Triple-acting Lytic Enzyme Treatment of Drug-Resistant and Intracellular Staphylococcus aureus

Abstract

Multi-drug resistant bacteria are a persistent problem in modern health care, food safety and animal health. There is a need for new antimicrobials to replace over used conventional antibiotics. Here we describe engineered triple-acting staphylolytic peptidoglycan hydrolases wherein three unique antimicrobial activities from two parental proteins are combined into a single fusion protein. This effectively reduces the incidence of resistant strain development. The fusion protein reduced colonization by Staphylococcus aureus in a rat nasal colonization model, surpassing the efficacy of either parental protein. Modification of a triple-acting lytic construct with a protein transduction domain significantly enhanced both biofilm eradication and the ability to kill intracellular S. aureus as demonstrated in cultured mammary epithelial cells and in a mouse model of staphylococcal mastitis. Interestingly, the protein transduction domain was not necessary for reducing the intracellular pathogens in cultured osteoblasts or in two mouse models of osteomyelitis, highlighting the vagaries of exactly how protein transduction domains facilitate protein uptake. Bacterial cell wall degrading enzyme antimicrobials can be engineered to enhance their value as potent therapeutics.

Figure 1. PGH construct schematics, resistance development and eradicating S. aureus in a rat colonization model.

(A) Schematic of PGH constructs. Domains: CHAP endopeptidase (CHAP, red box); N-acetylmuramoyl-L-alanine amidase (AMID, green box); M23 endopeptidase (PEP, blue oval); CBDs (LysK SH3b, gold diamond; lysostaphin SH3b, gold diamond with dot); protein transduction domain (PTD, blue circle); hexahistidine purification tag (His₆). Domains not to scale. Specific PG cut sites are illustrated in Supplementary Fig. 2B. B. In vitro antimicrobial resistance development. Engineered triple fusions K-L and L-K suppress antimicrobial resistance development compared to LysK (K), lysostaphin (L), or a combination of equimolar concentrations of both (L+K). Changes in MIC are depicted as a fold-change at the tenth round of sublethal exposure compared to the first exposure, with the average fold-change of 4 replicates in red. Error bars = SEM. First exposure MICs: Lysostaphin, 0.77 μg/ml (27 nM); LysK, 47 μg/ml (840 nM); Lysostaphin and LysK (L+K) in combination 0.2 μg/ml (7 nM and 3 nM respectively) ;triple fusion K-L, 7 μg/ml (97 nM); triple fusion L-K, 7.8 μg/ml (107 nM). C. Colonization reduction in a rat nasal carriage model. Rats were inoculated with S. aureus strain ALR on day 1. After 5 days, the rats were treated twice daily for 3 days with 20 μl of a 10 mg/ml solution of each enzyme. The rat noses were excised on day 10, homogenized, and quantitative cultures were performed. Each point represents the CFU recovered from an individual rat. Bars indicate the median CFU/nose recovered from treated rats. Triple fusion L-K showed a significant reduction in colonization (98%) of treated rats compared with rats treated with buffer alone. Data were compiled from five independent experiments. Lyso = commercially purchased lysostaphin (AMBI, Tarrytown, NY). Statistical comparisons were made with the Mann-Whitney test.

Figure 2. PGH-PTD eradication of intracellular S. aureus in cultured cells.

Cultured cells were infected, treated with gentamicin to kill extracellular S. aureus, and then treated with the PGHs indicated. All results are standardized to the gentamicin (GENT) only control. Nomenclature of the PGH constructs is as in Fig. 1. PTDs are listed in Table 1. Asterisks indicate statistical significance detected with single factor ANOVA (α = 0.05) with paired t-test posthoc analyses α = 0.05 adjusted with Šidák correction for multiple comparisons (for all cases p < 0.01). (A) Bovine mammary epithelial cell line (MAC-T) infected with strain Newbould 305 (N = 8). Error bars represent SEM. 1-Sigma is commercial lysostaphin (Sigma). (B) Human brain microvasculature epithelial cells (hBMEC) infected with S. aureus strain ISP479C (N = 3). (C) Murine primary osteoblasts (mOB) infected with S. aureus strain UAMS-1 (N = 3). (D) Single plane of confocal microscopy z-stack overlaid on bright field exposure of live cultured MAC-T cells exposed to both S. aureus and PGH 3-PTD1. Live S. aureus (~0.6–1.0 μM diameter) are labeled with green fluorescent wheat germ agglutinin (WGA); PGH K-L-PTD1 is labeled with Alexa Fluor (Red). Yellow staining in the combined panel represents intracellular co-localization (in a single plane as determined by z-stack analysis) of both S. aureus and K-L-PTD1 (blue arrow). (E) Confocal microscopy maximum intensity projections (with all z-planes represented) of a MAC-T cell exposed to S. aureus and K-L-PTD1 as in (D) The majority of the S. aureus are localized in the thickest part of the cytoplasm surrounding the zone of exclusion created by the nucleus. Many, but not all, S. aureus are co-localized with K-L-PTD1 (yellow) in the combined panel.

