Anti-Biofilm Activity of Carnosic Acid from Salvia rosmarinus against Methicillin-Resistant Staphylococcus aureus

Abstract

The Salvia rosmarinus "Eretto Liguria" ecotype was studied as a source of valuable bioactive compounds. LC-MS analysis of the methanolic extract underlined the presence of diterpenoids, triterpenoids, polyphenolic acids, and flavonoids. The anti-virulence activity of carnosic acid along with the other most abundant compounds against methicillin-resistant Staphylococcus aureus (MRSA) was evaluated. Only carnosic acid induced a significant reduction in the expression of agrA and rnaIII genes, which encode the key components of quorum sensing (QS), an intracellular signaling mechanism controlling the virulence of MRSA. At a concentration of 0.05 mg/mL, carnosic acid inhibited biofilm formation by MRSA and the expression of genes involved in toxin production and made MRSA more susceptible to intracellular killing, with no toxic effects on eukaryotic cells. Carnosic acid did not affect biofilm formation by Pseudomonas aeruginosa, a human pathogen that often coexists with MRSA in complex infections. The selected ecotype showed a carnosic acid content of 94.3 ± 4.3 mg/g. In silico analysis highlighted that carnosic acid potentially interacts with the S. aureus AgrA response regulator. Our findings suggest that carnosic acid could be an anti-virulence agent against MRSA infections endowed with a species-specific activity useful in multi-microbial infections.

Keywords: MRSA; Salvia rosmarinus; anti-virulence; biofilm; carnosic acid; quorum sensing.

Figure 1

Chromatogram of compounds in the methanolic extract of the rosemary “Eretto Liguria” ecotype. Peak data are shown in Table 1. The peaks of the identified metabolites are annotated with numbers. Key: 1: caffeic acid; 2: isorhamnetin–3–O-hexoside; 3: apigenin–7–O–glucoside; 4: homoplantaginin; 5: dihydrorabdosiin; 6: rosmarinic acid; 7: rabdosiin; 8: isorhamnetin; 9: rosmanol isomer; 10: rosmanol isomer/epirosmanol; 11: rosmanol isomer; 12: rosmarinic acid isomer; 13: rosmanol isomer; 14: hydroxy–octadecatrienoic acid; 15: carnosol; 16: rosmadial; 17: rosmadial isomer; 18: hydroxy–octadecadienoic acid; 19: carnosic acid; 20: methyl carnosate; 21: oleanolic acid.

Figure 2

Interference in quorum sensing signaling of MRSA. (A) Heatmap illustrating differential gene expression over non-treated (nt) controls. MRSA cultures (10⁶ CFU/mL) were incubated at 37 °C for 16 h with the isolated compounds (0.5 mg/mL) reported in Table 1 or vehicle (DMSO 0.05% v/v). RNA was extracted and subjected to qRT-PCR to evaluate agrA and rnaIII gene expression. Expression values for each gene were normalized to the expression of the housekeeping gene gyrB and reported as fold change over non-treated (nt) samples. Increased gene expression is reported in red, and reduced gene expression is reported in green (color scale at bottom). Data were obtained from three independent experiments, each performed in duplicate. See Table 1 for sample identification. (B) MRSA cultures (10⁶ CFU/mL) were incubated at 37 °C for 16 h with carnosic acid (CA) at 0.5, 0.05, and 0.005 mg/mL or 0.25 µg/mL clindamycin (CL). Expression of agrA and rnaIII genes was assessed by qRT-PCR and normalized to the expression of the housekeeping gene gyrB. Data are reported as mean ± st err of three independent experiments, each performed in triplicate, and calculated as fold change relative to gene expression in non-treated (nt) samples. (C) MRSA cultures (10⁶ CFU/mL) were incubated at 37 °C for 16 h with CA at 0.05 mg/mL or 0.25 µg/mL clindamycin (CL). Expression of hla and psmα genes was assessed by qRT-PCR and normalized to the expression of the housekeeping gene gyrB. Data are reported as mean ± st. err. of three independent experiments, each performed in triplicate, and calculated as fold change relative to gene expression in non-treated (nt) samples. * denotes p < 0.05 vs. nt; ** denotes p < 0.02 vs. nt.

