Vancomycin (VAN) was obtained from Sigma-Aldrich (St. Louis, MO, United States) and Rifabutin (RFB) from Pharmacy Biotech AB (Uppsala, Sweden). Thiazolyl Blue Tetrazolium Bromide (MTT) and Crystal violet (CV) were purchased from Panreac Applichem, ITW Reagents (Darmstadt, Germany). Culture media Mueller-Hinton Agar (MHA), Mueller-Hinton Broth (MHB), and Tryptic Soy Broth (TSB) were obtained from Biokar (Pantin, France) and Columbia Agar +5% sheep blood (CA) from BioMérieux (Marcy L’Ètoile, France). All the remaining chemicals used were of analytical grade.

A collection of 114 clinical isolates recovered from invasive Staphylococcal disease, was kindly provided by the “Laboratório de Microbiologia do Serviço de Patologia Clínica do Centro Hospitalar Universitário de Lisboa Central, Lisboa, Portugal.” The pathology laboratory was requested to recover all cases of invasive S. aureus isolates recovered during the period from April 14 to September 13, 2021. A case of invasive disease was defined as an isolate of S. aureus recovered from blood. All strains were identified as S. aureus by Matrix Assisted Laser Desorption Ionization Time-of-Flight (MALDI-TOF) though VITEK® MS system version 3.2 (Biomerieux, Portugal). A methicillin susceptible S. aureus ATCC®25923™ (MSSA) was also included in this study as a reference strain, obtained from the American Type Culture Collection (ATCC; Manassas, VA, United States). Bacterial stocks were prepared from overnight cultures on MHA or CA at 37°C and stored in MHB with 20% of glycerol at −80°C. Among the isolates collection, a total of 21 methicillin-resistant strains (MRSA) and 93 methicillin-susceptible strains (MSSA) were recovered.

All isolates were evaluated in terms of growth profiles. For this, bacterial suspensions of each isolate were prepared, from an overnight agar culture, in MHB at 0.5 McFarland turbidity standard, equivalent to 10⁸ colony forming unit per mL (CFU/mL) by measuring optical density (OD) at 600 nm. The bacterial suspensions were incubated at 37°C during 24 h. The OD at 600 nm of each suspension was measured at the following incubation time points: 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, and 24 h.

The MIC of RFB and VAN was determined for all clinical isolates, by the broth microdilution method, according to the Clinical and Laboratory Standards Institute guidelines (CLSI, 2020), followed by turbidity evaluation. This concentration range was selected based on a preliminary screen to appropriately define the concentration range (data not shown). The antibiotics were diluted in MHB to produce a 2-fold dilution with concentrations ranging from: 0.0002 to 25.0000 μg/mL and 0.02340 to 24.0000 μg/mL for RFB and VAN, respectively. Bacterial suspensions were performed from overnight cultures on MHA diluted in MHB until reaching a value of 0.5 in a McFarland scale. Bacterial suspensions of each isolate were placed in 96-well culture plate at 5 × 10⁵ CFU/mL and incubated with the respective, antibiotic at different concentrations, at 37°C during 24 h, under static conditions. A negative control containing a suspension of each isolate in MHB, without antibiotic, and a sterile control containing only MHB, were performed in parallel. Minimum inhibitory concentrations (MIC) were determined spectrophotometrically, at 570 nm in an iMark™ microplate absorbance reader (Bio-Rad laboratories, Inc., Hercules, CA, United States) and defined as the lowest antibiotic concentration able to prevent visible bacterial growth, resulting in the absence of turbidity. The MIC of the reference strain was assessed once a week as a control of the assay.

A subset of 40 isolates was selected for evaluation of biofilm formation after initial MIC screening for RFB and VAN on all clinical isolates. This subset included all methicillin-resistant strains (MRSA, n = 21), all strains with VCM MIC values of 1.5 μg/mL (n = 13), four randomly selected strains with VCM MIC values of 0.750 μg/mL (n = 4) and strains with RFB MIC values exceeding 0.025 μg/mL (n = 2). This assay was performed as described previously by Stepanovic et al. (2000) with small modifications. Briefly, bacterial suspensions were inoculated with a final concentration of 10⁶ CFU/mL in TSB supplemented with 0.25% of glucose (TSB 0.25%), in 96-well culture plate. Plates were incubated at 37°C for different time points (6, 24, and 48 h), under static conditions. TSB 0.25% without bacterial suspension was used as a sterile control. Biofilm assembly was evaluated by the crystal violet staining method (CV) (Pontes et al., 2016; Ferreira et al., 2017, 2021c). Following each time point, the content of the wells was washed twice with a sterile solution of phosphate buffered saline (PBS) to remove non-adherent bacteria. The attached bacterial cells were air-dried at room temperature (RT) for 15 min and subsequently stained with 200 μL of a CV solution 0.125% (w/v in water) and incubated at RT for further 15 min. Biofilms were washed twice with sterile PBS to remove excess dye, followed by a drying step at RT. Stained biofilms were then dissolved in 200 μL of ethanol and diluted at a ratio of 1:10. The OD was measured at 570 nm using an iMark™ microplate absorbance reader (Bio-Rad laboratories, Inc., Hercules, CA, United States).

