Discovery of a novel natural product inhibitor of Clostridioides difficile with potent activity in vitro and in vivo

Abstract

Clostridioides difficile infection is a global health threat and remains the primary cause of hospital-acquired infections worldwide. The burgeoning incidence and severity of infections coupled with high rates of recurrence have created an urgent need for novel therapeutics. Here, we report a novel natural product scaffold as a potential anticlostridial lead with antivirulence properties and potent activity both in vitro and in vivo. A whole cell phenotypic screening of 1,000 purified natural products identified 6 compounds with potent activity against C. difficile (minimum inhibitory concentration (MIC) range from 0.03 to 2 μg/ml). All these 6 compounds were non-toxic to human colorectal cells. The natural product compounds also inhibited the production of key toxins, TcdA and TcdB, the key virulence determinants of C. difficile infection pathology. Additionally, the compounds exhibited rapid bactericidal activity and were superior to the standard-of-care antibiotic vancomycin, in reducing a high inoculum of C. difficile in vitro. Furthermore, a murine model of C. difficile infection revealed that compound NP-003875 conferred 100% protection to the infected mice from clinical manifestations of the disease. Collectively, the current study lays the foundation for further investigation of the natural product NP-003875 as a potential therapeutic choice for C. difficile infection.

Fig 1. Data from the high-through screening (HTS) of the AnalytiCon MEGx library.

The natural products were screened against C. difficile at 3 μM. The natural product chemotypes that exhibited greater than or equal to 95% bacterial growth inhibition were deemed as hits. The HTS assay identified 9 hits.

Fig 2. Killing kinetics of hit natural products, vancomycin, and fidaxomicin (5X MIC) against C. difficile.

The standard deviation values for the triplicate samples of each natural product/control antibiotic have been represented by error bars.

Fig 3. C. difficile toxin inhibition by natural products and control antibiotics (vancomycin and fidaxomicin).

C. difficile was incubated with subinhibitory concentrations of drugs for 12 hours. Viable cells (log₁₀ CFU/ml, bars) in each sample were determined by plating and the amount of toxin in each supernatant (OD₄₅₀-OD₆₂₀, lines) was assessed using ELISA. The data represents the mean and standard deviation for triplicate samples of each treatment. Asterisk (*) indicates a statically significant difference in the toxin content of the supernatant between the fidaxomicin- or natural product-treated samples and the untreated control.

Fig 4. Cytotoxicity assay of natural products against Caco-2 cell line.

The percentage of viable Caco-2 cells was measured as the ratio of average absorbance of each sample relative to the negative control (DMSO). The absorbance values represent the mean of three samples of each natural product compound with error bars representing standard deviation values for the absorbance.

Fig 5. Murine model of CDI.

(A) Schematic of mice model. 6-week-old female mice were randomly divided into groups (n = 5 per group). Mice were given antibiotic water consisting of kanamycin (0.4 mg/ml), colistin (850 U/ml), gentamicin (0.035 mg/ml), vancomycin (0.045 mg/ml), and metronidazole (0.215 mg/ml) for 5 days. Clindamycin (10 mg/kg) was injected intraperitoneally a day before infection. C. difficile spores were administered via oral gavage on Day 0. Treatment was initiated on Day 0 and continued for 5 days. A single dose of each drug was administered via oral gavage daily. (B) The Kaplan-Meier survival curve was analyzed using a log-rank (Mantel-Cox) test. Significant difference was noted between the survival curves of the mice treated with NP-003875 vs. untreated control (100% vs. 20%; p = 0.0144).