Biofilm Assay for Clostridioides difficile with Applications for Drug Discovery

Summary

Biofilms of the bacterial pathogen Clostridioides difficile and their significance in disease are not well understood. Recent findings have suggested a role in recurrence, underscoring their importance. Here, we describe the adaptation of a biofilm assay coupled to a metabolic readout with applications for drug discovery against biofilms.

Abstract

Clostridioides difficile is a gastrointestinal bacterial pathogen able to take advantage of a dysbiotic microbiota environment to proliferate, secrete toxins, and damage the intestinal epithelium. A subset of C. difficile infection (CDI) patients will experience antibiotic (15%-30%) or fecal microbiota transplant (FMT) (<10%) treatment failure. Therefore, the development of additional therapeutic interventions is of critical importance. The role of C. difficile biofilms in recurrence is unclear. However, biofilms in other organisms are responsible for chronic and relapsing disease, suggesting this could also be the case in recurrent CDI. We hypothesize that biofilms of C. difficile present a valuable therapeutic target. The goal of the protocol presented here is to adapt a biofilm formation assay for the identification of repositionable compounds with activity against established C. difficile biofilms. The protocol refines a robust and reproducible assay for forming biofilms, couples it to a metabolic assay, and applies it to drug discovery. This protocol outlines the biofilm formation assay, biomass and metabolic activity readouts, drug susceptibility testing, drug screening of a repositioning library, and representative results.

Introduction

Clostridioides difficile is a spore-forming, anaerobic bacterial pathogen capable of inflicting severe damage to the human gastrointestinal (GI) tract by producing toxins. C. difficile is responsible for nearly half a million cases yearly, with ~30,000 fatalities (CDC, 2015). C. difficile infection (CDI) treatment adds a significant cost to the already strained healthcare system (~$4.8 billion)^1,2,3. Individuals most at risk are those with immunosuppression, antibiotic exposure, and/or the elderly, all populations that continue to significantly expand both in the U.S. and around the globe ^4,5,6,7.

C. difficile is highly dependent on the status of the host GI microbiota, and as such, antibiotic treatment is a key predisposing factor for CDI. The identification of antibiotic-resistant clinical isolates of C. difficile further complicates treatment^8,9,10. Moreover, between 15% and 30% of patients treated with antibiotics for CDI will fail initial treatment^{5,11,12,13,14,15,16}. Once a patient experiences a recurrence episode, the likelihood of recurrence episodes increases to 45%-64% (recurrence escalator) and is accompanied by an increased risk of further treatment failures^16,17. Importantly, most relapse cases of CDI are caused by the original strain that established the first episode of infection in the host^18,19, suggesting continued colonization by C. difficile even after successful treatment. While spores have been shown to persist in the GI tract and lead to relapse in ex vivo and murine models²⁰, C. difficile biofilms and their role in disease and relapse are underexplored. Biofilms are microbial communities protected by a self-produced extracellular matrix, a characteristic that makes them highly recalcitrant to environmental insults such as antibiotic treatment and immune responses²¹. Further, biofilm formation is a significant virulence factor^22,23,24,25 and in bacterial infections, such as those caused by Staphylococcus aureus and Pseudomonas aeruginosa, biofilms are key in recurrence and chronic manifestations^26,27. Recently, C. difficile biofilms have been suggested as a reservoir for recurrent infections^28,29,30 and therefore represent a high-value target for the treatment and/or prevention of CDI recurrence.

Current models to study C. difficile biofilms include colony biofilm models^31,32, microfermentors³³, chemostat gut models^34,35, and liquid culture systems using flasks or well-plates^36,37. All these models have yielded key information about C. difficile biofilms at distinct stages of development and in distinct environments. However, some of these models (i.e., microfermentors, chemostat gut models, colony biofilms) are not suitable for drug discovery efforts. To date, most studies involving drug discovery efforts have focused on the planktonic lifestyle of C. difficile^38,39,40,41. Therefore, the protocol described here aims to adapt, standardize, and validate the existing 96-well plate biofilm model with the goal of developing a platform for screening libraries to identify repositionable molecules with the ability to kill established (24 h old) C. difficile biofilms. Validation of the protocol was performed by comparing biofilm formation capabilities between a laboratory-adapted strain (630 Δerm) and an epidemic clinical isolate (UK1 strain isolated during a 2006 outbreak at Stoke-Mandeville Hospital in the United Kingdom⁴²). Biofilm formation was quantified using the classical Crystal violet assay to measure biomass and a metabolic assay to measure the viability of the biofilms. Significant differences in biofilm formation capabilities between the strains were identified, suggesting potential clinical implications. Further, the assay was utilized to determine antibiotic susceptibility profiles of established biofilms. Lastly, as proof of concept, the Global Health Priority box from Medicines for Malaria Venture (MMV), a non-profit organization focused on discovering new therapies against neglected diseases such as malaria, was screened to identify compounds with activity against established biofilms of C. difficile. The protocol described here adapts and validates a classical biofilm assay for use as a drug screening platform targeting mature biofilms.

