Putative role of beta-1,3 glucans in Candida albicans biofilm resistance

Abstract

Biofilms are microbial communities, embedded in a polymeric matrix, growing attached to a surface. Nearly all device-associated infections involve growth in the biofilm life style. Biofilm communities have characteristic architecture and distinct phenotypic properties. The most clinically important phenotype involves extraordinary resistance to antimicrobial therapy, making biofilm infections very difficulty to cure without device removal. The current studies examine drug resistance in Candida albicans biofilms. Similar to previous reports, we observed marked fluconazole and amphotericin B resistance in a C. albicans biofilm both in vitro and in vivo. We identified biofilm-associated cell wall architectural changes and increased beta-1,3 glucan content in C. albicans cell walls from a biofilm compared to planktonic organisms. Elevated beta-1,3 glucan levels were also found in the surrounding biofilm milieu and as part of the matrix both from in vitro and in vivo biofilm models. We thus investigated the possible contribution of beta-glucans to antimicrobial resistance in Candida albicans biofilms. Initial studies examined the ability of cell wall and cell supernatant from biofilm and planktonic C. albicans to bind fluconazole. The cell walls from both environmental conditions bound fluconazole; however, four- to fivefold more compound was bound to the biofilm cell walls. Culture supernatant from the biofilm, but not planktonic cells, bound a measurable amount of this antifungal agent. We next investigated the effect of enzymatic modification of beta-1,3 glucans on biofilm cell viability and the susceptibility of biofilm cells to fluconazole and amphotericin B. We observed a dose-dependent killing of in vitro biofilm cells in the presence of three different beta-glucanase preparations. These same concentrations had no impact on planktonic cell viability. beta-1,3 Glucanase markedly enhanced the activity of both fluconazole and amphotericin B. These observations were corroborated with an in vivo biofilm model. Exogenous biofilm matrix and commercial beta-1,3 glucan reduced the activity of fluconazole against planktonic C. albicans in vitro. In sum, the current investigation identified glucan changes associated with C. albicans biofilm cells, demonstrated preferential binding of these biofilm cell components to antifungals, and showed a positive impact of the modification of biofilm beta-1,3 glucans on drug susceptibility. These results provide indirect evidence suggesting a role for glucans in biofilm resistance and present a strong rationale for further molecular dissection of this resistance mechanism to identify new drug targets to treat biofilm infections.

FIG. 1.

Transmission electron microscopy of biofilm (right) and planktonic (left) cell walls. The images are representative of cell walls examined in more than 10 fields (50 cells from each condition). Brackets indicate cell walls. The magnification for the two images is ×110,000.

FIG. 2.

Biofilm and planktonic cell wall carbohydrate composition. Biofilm cells were collected from the in vitro silicone disk assay. Planktonic cells were collected in both log and stationary phases of growth. Each bar color represents a different cell wall fraction, with means and standard deviations from three biologic and three technical assay replicates. The data are expressed as micrograms of carbohydrate per 5 mg of dry cell weight. Carbohydrate concentrations of various fractions were estimated using the phenol-sulfuric acid methods with a glucose standard curve. *, biofilm cells contained significantly more β-1,3 glucan compared to planktonic cells in both stationary and log phases (P < 0.001).

FIG. 3.

Extracellular β-1,3 glucan concentrations from C. albicans biofilm and planktonic C. albicans in vitro and in vivo. The silicone disk assay was used for in vitro biofilm formation and for matrix isolation. The extracellular sample from the in vitro model represents cell-free culture supernatant. The rat central venous catheter model was used for in vivo biofilm formation. Biofilm formation was confirmed with microscopy. A disseminated candidiasis model was used as a control. All samples were collected after 12 h of biofilm formation. Viable cell counts were determined by plate counts and were similar for the biofilm and planktonic cultures at the time of sample collection. Each bar represents the mean and standard deviation from two biologic and three technical replicates. The black bars represent data from biofilm cultures, and the gray bars represent data from planktonic cultures. β-1,3 Glucan content was measured using the limulus lysate assay. Data are expressed as picograms of β-1,3 glucan per milliliter of sample. Both biofilm systems contained significantly more β-1,3 glucan compared to planktonic conditions (P < 0.001).

FIG. 4.

Fluconazole binding to biofilm and planktonic cell wall, culture supernatant, and biofilm matrix. The silicone disk assay was used for biofilm formation and for matrix isolation. Planktonic cells were collected in the log phase of growth. All samples were collected after 24 h of biofilm formation. For the cell wall study, 3 mg of dry cell wall weight from each condition was used. Viable cell counts were determined by plate counts and were similar for the biofilm and planktonic cultures at the time of supernatant collection. Following incubation of the culture component with fluconazole, unbound drug was separated using size exclusion dialysis. Fluconazole concentration was measured using a microbiologic assay. Each data value represents the mean and standard deviation from two biologic and three assay replicates. The black vertical bars represent data from the biofilm culture, and the gray bar represents data from the planktonic culture. Biofilm cell walls, supernatant, and matrix were associated with increased fluconazole binding compared to respective planktonic controls (P < 0.001).

FIG. 5.

Impact of glucanase on biofilm cell viability, determined using three β-glucanase preparations and two additional components (proteinase K and mannanase) of the predominant β-glucanase preparation (zymolyase). The 96-well polystyrene in vitro biofilm model was used. Cell viability was assessed using the XTT reduction assay. Viability is expressed as the optical density (OD) at 490 nm. The assays were performed in triplicate, and results are expressed as the mean and standard deviation. All drug exposures were of 48-h duration. (A) Biofilm cell viability following exposure to β-glucanase from Rhizoctonia over a concentration range from 0 to 5 units/ml. (B) Viability following exposure to β-1,3 glucanase from Arthrobacter luteus (zymolyase) over a concentration range from 0 to 5 units/ml. (C) Cell viability following exposure to β-1,3 glucanase from Tricoderma over a concentration range from 0 to 20 units/ml. (D) Biofilm cell viability following exposure to mannanase over a concentration range from 0 to 50 units/ml. (E) Viability following exposure to proteinase K over a concentration range from 0 to 2.5 units/ml.

FIG. 6.

Impact of fluconazole and β-1,3 glucanase alone and in combination as lock therapy against C. albicans biofilm cells in an in vivo catheter model. The SEM image perspective is from above the biofilm, imaging from the inside of the catheter lumen. The biofilms were grown for 24 h followed by instillation of the compound(s) into the lumen of the catheter for 24 h. Following compound exposure, the catheters were removed for SEM processing. For each of the panel groups (A, B, C, and D), the top row represents a ×50 magnification, and the bottom row represents a ×1,000 magnification. (A) Images are from a 48-h control biofilm with extensive biofilm formation. (B) Images are after a 1,000-μg/ml fluconazole exposure at 1,000 times the planktonic MIC. (C) Images are from catheters exposed to zymolyase at 1.25 units/ml alone. (D) Images are from a combination of zymolyase and fluconazole at 1.25 units/ml and 1000 μg/ml, respectively.

FIG. 7.

Impact of biofilm matrix and β-1,3 glucan (laminarin) on susceptibility of planktonic C. albicans K1 to fluconazole. Each vertical bar represents the log₁₀ CFU/ml following 48 h of incubation in the presence of twofold-escalating concentrations of fluconazole (range, 0.13 to 128 μg/ml) using the CLSI M27A methodology. Each value represents the mean from experiments performed on two separate days.