The minimum biofilm inhibitory concentration of 50% bacterial growth (MBIC₅₀) in the set of biofilm-forming isolates previous assessed, was subsequently determined for the two evaluated antibiotics. Thus, overnight cultures were inoculated at 10⁶ CFU/mL in 96-well culture plates in TSB 0.25%. Biofilms were assembled under static condition for 24 h at 37°C. Then each well was washed twice with sterile PBS solution. Serial dilutions of the antibiotics, ranging from 0.0002 to 25.0000 μg/mL and 6.25 to 200.00 μg/mL to RFB and VAN, respectively, were performed in TSB 0.25% and added to the respective wells. Plates were incubated for 24 h at 37°C. A negative control containing a suspension of each isolate in TSB 0.25%, without antibiotic, and a sterile control containing only TSB 0.25%, were performed in parallel. The bacterial cell viability was measured by the MTT reduction assay (Ferreira et al., 2021c). The assay was executed as previously described by Brambilla et al. (2017) with some exceptions. After biofilm rinsing with sterile PBS solution, 200 μL of a 125 μg/mL MTT solution in PBS was added and incubated at 37°C during 2 h. Then the MTT solution was removed and replaced by dimethyl sulfoxide (DMSO) to dissolve the MTT formazan product. The OD of the wells was measured at 570 nm using an iMarkTM microplate absorbance reader (Bio-Rad laboratories, Inc., Hercules, CA, United States). MBIC₅₀ was defined as the lowest antibiotic concentration able to inhibit more than 50% of bacterial growth compared to untreated controls. The determination of MBIC₅₀ was performed by sigmoidal fitting analysis considering a confidence level of 95%. The MBIC₅₀ of the reference strain was assessed once a week as a control of the assay.

Staphylococcus aureus stands out as one of the most significant pathogens worldwide, particularly due to the clinical and epidemiological relevance of invasive infections caused by MRSA, exhibiting high rates of morbidity and mortality (van Hal et al., 2012; Hassoun et al., 2017). The prevalence of systemic MRSA infections worldwide ranges from 20 to 50 per 100,000 population annually in countries with low and high incidence rates, respectively (van Hal et al., 2012; Hassoun et al., 2017). While there has been a decline in the incidence of MRSA bacteremia over the past decade, it continues to be associated with poorer clinical outcomes compared to infections caused by methicillin susceptible S. aureus (MSSA) (Hassoun et al., 2017).

To comprehensively evaluate the antibacterial potential of RFB against S. aureus originating from de facto infections, a collection of 114 staphylococcal clinical isolates were recovered from bloodstream infections. Among these, 93 (81.58%) isolates were susceptible to methicillin (MSSA) and 21 (18.42%) were methicillin resistant strains (MRSA) (Figure 1). Although, the current work was not designed to assess MRSA prevalence in the Portuguese population, the prevalence of MRSA in our study was consistent with the data from Gagliotti et al. (2021), who reported that, based on the European Antimicrobial Resistance Surveillance Network (EARS-Net), 16.3% of the S. aureus strains isolated from bloodstream infections in 2018 were MRSA. Yet, the rate was below the reported from the antimicrobial resistance surveillance report in Europe from 2022, elaborated by the European Center for Disease Prevention and Control (ECDC), which revealed that in 2020, 29.7% of invasive S. aureus isolates tested were MRSA in Portugal (WHO and ECDC, 2022).

Currently, the glycopeptide, VAN, remains the antibiotic of choice for treatment of MRSA infections. However, strains resistant to or with reduced susceptibility to VAN have been emerging (Shajari et al., 2017). Consequently, VAN is normally administered in combination with other antibiotics, such as rifampicin, to improve the therapeutic efficacy and reduce the emergence of resistant strains. Yet, it has severe adverse side effects associated, due to its toxicity profile. To overcome these drawbacks, studies have demonstrated the advantages of using RFB instead of rifampicin in Staphylococcal infections (Doub et al., 2020; Karau et al., 2020; Ferreira et al., 2021c; Thill et al., 2022).

To determine the susceptibility profile to VAN and RFB, of S. aureus clinical isolates recovered from invasive infections, the MIC values were determined. As shown in Table 1, all tested isolates were susceptible to VAN, presenting MIC values ranging from 0.375 to 1.500 μg/mL. The MIC₅₀ value, which is the concentration at which 50% of the clinical isolates are inhibited (Schwarz et al., 2010) was 0.750 μg/mL for both MSSA and MRSA strains, constituting 80.6% of MSSA (n = 75) and 90.5% of MRSA strains (n = 19).

