Inhibition of α-hemolysin activity of Staphylococcus aureus by theaflavin 3,3'-digallate

Abstract

The ongoing rise in antibiotic resistance, and a waning of the introduction of new antibiotics, has resulted in limited treatment options for bacterial infections, including these caused by methicillin-resistant Staphylococcus aureus, leaving the world in a post-antibiotic era. Here, we set out to examine mechanisms by which theaflavin 3,3'-digallate (TF3) might act as an anti-hemolytic compound. In the presented study, we found that TF3 has weak bacteriostatic and bactericidal effects on Staphylococcus aureus, and strong inhibitory effect towards the hemolytic activity of its α-hemolysin (Hla) including its production and secretion. A supportive SPR assay reinforced these results and further revealed binding of TF3 to Hla with KD = 4.57×10-5 M. Interestingly, TF3 was also able to protect human primary keratinocytes from Hla-induced cell death, being at the same time non-toxic for them. Further analysis of TF3 properties revealed that TF3 blocked Hla-prompting immune reaction by inhibiting production and secretion of IL1β, IL6, and TNFα in vitro and in vivo, through affecting NFκB activity. Additionally, we observed that TF3 also markedly attenuated S. aureus-induced barrier disruption, by inhibiting Hla-triggered E-cadherin and ZO-1 impairment. Overall, by blocking activity of Hla, TF3 subsequently subdued the inflammation and protected the epithelial barrier, which is considered as beneficial to relieving skin injury.

Fig 1. Bacteriostatic and bactericidal efficacy of theaflavins and catechins on Staphylococcus aureus USA300.

(A) Bacteriostatic effect of theaflavins and catechins was determined by standard macro-dilution assay after 24h. (B) Bactericidal effect of theaflavins and catechins was determined from broth macro-dilution tube test by sub-culturing it to blood agar plates that do not contain the test agent after 24h incubation; control—0.02% DMSO.

Fig 2. Inhibitory effect of theaflavins and catechins on Hla secretion by Staphylococcus aureus.

(A) Dose-dependent decrease in Hla secretion by S. aureus USA300 after 24h treatment with theaflavins and catechins was quantified by ELISA assay. (B) Dose-dependent decrease in Hla secretion by Staphylococcus aureus Wood 46 after 24h treatment with theaflavins and catechins was quantified by ELISA assay. Significant differences between treatment and control are represented as # p ≤ 0.05, Δ p ≤ 0.01, * p ≤ 0.001; control—0.02% DMSO; dash line—0.5-fold of change reflecting 50% of expression level of target gene.

Fig 3. Inhibitory effect of TF3 on Hla production by Staphylococcus aureus USA300.

(A) Inhibitory effect of TF3 on Hla production. Dose-dependent decrease in Hla protein secretion by S. aureus USA300 after 24h treatment with TF3 as illustrated by western blot. (B) Densitometry of western blot bands. (C) Dose-dependent decrease in hla and agrA genes expression by S. aureus USA300 after 24h of treatment with 12.5–50 μg/ml TF3 assessed by qPCR. Significant differences between treatment and control were assessed by densitometry and are represented as Δ p ≤ 0.01, * p ≤ 0.001; control—0.01% DMSO.

Fig 4. Inhibitory effect of TF3 on Hla activity.

(A) Dose-dependent decrease of rHla hemolytic activity by TF3 as demonstrated by rRBC hemolysis assay. 10% rRBCs were exposed to increasing doses of TF3 for 10 min. and treated with 0.5 μg/ml of Hla for 20 min. (B) Dose-dependent decrease of secreted Hla hemolytic activity by TF3 as demonstrated by rRBC hemolysis assay. 10% rRBCs were added to 100 μl of sample containing supernatant from S. aureus USA300 overnight culture that was supplemented with different concentrations of TF3 and incubated at RT for 10 min. Significant differences between treatment and vehicle are represented as # p ≤ 0.05, Δ p ≤ 0.01, * p ≤ 0.001; control sample—0.5 μg/ml rHla or 100 μl of supernantant+0.01% DMSO, negative controls—1 x PBS or 0.01% DMSO, positive control—1.0% Triton X-100. (C) Effect of TF3 on Hla oligomerization. Dose-dependent decrease in 5 mM deoxycholate-induced oligomerization of 20 μg/ml native Hla was assessed by western blot, controls—0.01% DMSO+native monomeric Hla treated with or without 5 mM of deoxycholate.

Fig 5. Binding of TF3 to Hla of S. aureus.

SPR result of Hla with TF3. Sensorgram curve of TF3 with concentration of 0 μM, 1.56 μM, 3.13 μM, 6.25 μM, 12.5 μM, 25 μM, 50 μM, 100 μM binding to Hla for evaluation of the binding affinity and affinity parameters such as KD, ka and kd.

Fig 6. Cytotoxic effect of Hla and TF3 on human primary keratinocytes.

(A) Cells were treated with native Hla at 0.5 μg/ml concentration alone or with different concentrations of TF3 in order to check the survival rate that was determined using MTT method and measuring absorbance at 570 nm after 24h. (B) Cytotoxic effect of EV-Hla at 0.5 μg/ml concentration alone or with different concentrations of TF3 on cells determined using LDH method and measuring absorbance at 450 nm after 24h. (C) Viability of cells treated with sHla at 0.5 μg/ml concentration alone or with different concentrations of TF3 in order to check the their apoptosis status determined using Annexin V method and measuring fluorescence Ex/Em = 488/530 nm after 24h. Significant differences between treatment and control are represented as * p ≤ 0.001; control—0.05% DMSO.

Fig 7. Cytokines status in vitro and in vivo.

(A) Effect of native Hla alone and with TF3 on activity of NFκB performed as described in Material and Method section; control—0.05% DMSO, positive control—control cells stimulated with PMA at 3 nm concentration. (B) Human primary keratinocytes were treated with different concentrations of TF3 upon EV-Hla and sHla stimulation, respectively. Levels of secreted pro-inflammatory cytokines were assessed by ELISA after 24h post-treatment. (C) Mouse skin tissue samples injected with native Hla at 5.0 μg/ml concentration alone or with 50 μg/ml TF3. Levels of secreted pro-inflammatory cytokines were assessed by ELISA after 48h. (D) Images or representative skin alterations after treatment with 5.0 μg of native Hla alone together with different concentrations of TF3 for 48h stained with H&E; black arrows—noticeable skin pathology as described in Result section, control animals treated with DMSO only n = 4, control animals treated with TF3 only n = 4, animals treated with Hla only n = 8, animals treated with Hla and TF3 n = 8. Significant differences between treatment and control are represented as # p ≤ 0.05, Δ p ≤ 0.01, * p ≤ 0.001.

Fig 8. Effect of Hla and TF3 on skin barrier.

(A) Status of key molecules of adherence and tight junctions at the cellular level analyzed by western blot using antibodies against E-cadherin, ZO-1, and claudin. (B) Densitometry of western blot bands. (C) Representative images of mouse skin after Evans blue dye penetration upon treatment with native Hla at 5.0 μg/ml concentration alone or together with 50 μg/ml concentration TF3 (see Table 1). (D) Quantification of Evans blue dye and fluorescein-labeled OVA penetration into mouse skin after treatment with native Hla at 5.0 μg, respectively. Significant differences between treatment and control are represented as Δ p ≤ 0.01, * p ≤ 0.001; control—0.05% DMSO.