Evaluating the Effects of Surotomycin Treatment on Clostridium difficile Toxin A and B Production, Immune Response, and Morphological Changes

Abstract

Surotomycin is a cyclic lipopeptide in development for Clostridium difficile-associated diarrhea. This study aimed to assess the impact of surotomycin exposure on C. difficile toxin A and B concentrations and the associated changes in immune response in comparison to vancomycin and metronidazole. Time-kill curve assays were performed using strain R20291 (BI/NAP1/027) at supra-MICs (4× and 40×) and sub-MICs (0.5×) of surotomycin and comparators. Following treatment, CFU counts, toxin A and B concentrations, and cellular morphological changes using scanning electron microscopy were examined. Inflammatory response was determined by measuring interleukin-8 (IL-8) concentrations from polarized Caco-2 cells exposed to antibiotic-treated C. difficile growth media. Supra-MICs (4× and 40×) of surotomycin resulted in a reduction of vegetative cells over 72 h (4-log difference, P < 0.01) compared to controls. These results correlated with decreases of 77% and 68% in toxin A and B production at 48 h, respectively (P < 0.005, each), which resulted in a significant reduction in IL-8 concentration compared to controls. Similar results were observed with comparator antibiotics. Bacterial cell morphology showed that the cell wall was broken apart by surotomycin treatment at supra-MICs while sub-MIC studies showed a “deflated” phenotype plus a rippling effect. These results suggest that surotomycin has potent killing effects on C. difficile that results in reduced toxin production and attenuates the immune response similar to comparator antibiotics. The morphological data also confirm observations that surotomycin is a membrane-active antibiotic.

INTRODUCTION

Clostridium difficile is a Gram-positive, spore-forming anaerobic bacterium that is considered an urgent threat by the U.S. Centers for Disease Control and Prevention (CDC) (1). C. difficile infection (CDI) accounts for one-quarter of a million infections per year and 14,000 deaths. Since the emergence of the epidemic strain (ribotype BI/NAP1/027), deaths attributed to CDI have risen 400% (2). Upon infection by the bacterium, C. difficile secretes active toxins (A and B) that cause a systemic immune response coordinated by several immune circulating factors, including the chemokine interleukin-8 (IL-8) (3, 4).

There are currently three antibiotics used clinically to treat CDI: vancomycin, fidaxomicin, and metronidazole. The increased incidence and severity of CDI have prompted the development of new compounds with activity against C. difficile. Surotomycin (formerly CB-183,315) is a novel narrow-spectrum, minimally absorbed antibiotic that is currently being evaluated for the treatment of CDI (5). Surotomycin affects the cell wall integrity of C. difficile by dissipating the membrane potential, thereby killing the bacteria (6). Other studies have demonstrated killing activity against C. difficile; however, the effects on cell morphology, toxin A and B secretion, and associated changes in the immune response have not been examined. Therefore, the purpose of this study was to assess these pharmacologic activities of surotomycin compared to metronidazole and vancomycin.

MATERIALS AND METHODS

Antibiotics used and cell growth conditions.

Surotomycin was provided by Merck and Co. Metronidazole and vancomycin were purchased from a commercial supplier (Sigma-Aldrich). C. difficile strain R20291 (BI/027/NAP1) was used for all the experiments. The strain was grown in brain heart infusion (BHI) medium (Hardy Diagnostics), supplemented with 0.1% sodium taurocholate (Alfa Aesar) in a vinyl anaerobic chamber (Coy Lab Products) at 37°C. Bacteria were grown on blood agar plates (Hardy Diagnostics) to determine colony recovery and counts.

Time-kill curves.

MICs were determined for surotomycin, metronidazole, and vancomycin by broth microdilution. The medium was adjusted to 50 mg/liter of Ca²⁺ for testing surotomycin. Determination of supra-MICs (4× and 40×) and sub-MIC (0.5×) was performed for each of the antibiotics. For these studies, cultures were prepared by inoculating one colony into BHI medium. Following a 24-h incubation, cultures were diluted 1:100 (approximately 10⁶ CFU/ml) in fresh BHI supplemented with 0.1% sodium taurocholate and the designated concentration of antibiotic at time zero (T0). CFU counts were determined at T0, 24 h (T24), T48, and T72.

