Visualizing the Effects of Sputum on Biofilm Development Using a Chambered Coverglass Model

The methods described herein allow for visualization of bacterial biofilms grown in the presence of exogenous products. Not surprisingly, the production of the exoproducts is of importance when using this type of system. For instance, Dithiothreitol (DTT), is often used on human sputum samples to help liquefy the samples. However, the effect of DTT alone can decrease biofilm development and viability (data not shown). Thus, proper controls for all conditions are necessary. Furthermore, the addition of human sputum products creates inherent variability on the experiment due to the fact that the sputum of each patient has a unique microbiome. To adjust for this, we have used pooled sputum samples to avoid patient specific results. Additionally, the choice of appropriate media is important when using any model system. Standard media for biofilm growth of the intended organism are recommended for initial setup of the system. If the goal of the experiment is to identify how exogenous products are affecting biofilm formation, a nutrient rich media that provides adequate biofilm formation is suggested. This will allow sufficient nutrition for growth while determining how exogenous factors may affect the biofilm. Other media, such as minimal or defined media, can be used to better mimic certain conditions. Other media have been used with good results in this system (data not shown). The attachment of the biofilms to the chambered coverglass is robust, however great care must be taken when removing/adding media or during washing steps of the biofilms to prevent disrupting the biofilms.

The use of the live-dead stain as per manufacturer's protocol has yielded good results for the system described. A number of factors can affect the fluorescence ratio of the images (including dye uptake of bacteria, relative density of biofilm and detector gain settings), however, this method should relate to what is observed in the images. Other measures may be used to represent the relative amount of dead/live biofilm, such as the biomass from the dead cells as a ratio of the total biomass present in the biofilm (as derived from COMSTAT). Using bacterial counts to confirm biofilm viability in repeated experiments is recommended to confirm the visual observations of this model. Due to inherent heterogeneity of biofilms, multiple images from each chamber and multiple chambers should be used for each condition in an experiment. This adds to the cost and duration of these experiments.

The acquisition of images is important prior to the analysis of data from these experiments. Care must be taken to not over saturate images and when determining the microscope setup. Depending on the level of detail required and the microscope available, a number of different objectives (10X, 25X, 40X and 63X) can be used to acquire images, though we find that 25X objective gives better images. The thickness of the Z-stack can also affect overall image quality and level of detail. Having images taken every 0.5-1 μm seems to provide clear images at 25X objective, while keeping the Z-stack images at a size that can be analyzed by COMSTAT on standard computer systems.

These experiments are more costly and time consuming than other attachment assays, and cannot be done in a high throughput manner. In this procedure, bacteria attach to a borosilicate surface, which is not reflective of an in vivo condition and may affect biofilm formation and development. However, they provide additional information and can generate hypotheses for mechanism related to biofilm antibiotic resistance and physiological response to exogenous signal. If the goal of the experiment is to understand how biofilms change in response to different factors, rather than identifying anti-biofilm compounds using high-throughput methods for instance, the additional information gained using this technique makes the method worthwhile. Thus the current method is a significant advancement over previous limited attachment assays such as the crystal violet assay, which is most commonly used to study biofilms.

Critical steps to this procedure include: 1) Ensuring timely processing of sputum samples to avoid degradation; 2) Being gentle with media changes on the biofilms to avoid disruption of the biofilm and 3) Doing repeated measurements of biofilm imaging to ensure an accurate representation of the biofilm thickness due to the heterogeneity of biofilm formation.

Once the system is set-up for a given bacterial species, one can test a number of different conditions, including the effects of exogenous factors on multiple clinical isolates. This system has been used to study the effect of antibiotics on biofilms exposed to human sputum¹⁶ and to compare highly resistant and intermediately resistant clinical isolates.¹⁷ Modifications to the chambered coverglass, such as coating the bottom with mucin or collagen micro-scaffold (e.g., puracol), can further extend the range of experiments possible. It may be possible to adapt this system to study direct interactions, such as those between epithelial cells and bacteria, or between different bacterial species. This is an expansion of the model described by Jurcisek et al.²⁰ The method uses slide chamber growth similar to that described here and uses formalin to fix biofilms prior to visualization. Alternatively, in this method we visualize biofilms immediately after staining and do not use a fixative. In addition, we add exogenous factors to test the effect of antibiotic on biofilms. This model thus has the potential to considerably further our understanding of bacterial biofilms in human disease.