Figure 3

Carnosic acid (CA) exerts no effect on prokaryotic and eukaryotic cells. (A) MRSA (10⁶ CFU/mL) cultures were incubated at 37 °C with DMSO (0.05% v/v), CA (0.05 mg/mL), or were left untreated (nt). The growth kinetics were recorded for 16 h at 620 nm. (B) At the end of the incubation reported in (A), bacterial cultures were diluted and plated on agar media. Plates were incubated for 16 h at 37 °C and colonies were counted. (C,D) THP-1 and A549 cell lines were incubated with DMSO (0.05% v/v), CA (0.05 mg/mL), or were left untreated and cell viability was assessed by MTT assay at 0, 12, 24, 36, and 48 h. Data are reported as mean ± st err of three independent experiments, each performed in triplicate.

Figure 4

Effect of carnosic acid on MRSA biofilm. (A) To assess biofilm formation by MRSA, bacterial cultures were incubated under static conditions and biofilm formation was assessed at different time points by crystal violet staining. (B) MRSA cultures were treated with CA (0.05 mg/mL) and incubated at 37 °C under static conditions. Biofilm formation was evaluated 16 h later by crystal violet staining. (C) MRSA cultures were incubated for 8 h under static conditions and then treated with CA at 0.05 mg/mL. Cultures were incubated for 8 h more and biofilm formation was evaluated by crystal violet staining following 16 h of total incubation. (D) The experiments described in B were also performed with non-antibiotic-resistant S. aureus. Data are reported as mean ± st err of three independent experiments, each one performed in triplicate. * denotes p < 0.05 vs. DMSO.

Figure 5

Effects of carnosic acid (CA) on MRSA intracellular killing. (A) MRSA cultures (10⁶ CFU/mL) were incubated for 16 h with CA (0.05 mg/mL), DMSO (0.05% v/v), or were left untreated (nt). Bacteria were then cultured with differentiated THP-1 (MOI 1:1) cells to assess phagocytic activity. Viable bacteria were enumerated by seeding the samples on agar plates. (B) The intracellular killing assay was also performed with non-antibiotic-resistant S. aureus. The experimental protocol was the same as described above. Data are reported as mean ± st err of three independent experiments, each performed in triplicate. * denotes p < 0.05 vs. nt. (C) To exclude any possible effects of CA in reducing the phagocytosis of THP-1 cells, fluorescent latex beads with 0.5 μm mean size were incubated (CA treatment) or not (no treatment) with CA, as described above. The phagocytic activity was determined by cytofluorimetric analysis on 10,000 collected events. FL-1, FITC fluorescent channel. The image is representative of 3 independent experiments.

Figure 6

Binding pose (A) and interactions (B) of carnosic acid (CA) at the conserved AgrA active site. (A) The protein is reported as green-purple ribbons; the ligand is reported as purple capped sticks; H-bonds are presented as yellow dotted lines. (B) The ligand is surrounded by the protein residues represented as follows: the negatively charged residues are indicated in red, polar residues are in cyan, and hydrophobic residues are shown in green. H-bonds are depicted as purple arrows.

Figure 7

Molecular dynamics simulation results. (A) Root-mean-square deviation (RMSD) plot for the carnosic acid (CA)/AgrA complex along 100 ns molecular dynamics simulation related to Cα positions of residues belonging to the protein backbone (blue) and the ligand (purple). (B) Protein–ligand interactions (or ‘contacts’) plot for the CA/AgrA complex along 100 ns molecular dynamics simulation. Contacts are categorized into four types: hydrogen bonds, hydrophobic interactions, ionic bonds, and water bridges. (C) Ligand–atom interactions with the protein residues on chain A (Lys223 and Lys236) and on chain B (Arg233 and Asn234). Interactions that occurred more than 30.0% of the simulation time in the selected trajectory (0.00 through 100.00 ns) are shown.