Protocol

1. Preparing cells for the biofilm assay

CAUTION: Clostridioides difficile is a human pathogen and requires BSL-2 containment.

NOTE: Spores utilized for this study were acquired from Dr. Carol Kumamoto at Tufts University School of Medicine. C. difficile spore stocks were prepared as previously described^43,44,45. Spores are then aliquoted as 10 µL volumes into PCR tubes for streaking out plates. All media and Phosphate Buffer Saline (PBS) required for the experiment should be pre-reduced in the anaerobic chamber 24 h prior to use. This is accomplished by loosening the cap of the bottle containing the media or PBS inside the chamber.

1. Pre-reduce a Brain Heart Infusion agar plate supplemented with 0.1% (w/v) taurocholate and 0.1% (w/v) L-cysteine (BHIS) in the anaerobic chamber, preferably overnight or for at least 3 h to allow for full reduction. Set the temperature of the anaerobic chamber to 37 °C.
2. Heat the PCR tube containing 10 µL of spores for 20 min at 70 °C. This can be done in a water bath or thermocycler.
3. Bring spores into the anaerobic chamber, mix by gently pipetting 5x-6x, and pipette 10 µL on the pre-reduced BHIS plate. Perform a four-quadrant streak as previously described⁴⁶, and incubate the plate overnight at 37 °C (temperature of the anaerobic chamber).
4. After 24 h of growth, isolated colonies should be visible. Select a single colony and, using a disposable loop, transfer to a tube containing 10 mL of BHIS liquid. Incubate overnight (14-18 h).
5. Determine cell density by serially diluting the overnight culture in the chamber.
   1. Perform 10-fold dilutions of the overnight culture. The 10³ dilution yields a countable sample for each of the strains utilized in this protocol. However, standardization of this step might be required for clinical isolates with growth differences.
   2. Bring the tubes out of the anaerobic chamber, place 10 µL of each sample in a hemocytometer, and count using a brightfield microscope as described by the hemocytometer manufacturer.
   3. Follow the manufacturer's instructions to determine cell concentration after counting.

2. Biofilm formation

NOTE: While this protocol is adaptable to distinct well sizes, it was optimized for the 96-well platform. If the well size is modified, the cell concentration will need to be adjusted based on the surface area of the well.

1. After determining the cell concentration from the overnight culture, adjust the cell concentration to 1 x 10⁷ cells/mL in BHIS (supplemented with 10% (w/v) L-cysteine and 0.1M glucose).
   NOTE: The 0.1M glucose supplementation has previously been shown to increase biofilm formation by the C. difficile 630 Δerm strain⁴⁷. For this protocol, the concentration of 1 x 10⁷ cells/mL was selected after testing distinct starting concentrations (Figure 1). Robust biofilm formation can still be achieved without glucose supplementation.
2. Dispense 100 µL of cells into the wells of a 96-well microtiter plate (flat bottom and tissue culture treated), leaving control (cells only) and blank (media only) columns. Incubate for 24 h in the anaerobic chamber (37 °C).
   1. If media evaporation becomes an issue, place biofilm plates away from any heating unit. Alternatively, place biofilms inside an additional container or cover with permeable 96-well plate membranes.
3. Gently remove the media from all the wells, starting with the blank. This will remove non-adherent cells and leave behind a biofilm. This step is crucial as only cells within the biofilm are of interest.
4. Gently wash the biofilms 2x with 200 µL of pre-reduced PBS. To perform this wash gently, tilt the 96-well plate slightly, insert the pipette tips against the wall of the well, slide the tips close to the bottom of the well without scraping the bottom, and slowly aspirate. If PBS is left at the bottom, repeat the process again. At this stage, non-adherent cells can be removed, and it is normal. Remove the PBS before taking the plate out of the anaerobic chamber.
5. At this point, use biofilms for: Biomass measurements (See Step 3); metabolic activity measurements (See Step 4); drug susceptibility and drug screening assays (See Step 5).
   NOTE: This protocol has been validated for the lab adapted 630 Δerm strain and the epidemic UK1 strain. Distinct clinical isolates and strains might display weak biofilm forming capabilities. To minimize biofilm disruption, wash with PBS 1x instead of 2x and reduce washing volume to 100 µL.

