reference number: 09/H1006/41, protocol number 1.0. Twenty-six wound tissue debridement samples were collected from patients with chronic diabetic foot wounds (defined as being distal to the medial and lateral malleoli, with a known duration greater than four weeks), attending a specialist foot clinic. Wound tissue samples were taken from the wound bed and surrounding tissue using a sterile scalpel by the attending clinician and placed sterile 0.85% (w/v) saline for transportation. All samples were transported to the laboratory at 2°C and processed within 3 h of collection.
Twenty-six tissue samples were processed as previously described [
Residual chronic wound tissue from four samples (which were of sufficient quantity for multiple microscopic analyses) were divided transversely (50 : 50) with a sterile scalpel and one section was embedded in optimal cutting temperature (OCT) embedding matrix and frozen at –80°C for ≥24 h. The remaining tissue sections were placed in a sterile Bijou bottle and transported immediately for ESEM imaging. To produce slide-mounted tissue sections to visualise microcolonies and biofilm architecture, OCT-embedded whole tissue samples were sectioned to a thickness of 5
Slide-mounted tissue sections were fixed in 4% paraformaldehyde for 3 h and then subjected to a prepermeabilization step, consisting of lysozyme enzymatic buffer (100 mM Tris HCl [pH8], 50 mM EDTA, and lysozyme [5 mg/mL]), for 4 h at 45°C. Slides were then washed in wash buffer consisting of 0.9 M NaCl and 20 mM Tris and air-dried. Slides were then incubated in FISH buffer containing 50% formamide, 0.9 M NaCl, 20 mM Tris, 0.01% SDS (w/v), and 50 ng of the general eubacterial probe (EUB 338)-cy3 probe-GCT GCC TCC CGT AGG AGT [
FISH images were captured using an Olympus BX51 upright microscope using a 60x and 100x objective and captured using a Coolsnap ES camera (Photometrics, AZ, USA) through MetaVue Software (Molecular Devices, CA, USA). Specific band pass filter sets for DAPI (Ex. BP365/12 nm, Em. LP397 nm), FITC (Ex. BP450–490 nm, Em. BP515–565), and Texas red (Ex. BP546/12 nm Em. LP615 nm) were used. Gram-stained images were visualized using a Zeiss Axioscop 2 microscope, Axiocam, and Axiovision Version 4.8 (Carl Zeiss Ltd., Herefordshire, UK). All images were then processed using ImageJ (
Chronic wound tissue was placed in a sterile Bijou and transported immediately for ESEM imaging. ESEM of chronic wound tissue samples was performed using a FEI Quanta 200 environmental scanning electron microscope under a low vacuum (<0.75 Torr) permitting inspection of putative biofilm structures and microcolonies whilst conserving the hydrated state of the sample.
All 26 tissues samples harboured aerobic and facultative anaerobic species at cell densities equal to or greater than 5 log10 CFU/g of tissue, as shown in Figure
Differential viable counts of selected bacterial groups from the four imaged samples.
Sample | Total: | ||||
---|---|---|---|---|---|
Aerobic count | Anaerobic count | Staphylococci | Coliforms | Streptococci | |
1 | 10.31 | 9.07 | 8.93 | ND | 7.73 |
2 | 10.46 | 10.49 | 5.93 | ND | 10.40 |
3 | 7.28 | 7.33 | 7.24 | ND | ND |
4 | 8.39 | 8.36 | ND | 8.49 | ND |
Mean* |
8.15 (1.30) | 7.87 (1.72) | 7.09 (1.77) | 3.20 (4.27) | 3.37 (3.90) |
Values are
Differential viable counts of selected bacterial groups from 26 chronic wound samples. The lower and upper boundaries of the boxes represent quartiles 1 and 3, respectively, and horizontal bars within the boxes represent median values. ○ represents minimum outliers and ● the maximum outliers. White bars represent samples from which pathogens [
Figure
Images acquired from Sample 1. (a) and (b) are replicate images from Gram-stained sections; (c) and (d) (replicates) have been visualized using a combination of FISH (red), to indicate eubacteria, ConA (green) to indicate biofilm matrix, and other ConA-reactive material, and with Hoechst 33252 (blue) for the detection of nucleic acids. (e) and (f) show replicate ESEM images. Presumptive bacterial microcolonies and biofilm matrix have been indicated by arrows.
Images acquired from Sample 2. See legend to Figure
Images acquired from Sample 3. See legend to Figure
Images acquired from Sample 4. Putative biofilm matrix is indicated by arrows. See legend to Figure
Based on localization of reactive material, the utility of ConA and Hoechst 33252 as specific biofilm indicators is limited by the reactivity of structures associated with host cells. However, it is likely that Con-A-reactive material adjacent to bacterial microcolonies (as indicted by the FISH probe) is biofilm matrix. This is particularly evident in Figures
The taxonomically diverse microbial communities which occur in diabetic foot wounds may include both aerobic and anaerobic organisms many of which are potentially pathogenic [
A limitation of biofilm matrix staining using a carbohydrate marker such as the (fluorescently labeled) carbohydrate-binding lectin, Concanavalin-A, is the fact that reactive materials are also commonly associated with mammalian cells. It is, therefore, important to consider the location of reactive material. The feasibility of this approach may be enhanced by using FISH-probes for bacteria and a nucleic acid stain such as Hoechst 33252.
When exploring biofilms using scanning electron microscopy, a high level of resolution and detail can be obtained, potentially revealing biofilm-specific morphology, but also individual cells and their spatial location. Exopolymer matrix is amorphous material which may appear as a layer covering the biofilm, or as a fibrous material. Preparation of the sample for SEM involves dehydrating the sample which can affect the overall morphology of the biofilms and the appearance of the biofilm matrix. These considerations can partially be overcome with the application of ESEM or cryo-SEM which preserve the hydrated state of the biofilm.
In the present study, examination of slide-mounted, Gram-stained tissue sections revealed microcolonies attached to tissues which are indicative of the biofilm phenotype. These microcolonies comprised Gram-positive cocci (Samples 1–3), which corresponds to the organisms isolated by culture, coagulase negative staphylococci (Sample 1), and
The tissue samples were imaged further using ESEM. To conserve their hydrated state inspection of tissue surfaces was performed using an ESEM under a low vacuum (<0.75 Torr). Whilst ESEM is a method which requires access to the specialised equipment and training to ensure conservation of biofilm architecture and tissues, the images generated generally agreed with those gathered using the less complex methods using Gram and fluorescent staining, with microcolonies and/or amorphous substances (indicative of biofilms) identified in all samples.
The three visualisation techniques involved staining, fluorescence, and high-resolution microscopy to identify structures typical of the biofilm phonotype. Whilst the data presented represent a relatively small sample size, the outcomes of each method were broadly congruent. Since each method detects at least two of the three criteria (i) microbial surface attachment [
The growing interest in the role biofilms play in chronicity and impaired healing of diabetic wounds has led to an increased clinical requirement for a simple means of identifying biofilms in wound samples. More readily available methods such as Gram staining and bright-field microscopy can efficiently detect microcolonies associated with the biofilm phenotype and may therefore be of use for the identification of biofilms where expediency and cost-effectiveness are required.
The authors declare that there is no conflict of interests regarding the publication of this paper.