Figure 3. Triple-acting fusions can eradicate intracellular S. aureus in murine osteoblasts.

(A) Intracellular eradication of GFP expressing S. aureus. Neonatal mouse whole calvaria were treated for 4 hours with chimeric PGHs or buffer alone, 24 h post inoculation with GFP expressing S. aureus. Calvaria were embedded in freeze media, sectioned, and subjected to fluorescence imaging. The image shown is a representative figure for triplicate sections of three separate calvaria. (B) Fluorescence intensity from these sections was measured in ImageJ and defined as arbitrary fluorescence standardized to the untreated control. Error bars represent SEM of four replicate experiments. (C) Ex-vivo intracellular S. aureus eradication. Infected calvaria were homogenized and CFU were counted post treatment. Error bars represent SEM. (D) Murine model of staphylococcal osteomyelitis. C57BL/6J mice were anesthetized and their femurs were surgically exposed. A trough was drilled through the bone cortex, and the damaged bone sites were inoculated with 1 × 10³ CFU S. aureus in agarose beads. After 24 hours, mice were treated (i.m. to site of infection) twice in a 24 h period with PBS or 5 mg/kg of triple fusion K-L or K-L-PTD1. The femurs were removed, homogenized, and plated to quantify the bacterial load. Bars indicate the average CFU recovered (N = 6). Both triple fusion K-L (p = 0.012) and K-L-PTD1 (p = 0.021) significantly reduced bacterial load as compared to no treatment, but the presence or absence of PTD1 had no significant effect (p = 0.73). Asterisks represent statistical significance as determined by one-way ANOVA followed by Tukey’s posthoc test.

Figure 4. PTD1 enhances triple fusion K-L eradication of dynamic MRSA biofilms.

(A) Confocal microscopy of biofilm with Live/Dead staining. Single 1 μm z-stack images in the middle of NRS382 (USA100) biofilms treated with 100 μg/ml (1.4 μM) triple fusion K-L, K-L-PTD1, (69 μM) vancomycin (Van) at a flow rate of 0.5 ml/min for 0, 60, or 120 minutes. Biofilms were pre-stained with the Live/Dead stain (see methods) and viewed with 20X magnification. (B) Bacterial viability. Analysis of bacterial viability in NRS382 dynamic biofilms based on mean fluorescent intensities of the Live/Dead viability stain when exposed to 100 μg/ml of PGH K-L (1.4 μM), K-L-PTD1 (1.4 μM), or vancomycin (69 μM) at a flow rate of 0.5 ml/min for 2 h and compared to PBS. Error bars represent SEM (N = 3) of three independent 100 × 100 pixel squares identically located in each biofilm z-stack. All values were found to be significantly different from PBS and each other using a two-tailed, unpaired, t-test; K-L vs. KL-PTD1 (p = 0.00064), K-L vs. vancomycin (p = 0.00036), and K-L-PTD1 vs. vancomycin (p = 0.000023), as indicated with an asterisk.

Figure 5. Triple-acting fusions can reduce S. aureus induced mastitis in a mouse model.

(A) Specific activity of PGH fusion constructs in vitro. Turbidity reduction assay with S. aureus Newbould 305 resuspended in lysis buffer to an OD_600nm of 1.0 and treated with 1 μM final concentrations of each PGH. The maximum specific activity for three experiments is represented as an average with SEM error bars. Single asterisks indicate significant difference from buffer control (p < 0.05); double asterisks indicate significant difference from parental enzyme (p < 0.05). (B) Bacterial burden in mice with mastitis treated with recombinant PGHs. CFUs from infected murine mammary glands following a single treatment with 50 μl of 25 μM PGH (1.25 nmol). Each data point represents the average of duplicate bacterial platings. The horizontal bars represent the average CFU/mg for each group. (C) TNFα response in mice with mastitis that were treated with PGHs. Data represent a minimum of 7 measurements in duplicate. Log TNFα was used for statistical analysis. Single asterisk indicates a significant difference from buffer control (p < 0.05); double asterisk indicates significant difference from parental enzyme (p < 0.05).