At each collection time point, 100 μl of killed yeast solution was mixed with one ml of culture and centrifuged for 1 min at 10,000 rpm to ensure coaggregation of Candida albicans with C. difficile to maximize the ability of low concentrations of C. difficile to be trapped into the pellet (7). For the yeast solution preparation, C. albicans suspension was centrifuged 1 min at 10,000 rpm. The supernatant was removed; the pellet was resuspended in 70% isopropanol and incubated at room temperature for 10 min. The solution was centrifuged 1 min at 10,000 rpm, the supernatant removed, and the pellet resuspended in phosphate-buffered saline (PBS). The sample supernatant was removed, and the pellet was resuspended in fresh BHI. One hundred microliters of total sample serial dilutions was spread on blood agar plates in duplicate and incubated at 37°C in anaerobic chamber for 48 h prior to colony counting.

Spore viability was determined by plating 100 μl of sample serial dilutions on blood agar plates in duplicate and counting colonies at T0 and T24. From the total samples, vegetative cells were killed using 100 μl of 70% isopropanol. The pure spore samples were then diluted and spread on blood agar plates, and colonies were counted thereafter. The limit of detection (LOD) for these assays was 500 CFU/ml. All experiments were performed at least in duplicate.

Electron microscopy.

At each collection time point for the time-kill studies, 5 ml of samples was centrifuged for 5 min at 5,000 rpm. The supernatant was removed, and the pellets were then resuspended in 200 μl of 4% paraformaldehyde (PFA) and incubated for 1 h. The samples were again centrifuged for 1 min at 10,000 rpm, resuspended in H₂O, and subsequently incubated on poly-l-lysine-coated coverslips for 30 min. Dried coverslips covered with the cells were coated with 20 nm of gold using a Denton Desk II Sputtering System (Denton Vacuum). Following coating, coverslips were transferred to a scanning electron microscope (FEI XL-30FEG). Images were taken at 15.0 kV at the designated magnifications.

Determination of toxin A and B production.

An enzyme-linked immunosorbent assay (ELISA) for C. difficile toxin A or B (tgcBIOMICS) was used according to the manufacturer's instructions to measure toxin levels. The limit of detection was determined to be 0.31 ng/ml for the toxins.

Measuring IL-8 concentrations.

Caco2 cells (ATCC) were grown in Eagle's minimum essential medium (EMEM; ATCC) with antibiotics and supplements according to the ATCC guidelines. Experiments were carried out in a 24-well plate between cell passage numbers 5 to 15 in order to obtain a two-dimensionally (2D) polarized epithelium before adding toxin. Complete medium supplemented with 20% fetal bovine serum (FBS) was changed every other day until cell polarization (8). For the interleulkin-8 (IL-8) measurements, 300 μl of filtered supernatants (collected from the time-kill experiments) was added to the culture. Following 24 h of incubation, the medium from the Caco-2 cells was collected and IL-8 release was quantified by an IL-8 ELISA kit (Thermo Fisher Scientific). The limit of detection for IL-8 levels was 5 pg/ml.

Statistics.

Statistical analyses were performed using SPSS statistics software (IBM). Data are expressed as the means ± standard errors of the means (SEM). Data were assessed for significance using a Student t test or one-way analysis of variance (ANOVA) as appropriate. A P value of <0.05 was considered significant.

RESULTS

Surotomycin killing effects.

