Bacterial Strains

For the in vitro assessment of the antimicrobial activity of (S)-JBC 1847 a diverse collection of 19 strains of S. aureus were chosen (Table S1). Of these, 16 strains were obtained from our in-house strain collection of clinical isolates. Fourteen of the in-house strains were resistant to at least one antimicrobial class, seven were categorized as MRSA strains whereas two strains exhibited no detectable antimicrobial resistance (Table S1). To compare the activity of JBC 1847 (racemic) with (S)-JBC 1847, three well-characterized reference strains of S. aureus were chosen for Minimal Inhibition Concentration (MIC) determination. The strains represent a livestock associated MRSA (CC398), a community associated MRSA (USA300) (GenBank: CP020619.1) and the MRSA strain ATCC BAA-1556 (GenBank: CP000255). A vancomycin resistant Enterococcus (VRE) strain (E. faecium ATCC 700-221) was also included in the assessment. In general, all 19 S. aureus strains and the single E. faecium strain, were handled as previously described by Ronco et al. [7].

Purification of (S)-JBC 1847

A racemic mixture of JBC 1847 was synthesized from promazine at The Department of Chemistry, Faculty of Science, University of Copenhagen as previously described [7]. Subsequently, (S)-JBC 1847 was purified from this racemic mixture of JBC 1847 from (S)-Citronellol tosylate (Sigma aldrich, Roskilde, Denmark), prepared according to the protocol published by Shirai et al. 1999 [11].

In Vitro Analyses of Antimicrobial Activity

Determination of MIC values was performed based on the CLSI guidelines references for the broth dilution method [12]. All MIC assays were run in duplicates and performed as previously described [7].

For determination of MBC, the broth dilution method was performed slightly modified but based on the procedure described by Rodríguez-Melcón et al. [13]. The MBC values were determined for the three MRSA reference strains (CC398, USA300 and ATCC BAA-1556) and the single VRE strain (E. faecium ATCC 700-221). Briefly, for MBC determinationa total of 100 μL suspension from each well in the MIC assay was seeded with loop on agar plates (agar base, Sigma, Copenhagen, Denmark) supplemented with 5% bovine blood. The plates were incubated at 37 °C for 48 h. MBC was defined as the lowest concentration of JBC 1847 that reduced the CFU of the original inoculum (1.5 × 105 CFU) by ≥ 99.9%.

Generation of Mutants with Increased Phenotypically Tolerance to (S)-JBC 1847

The frequency of spontaneous single-step mutations in S. aureus and E. faecium, respectively, mediating increased tolerance to (S)-JBC 1847, was determined using the gradient plate technique. The plates were prepared in Petri dishes, which were poured with two layers of agar. The bottom layer consisted of Mueller–Hinton agar, allowed to harden with the plate slanted sufficiently to cover the entire bottom. The top layer, added to the dish in the normal position, were supplied with (S)-JBC 1847 to generate concentrations of ∼0.5–8 × MIC through the slope of the plate. An inoculum of 109 CFU of S. aureus USA 300 and E. faecium ATCC 700–221, respectively, was homogeneously spread on each plate, and incubated for 48 h at 37 °C. Colonies growing at the highest antibiotic concentration were sampled, checked for purity, grown overnight in antibiotic-free broth, and determined for MIC as described above. All experiments were carried out in triplicates.

Tolerance of (S-) JBC 1847 in a Larvae Model

High concentration tolerance (32 mg/kg) was initially assessed in a Galleria mellonella model. The study was performed in accordance to a method previously published by Desbois and Coote, 2011 [14], with minor modification and in technical and biological duplicates.

The larvae were purchased from Monis Pet Store, Viborg, Denmark and upon arrival stored at 5 °C and used within 3 days. The larvae did not receive either food nor water at any timepoint after arrival and were only exposed to light when handled and kept in petri dishes. The dishes were cleaned for fecal discharge and dead larvae were removed every day during the trials. At the end of all trials the larvae were euthanized by freezing to − 20 °C for at least 24 h.

Before injection, the larvae were left to acclimatize for 30 min at room temperature. The larvae were inspected for signs of melanization and larvae with excess darkening of the cuticle were excluded. The larvae weighing between 300 and 400 mg were included in the study only.

The larvae were randomly distributed into four different treatment groups (unexposed control group, vehicle injected, (S)- JBC 1847 (32 mg/kg) and vancomycin injected (32 mg/kg)), each group including 20 larvae.

The larvae were injected intraperitoneally using ‘Insumed’ syringes (Fisher scientific, Roskilde, Denmark) with a 31-gauge needle. The first injection was made by penetrating last proleg on the left side of the larva and the needle was subsequently pushed in cranial direction just beneath the cuticle until the needle-tip reached the adjacent proleg on same side. Subsequent injections were performed the same way only the site of injection was moved one proleg in cranial direction. Following the first injection larvae were placed in petri dishes in an incubator at 37 °C. The larvae received two injections with an interval of 24 h and observations for viability in all groups were observed for 72 h and scored for mortality. Larvae were considered dead and discharged when completely melanized and showed no movement when poked with a needle cap.

A statistical analysis of variance (ANOVA) was used to establish if the groups differed in total mortality.

Tolerance (S)-JBC 1847 in a Murine Model

Following the compound tolerance evaluation in G. mellonella, tolerance to (S-) JBC 1847 and JBC 1847 racemic, respectively, was assessed in a vertebrate (murine) model. The maximum tolerable dosis (MTD) was determined for (S-) JBC 1847 and JBC 1847, respectively by injecting NMRI female mice intraperitoneal (IP) (n = 2) with escalating doses from 0.5 to 25 mg/kg of the compound. The mice were scored for clinical signs of discomfort for 4 h. Mice with no clinical scores within 2 h, were treated with a similar IP dose once again. Mice were scored additionally following 2 h after the last dose. When moderate discomfort was observed the mice were euthanized and the dose was considered “not tolerated”. The MTD for (S)-JBC 1847 and JBC 1847, respectively, were determined as the dose proceeding the “not tolerable” dose.

Treatment Efficacy in a Murine Peritonitis Model

The treatment efficacy of (S)-JBC 1847 was assessed in a murine peritonitis infection model as previously described by Lundberg et al., 2010 [15], with minor modifications. XX Female NMRI mice (18–22 g) were infected with MRSA 43484 and treated with a dosage of 20 mg/kg (S)-JBC 1847 and treatment efficacy was investigated in two separate assays ((S)-JBC 1847 dissolved in 5% DMSO or β–cycklodextrin). The treatment control group was treated with vancomycin (80 mg/kg) subcutaneously.