Rifabutin MIC values (Table 2) ranged from 0.002 to 6.250 μg/mL, with n = 43 (46.2%) and n = 15 (71.4%) of the MSSA and MRSA strains presenting MIC values of 0.013 μg/mL, consistent with the RFB MIC₅₀ value. To the best of our knowledge, there are currently no established breakpoints for RFB against S. aureus. Notwithstanding, the data obtained in this study revealed low inhibitory concentrations for RFB for both, planktonic MSSA and MRSA strains and are in line with the previous study from Albano and collaborators who reported a MIC₅₀ = 0.016 μg/mL for S. aureus strains recovered from periprosthetic joint infections (Albano et al., 2019).

Similar susceptibility profiles were obtained for the reference S. aureus strain ATCC®25923™ (MSSA), which was tested in parallel. The reference S. aureus strain presented MICs of 1.500 and 0.006 μg/mL for VAN and RFB, respectively.

It has been known that bacteria organized in biofilms are resistant to most antibiotics, reaching a resistance 1,000 times higher than their planktonic counterparts (Divakar et al., 2019; Doub et al., 2020; Ferreira et al., 2021a; Peng et al., 2022) and researchers believe that more than 80% of chronic infections are associated to bacterial biofilms (Divakar et al., 2019).

In light of the aforementioned considerations, the present study aimed to validate the anti-biofilm activity of RFB in parallel to VAN. For this purpose, a subset of 40 clinical isolates was selected based on the resistance profile previously obtained, with the objective of determining the MBIC₅₀ of RFB and VAN through the evaluation of bacterial cell viability. All MRSA clinical isolates, strains exhibiting VAN MIC values of 1.500 μg/mL, four randomly selected strains with VAN MIC of 0.750 μg/mL, and the two strains with the highest RFB MIC values (RFB MIC = 6.250 and 0.391 μg/mL) were included in the sub-analysis.

The ability of the 40 selected isolates to form biofilms structures was initially confirmed, with mature biofilms observed after 24 h of incubation for all isolates (data not shown). The 24 h old biofilm susceptibility to VAN ranged from 10.00 to more than 200.00 μg/mL, while for the RFB from 0.005 to more than 25.000 μg/mL (Tables 3, 4, respectively). As expected, all clinical isolates in this subset displayed higher inhibitory concentration to VAN in biofilm compared to planktonic state (Tables 1, 3). Indeed, MBIC₅₀ superior to 200.00 μg/mL was observed for 47.4% (n = 9) and 23.8% (n = 5) of the selected MSSA and MRSA strains, respectively. For the reference strain ATCC®25923™, a MBIC₅₀ higher than 200.00 μg/mL was achieved, exhibiting a biofilm susceptibility profile similar to the one observed for the selected set of clinical isolates.

Previous studies corroborate this increased resistance in bacteria organized in biofilms compared with their planktonic counterparts. A study conducted by Rose and Poppens (2009) demonstrated an 8-fold increase in VAN anti-biofilm activity in relation to VAN MIC values in a collection of 40 MRSA clinical isolates. In more recent publications, MICs values for VAN lower than 4 μg/mL were found in a set of clinical isolates recovered from healthy, subclinical and clinical mastitic human milk samples from lactating women, while in the same isolates organized in biofilms, treatment with VAN at 4x MIC led to significant increase in biofilm biomass compared with the untreated control (Angelopoulou et al., 2020). Furthermore, Douthit et al. (2020) suggested that VAN can effectively eradicate S. aureus biofilms, but only in concentrations above 6,000 μg/mL.

The majority of MSSA (n = 9; 47.4%) and MRSA (n = 13; 61.9%) isolates presented a decrease of bacterial cell viability above 50% for RFB concentrations ranging from 0.010 to 0.050 μg/mL. Surprisingly, one MSSA (5.3%) and three (14.3%) MRSA strains did not show a biofilm decrease above 50% with the tested concentrations, leading to the conclusion that these strains presented RFB MBIC₅₀ higher than 25.000 μg/mL. As observed for some clinical isolates, the reference strain showed RFB MBIC₅₀ values similar to the respective MIC (RFB MIC ATCC®25923™ = 0.006 μg/mL; RFB MBIC₅₀ ATCC®25923™ = 0.005 μg/mL).

To the best of our knowledge, the in vitro anti-biofilm activity of RFB have not yet been assessed in S. aureus clinical isolates recovered from bloodstream infections. However, previous studies have already demonstrated the antibacterial potential of RFB in the treatment of S. aureus associated to biofilm infections. Indeed, Doub et al. (2020) validated the in vivo benefit of RFB instead of rifampicin in patients with chronic staphylococcal infections associated to prosthetic material. Remarkably, none of the patients experienced recurrence of staphylococcal infections or reported any adverse side effects with RFB therapy (Doub et al., 2020). Moreover, Karau et al. (2020) conducted a study investigating the therapeutic efficacy of RFB either alone or in combination with VAN in a rat model of foreign body osteomyelitis. Their results demonstrated a higher reduction in colony forming units (CFU) counts in the animals treated with RFB in combination with VAN than the common treatment, rifampicin with VAN, suggesting once more the therapeutic potential of rifabutin in these type of infections (Karau et al., 2020).