3. Measurement of biofilm biomass using the Crystal violet assay

NOTE: Crystal violet staining was based on a previously described protocol⁴⁸.

1. After washing the biofilms in the anaerobic chamber (see step 2.4), remove the plate and transfer it to a chemical fume hood.
2. To fix the biofilms, gently add 150 µL of methanol to each well. Cover the plate with the lid for 10 min.
3. Remove methanol and allow wells to completely dry with the lid removed. This should take between 30 min and 45 min.
4. Add 100 µL of 0.3% Crystal violet to all wells and stain for 10 min. Remove Crystal violet carefully to prevent staining of the well rim and walls, as this can impact readout.
5. Allow the plate to dry with the lid removed. This should take between 30 min to 45 min.
6. Wash the wells 3x with 200 µL of distilled water and allow the plate to dry completely. This should take 30 to 45 minutes.
7. Once dry, use a brightfield microscope to image the biofilms (Figure 2A). While this step is optional, it allows for more thorough observation of the cells in the biofilm.
   1. Place the 96-well plate on an inverted brightfield microscope with a plate adaptor. Identify representative wells for each strain tested.
   2. Take images for each well at the desired magnification. The images presented in this protocol were taken at 40x magnification.
8. Once the desired images have been acquired, transfer the plate back to a chemical fume hood for de-staining.
9. To de-stain the biofilm, add 150 µL of 33% acetic acid and place on a shaker at medium speed for 10 min.
10. Remove 90 µL from each well and transfer to a new 96-well flat-bottom plate. Read absorbance at 595 nm using a plate reader (Figure 2B).

4. Measurement of biofilm viability using the metabolic dye PrestoBlue

1. Conduct the measurement of metabolic activity following the manufacturer's protocol for the assay with key modifications. Warm the plate reader to 37 °C before preparing the biofilms for the assay. A plate reader capable of measuring fluorescence is recommended.
   NOTE: Alternatively, metabolic activity can be read at OD₆₀₀; however, we have not found that approach to be sensitive enough for these assays. It is recommended that users review the manufacturer's protocol and standardize if necessary.
2. After washing the biofilms with PBS in the anaerobic chamber as described above (see step 2.4), remove the plate from the chamber.
3. Along with the plate, remove 9 mL of pre-reduced BHIS (supplemented with 10% (w/v) L-cysteine and 0.1 M glucose).
4. Quickly add 1 mL of PrestoBlue reagent to 9 mL of pre-reduced BHIS and vortex. Do not pre-reduce the reagent as this can affect reagent coloration (See Discussion).
5. Add 100 µL of BHIS-PrestoBlue mixture to each well. Place the plate in a plate reader with the following settings: Temperature at 37 °C, assay type as Kinetic; read time as 30 min; read frequency at every 2 min; fluorescence intensity depending on the type of plate reader available. The measurements described in this protocol were performed using excitation/emission wavelengths of 560 nm/ 590 nm, respectively.
6. After reading, discard the plates.
7. Once fluorescence values are acquired from the plate readouts, use blank (media only) wells to subtract the noise from all sample wells. After this, divide the biofilm wells treated with each compound by the control well containing an untreated (100%) biofilm to obtain metabolic activity.

5. Addition of drugs to biofilms

NOTE: The setup described below can be modified for drug susceptibility assays and fixed concentration drug screening. For both methods described below, the PrestoBlue assay is recommended.