Bactericidal activity of surotomycin was determined by measuring supra-MIC killing kinetics. For C. difficile strain R20291, the MIC for surotomycin was 0.03 μg/ml, that for metronidazole was 0.5 μg/ml, and that for vancomycin was 1 μg/ml. Under control conditions without antibiotic, C. difficile cells grew exponentially, reaching a plateau after 48 h of incubation. In contrast, 4× and 40× MIC treatment with surotomycin significantly reduced total vegetative cell number by 4 log₁₀ (P < 0.01) or greater by 72 h, demonstrating potent and rapid bactericidal activity (Fig. 1A). C. difficile cells were imaged following 24 h of treatment. Consistent with exponential cell growth, control cells appeared to be intact and numerous and were dividing (Fig. 1B). Following surotomycin treatment, cells were sparse and the remaining cell's outer wall appeared rough and punctured (Fig. 1B). In addition to investigating the effects on vegetative cells, spore viability was evaluated following 24 h of surotomycin treatment. The proportion of spores increased following surotomycin treatment (Fig. 1C); however, the absolute number of spores remained constant, indicating that surotomycin does not induce sporulation. Collectively, these data provide evidence that surotomycin has potent in vitro effects on exponentially growing C. difficile cells.

FIG 1.

(A) Surotomycin time-kill experiments against C. difficile strain R20291 at 4× and 40× MIC (mean values of two experiments ± standard deviations [SD]) compared to control, nontreated cells (CTR). The limit of detection (LOD) is denoted by the dashed line. (B) Following 24 h of treatment/nontreatment, cells were collected and imaged on a scanning electron microscope. Magnifications for the control image (top) and surotomycin-treated image (bottom) are ×25,000 and ×50,000, respectively. (C) Ratio of spores to vegetative cells for each of the groups at 0 and 24 h of treatment.

Changes in toxin production and immune response following supra-MIC antibiotic treatment.

To determine whether surotomycin treatment affected toxin concentrations, toxin A and B concentrations were measured and compared with those of metronidazole and vancomycin at supra-MICs. In all treatment groups, there was a significant time-dependent decrease in toxin production. Following 4× and 40× MIC surotomycin treatments, there were >77% and >63% decreases in toxin A and B production at 48 h, respectively (P < 0.005, each) (Fig. 2A). Following 4× and 40× MIC vancomycin treatments, there were >35% and >56% decreases in toxin A and B production at 48 h (P < 0.01, each) (Fig. 2B). Following 4× and 40× MIC metronidazole treatments, there were >71% and >56% decreases in toxin A and B production at 48 h (P < 0.05) (Fig. 2C).

FIG 2.

Toxin A and B levels are graphed following supra-MIC treatment with surotomycin (A), vancomycin (B), and metronidazole (C) at 24 h and 48 h of exposure. Error bars represent standard errors.

To test the effects of surotomycin and comparators on the immune response, IL-8 concentrations were measured in cultured polarized colonic epithelial cells following treatment with the supernatant from the C. difficile culture studies as described previously (9). Consistent with toxin levels, treatment with surotomycin, vancomycin, and metronidazole reduced IL-8 release by >55% (P < 0.05) in all cases, demonstrating a reduction in the IL-8 response with exposure to antibiotics. No differences were observed between the groups. Cumulatively, these data suggest that the in vitro killing effects of surotomycin, vancomycin, and metronidazole can sufficiently reduce toxin concentrations and immune response activation when used in the exponential growth stage.

Comparing the effects of surotomycin, vancomycin, and metronidazole at sub-MIC.

While the supra-MIC killing experiments demonstrate the efficacy of surotomycin, they do not provide insight into the antibiotic's direct pharmacologic effect independent of its in vitro killing properties. Therefore, sub-MIC experiments were performed in order to determine potential independent pharmacologic properties. Following treatment with 0.5× MIC of surotomycin, vancomycin, and metronidazole, cell growth was marginally reduced (Fig. 3A). While cell growth was slightly lowered, there were no differences in toxin production (Fig. 3B) or Il-8 release (Fig. 3C) following antibiotic treatment with surotomycin or comparator antibiotics. Cell morphology was affected in antibiotic-treated cells compared to controls (Fig. 3D). Surotomycin-treated cells appeared “deflated” and exhibited a rippling effect, likely consistent with the antibiotic's proposed mechanism of action on the cell membrane. Vancomycin also affected the cell wall, whereas metronidazole-treated cells appeared smaller, indicating an effect on cell growth. These results indicate that lower doses of these antibiotics may not be effective at killing C. difficile, although some of the cells are morphologically affected.

