The S. aureus teicoplanin- and vancomycin-intermediate-resistant clinical strain, hereafter referred to as wild-type (WT), was obtained from Lady Reading Hospital Peshawar. Tryptic soy broth (TSB), Luria Bertani broth (LB), and cation-adjusted Mueller Hinton broth (MHB) were used for S. aureus, E. coli, and MIC testing, respectively. Bacterial cultures were refreshed from −80°C and were grown in TSB (S. aureus) and LB (E. coli) at 37°C with shaking at 220 rpm. Bacterial strains were grown in round bottom tubes (14 mL tube with 3 mL culture media) and persister assays were performed in a conical flask (100 mL flask with 30 mL culture media). Cells were washed with ddH₂O. Temperature-sensitive (at 30°C) vector pBTs were used for mutant construction in S. aureus, pRMC-2 as an inducible vector for gene expression, and pALC with green fluorescent protein (GFP) gene was used for fluorescence assay. For plasmid maintenance, media were supplemented with appropriate antibiotic concentration, i.e., ampicillin 150 μg/mL and chloramphenicol 20 μg/mL. The pSD1 is a pRMC2 derivative vector that contains dCas9 and sgRNA expression cassettes. S. aureus RN4220 and E. coli DH5α were used for transformation. The primers, plasmids, recombinant vectors, and bacterial strains used in this study are listed in Table 1.

We performed RNA sequencing to study the transcriptome of the resistant isolate. S. aureus 24 h culture was 100 times diluted and was grown for 2 h (OD₆₀₀ = 2) in TSB medium, cells were harvested, washed, and dissolved in RNAiso Plus (Takara Tokyo, Japan) and preserved at −80°C. RNA sequencing was performed in two biological replicates by the core sequence facility of the Institute of Microbiology Chinese Academy of Sciences. A NanoPhotometer® (Implen CA, USA), the Qubit® RNA Assay Kit and Fluorometer (Life Technologies CA, USA), and the RNA Nano 6,000 Assay Kit from the Agilent Bioanalyzer 2,100 system were used to measure RNA purity, concentration, and integrity, respectively. RNA degradation or contamination was checked by agarose gel (1%).

Staphylococcus aureus cells were collected and dissolved in 1 mL of RNAiso Plus. The cell lysis was performed with the help of 0.1-mm silica beads in the FastPrep-24 automated system, and then the lysate was treated with DNase I to remove the remaining DNA. For reverse transcription and qPCR, the PrimeScript cDNA synthesis kit and SYBR Premix Ex Taq reagent kit (Takara Tokyo, Japan) were used, respectively. The StepOne real-time PCR system was used for RT-qPCR analysis. The hu gene cDNA abundance was used for normalization (Valihrach and Demnerova, 2012).

For gene knockdown, the anhydrotetracycline (ATC) inducible vector pSD1 was used which expresses dcas9 and custom-designed sgRNA. The pSD1 vector was derived from Streptococcus pyogenes cas9 and was used as a CRISPR interference system (CRISPRi) (Zhao et al., 2017). From the vector design, the dcas9 and sgRNA are under the control of the P_tetO and P_pflB inducible and constitutive promoters, respectively. For tcaA/tcaB/tcaR knockdown, gene specific complementary oligonucleotide sequences were synthesized and cloned into the SapI-digested pSD1 site. The sgRNA 5′ variable region binds to the target gene while the 3′ constant region binds to dCas9. The dCas9 functions as a DNA-binding protein guided by sgRNA which is complementary to each target gene. The pSD1 plasmid without the target gene (WT-dCas9) was used as a negative control that can also express the sgRNA which has no specific target site in S. aureus.

To create the tcaA mutant, the upstream and downstream regions were amplified by primer pairs PBts-tcaA-F-KpnI and PBts-tcaA-R-0, and PBts-tcaA-F-0 and PBts-tcaA-R-SacI from S. aureus WT genome, respectively, (Table 1). The PBts-tcaA-R-0 and PBts-tcaA-F-0 are the overlap primers that contain the overlap segments, and a 1,400-bp fragment was generated by PBts-tcaA-F-KpnI and PBts-tcaA-R-SacI in Prime Star PCR conditions: 95°C for 5 min, 95°C for 30 s, 55°C for 1.5 min, 72°C for 1 min, 35 cycles, and 72°C for 10 min. A 1400-bp joint fragment was constructed, sequenced, and purified to clone into pBTs, a modified vector of pBT2 (Brückner, 1997; Bae and Schneewind, 2006) via restriction enzymes. The 1,400 bp fragment and pBTs vector were digested with KpnI and SacI for 1 h at 37°C, the mixture was purified by a DNA purification kit separately and ligated into pBTs by DNA ligase. The constructed recombinant vector was initially transferred to DH5α, then transferred to RN4220 for genomic modification, and lastly to S. aureus WT. S. aureus colonies carrying the recombinant vector were selected by tryptic soy agar plus chloramphenicol (20 μg/mL) at 37°C. S. aureus carrying the recombinant vector pBTs-tcaA was grown at 30°C with shaking at 220 rpm for 24–72 h. Daily screening was performed on ATC inducible tryptic soy agar plates at 37°C. PCR was performed to screen the mutant colonies. The tcaA markerless mutant was confirmed by DNA sequencing.