1. Determination of minimum inhibitory concentrations (MIC) of distinct antibiotics
   ​NOTE: Biofilms are handled over a 48 h period, increasing the possibility of contamination. It is of high importance to process blank wells first and follow aseptic procedures. MIC values reported in the literature for other C. difficile strains and test dose response assays were used to inform starting antibiotic concentrations for the MIC assays presented in this protocol. This approach will be useful for determining starting antibiotic concentrations when working with clinical isolates.
   1. Wash the mature (24 h old) biofilms with PBS in the anaerobic chamber as described above (See step 2.4).
   2. Add 200 µL of pre-reduced BHIS (supplemented with 10% (w/v) L-cysteine and 0.1M glucose) media containing two-fold dilutions of the test antibiotic to each well except the control (cells only) and blank (media only) columns in the plate.
   3. Incubate for an additional 24 h. In this study, vancomycin (starting concentration of 200 µg/mL), metronidazole (starting concentration of 16 µg/mL), and fidaxomicin (starting concentration of 4 µg/mL) were utilized to validate the assay. MIC values are reported in Table 1.
   4. To prevent drying of plates, keep away from the heating unit in the anerobic chamber or use a gas permeable membrane.
      NOTE: Alternatively, a pre-reduced PBS or media moat (300 µL) could be used on the outer wells of the plate.
   5. After 24 h of incubation with the drugs, remove the media and gently wash the biofilms 1x with PBS. Process biofilms using the Crystal violet assay or the PrestoBlue assay.
2. Screening drug libraries
   1. Wash the mature (24 h old) biofilms with PBS in the anaerobic chamber as described above (See step 2.4).
   2. Add 200 µL of media containing a fixed concentration of drug library compounds to each well except the control (cells in media without antibiotics) and blank (media without antibiotics) columns in the plate. Incubate for an additional 24 h. To prevent drying of plates, keep away from the heating unit in the anerobic chamber.
   3. After 24 h of incubation with the drugs, remove the media and gently wash the biofilms 1x with PBS. Process biofilms using the Crystal violet assay or the metabolic assay.

Results

The aim of the protocol described above is to adapt and validate the 96-well plate biofilm formation model as a drug screening platform for the identification of repositionable compounds with activity against established biofilms of C. difficile. The impact of distinct cell densities (10⁴, 10⁵, 10⁶, and 10⁷ cells/mL) on biofilm formation was determined, and biofilm formation was quantified by Crystal violet staining as previously descri...

Discussion

C. difficile biofilm formation and the role it plays in disease are not well understood. Recent findings suggest a role for biofilms in disease relapse^28,34. Biofilm formation is a significant virulence factor^22,23,24,25 and in bacterial infections, such as those caused by Staphylococcus aureus and Pseudomonas aeru...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
ACETIC ACID SAF-CT ACS 500ML	Fisher Scientific	A38S500
AGAR BACTERIOLOGICAL GR 1KG	Fisher Scientific	AC443570010
BioTek Synergy H1 plate reader	Fisher Scientific	BTH1MG
BOTTLE YEAST EXTRACT 500G	Fisher Scientific	DF0127179
BRAIN HEART INFUSION BROTH	Fisher Scientific	CM1135R
Crystal Violet USP grade 100G	VWR	97061-850	For staining biofilms for biomass assays
CytoOne 96-well TC plate, flat bottom, clear, indiv.wrapped withlids, 50/case	USA Scientific	CC7682-7596
D-(+)-Glucose, anhydrous	Fisher Scientific	50712744
Dimethyl Sulfoxide for Molecular Biology (DMSO)	Millipore Sigma	D8418-250ML
FIDAXOMICIN 250MG	Fisher Scientific	AC468312500
L-CYSTEINE 98+% 1KG	Fisher Scientific	AAA104350B	For C. difficile media
METHANOL 99.8 ACROSEAL 1LT	Fisher Scientific	AC364390010
METRONIDAZOLE, 99% 5GR	Fisher Scientific	AC210340050
Phosphate buffer saline (PBS) 10X	Fisher Scientific	BP3991
PrestoBlue Cell Viability Reagent	Fisher Scientific	A13262
SILVERSEAL OPAQUE ADHES 100/CS	Fisher Scientific	7000379	Used to seal mother and daughter drug plates before storage
SODIUM TAUROCHOLATE 100MG	VWR	100291-598
VANCOMYCIN	Fisher Scientific	BP29581