FIG 3.

Following 24-h exposure of 0.5× MIC of either surotomycin, vancomycin, or metronidazole, CFU were determined (A), levels of toxins A and B were measured (B), IL-8 concentrations from Caco-2 exposed cells were assessed (C), and the effects on cell morphology were documented (D). Error bars represent standard errors.

DISCUSSION

The goal of the current study was to evaluate the efficacy of surotomycin on toxin A and B concentrations, the immune response measured using IL-8 concentrations, and the morphological changes associated with treatment using SEM. In this study, surotomycin exhibited potent and rapid bactericidal activity against exponentially growing cells without effect on spores, confirming previous observations (6, 10–12).

Novel findings from this study include reduced toxin A and B concentrations in exponentially growing C. difficile cells exposed to surotomycin and comparators, which reduced the immune response as evidenced by decreased concentrations of IL-8. These effects are likely to be a direct effect of the in vitro killing effects of the antibiotics, as similar effects were not observed with C. difficile cells exposed to sub-MICs of the antibiotics. Finally, morphological changes in C. difficile cells exposed at supra- and sub-MICs of surotomycin provide further evidence for the cell wall-active properties of this antibiotic.

Bouillaut et al. recently investigated toxin A concentrations of C. difficile after exposure to supra-MICs of surotomycin (13). Using C. difficile cells in the early stationary phase, these investigators demonstrated minimal change in CFU counts after exposure to surotomycin at 8× MIC. Similarly, no associated decrease in toxin A concentrations were observed after surotomycin exposure. The observations are in concordance with our sub-MIC observations and help confirm that the decreased toxin production observed in our study was due to the killing effect of the antibiotics and not an independent pharmacologic property. These observations are also concordant with Chilton et al., who demonstrated reduced C. difficile cytotoxicity in a human gut model after exposure to surotomycin (10). Taken together, it can be concluded that surotomycin can reduce toxin A and B concentration by its direct killing effects and not by an independent property of the antibiotic. In our study, this killing effect and associated decreased toxin production led to a reduced immune activation measured using IL-8.

Alam et al. recently investigated the mode of action of surotomycin against growing and nongrowing C. difficile (6). In that study, surotomycin inhibited macromolecular synthesis of DNA, RNA, and protein simultaneously and dissipated the membrane potential of the bacterial cell wall. Our observations agree with and extend these findings by providing evidence of morphological changes to the cell wall that are consistent with these previous findings. Phenotypically, the effects of surotomycin on C. difficile were distinct from those of the other antibiotics, demonstrating a novel mechanism of action and providing insight into its killing action. Taken together, it can be concluded that surotomycin has potent in vitro killing effects of C. difficile with a unique mechanism of action, different from that obtained with other currently available antibiotics used for the treatment of patients with CDI.

This study has certain limitations. We used the current epidemic strain 027 for all experiments. These results will need to be confirmed using other strains. The clinical significance of these beneficial pharmacologic properties will need to be confirmed, especially the host IL-8 response, for which clinical data are lacking. We did not compare surotomycin to fidaxomicin, which will be required in future studies. Finally, we did not simultaneously assess the changes in microbiota due to surotomycin and comparator antibiotics. This beneficial property also requires additional studies in other models, such as mixed cultures and/or animals.

In conclusion, this study demonstrated that surotomycin has potent in vitro killing effects against exponentially growing C. difficile cells, which was associated with reduced toxin A and B production and decreased IL-8 production. The toxin and immune response effects were not observed at sub-MICs, suggesting that these effects are a direct result of the killing effect of the antibiotic and not due to an independent pharmacologic property.