We followed our previous method of antibiotic susceptibility testing in S. aureus (Habib et al., 2020). Initially, the tcaA complementary strain (C-tcaA) was constructed by amplifying the tcaA gene through primer pairs pRMC-tcaA-F-KpnI and pRMC-tcaA-R-SacI, and the vector WTpRMC-tcaA was constructed and transferred to S. aureus WT. The 24 h fresh culture of S. aureus, ∆tcaA, and C-tcaA (complementary strain) was grown for 2 h in TSB (OD₆₀₀ = 2.0) and 2 × 10⁹ CFU/mL were challenged with 100 fold MIC of teicoplanin, vancomycin, and azithromycin at 37°C. The inhibition zones were mapped after 24–36 h. MIC was determined in MHB medium using the two-fold broth microdilution method (Clinical and Laboratory Standards Institute, 2018). Briefly, the double dilutions of antibiotics were distributed into microtiter plates wells, and freshly prepared bacterial suspension (5 × 10⁵ CFU/mL) was added to each well, and plates were incubated at 35°C. S. aureus colonies suspension with a density of 5 × 10⁵ CFU/mL required a transfer of 100 μL of the 0.5 McFarland equivalent suspension to 10 mL of broth. Of note, CLSI does not recommend performing vancomycin susceptibility testing using the disk diffusion method.

The protein expression levels of tcaA, tcaB, and tcaR were detected by specific antisera for tcaA, tcaB, and tcaR through Western blotting. The dCas9-tcaA, dCas9-tcaB, and dCas9-tcaR were induced at OD₆₀₀ = 0.4 with ATC 100 ng/mL for 1 h. The cells were collected, and media were removed and washed twice with ddH₂O. An equal number of cells were taken and lysed by lysostaphin for 1 h at 37°C. The mixture was washed with normal saline and total proteins were collected in lysis buffer. A 12% SDS-PAGE was used for whole-cell protein separation and proteins were electrotransferred to a polyvinylidene difluoride membrane. The tcaA, tcaB, and tcaR proteins were detected with rabbit anti-tcaA, anti-tcaB, and anti-tcaR antibodies (1:1000) followed by horseradish-peroxidase conjugated antibody (1:10,000 dilution). The membrane was analyzed using a Thermo Fisher chemiluminescent detection kit and spots were detected using an ImageQuant LAS 4000.

A persistence assay was adopted from our previous protocol (Habib et al., 2020). The cells were grown to OD₆₀₀ = 0.40 in the TSB medium. The pSD1 containing dCas9-tcaA and dCas9-tcaB were induced with ATC 100 ng/mL for 1 h, the medium was removed, and fresh TSB and antibiotic were added whereas ΔtcaA cells were challenged with antibiotic at OD₆₀₀ = 0.40. After 12 h, a 200 μL sample was taken and CFU counting was performed. Each antibiotic concentration was 10 fold of MIC, and S. aureus WT and S. aureus WT-dCas9 were used as controls.

Fluorescence microscopy was used to observe the persister cell resuscitation ability of the whole population. S. aureus WT and ΔtcaA were transferred with pALC fluorescence shuttle plasmid with GFP and challenged with 20-fold MIC for 48 h. The cells were harvested, washed with ddH₂O, and resuspended in TSB media. The cells were grown for 2 h at 37°C and the S. aureus WT and ΔtcaA were observed under the fluorescence microscope.

The overnight culture was 100 times diluted in TSB medium and a 200 μL was inoculated into a 96-well microtiter plate. The plates were incubated at 37°C with shaking at 200 rpm. At different time intervals (1–18 and 24 h), OD₆₀₀ was measured using a microplate reader (Thermo Fisher, Waltham, MA, USA).

Experiments were performed in biological triplicates unless otherwise stated. The data values were analyzed by Student’s t-test for two groups (unpaired, two-tailed) and a one-way analysis of variance for more than two groups. The gene sequences were analyzed by vector NTI advance and data were analyzed by GraphPad prism 8. * p < 0.05; ** p < 0.01; *** p < 0.005.

Staphylococcus aureus isolated from a COVID-19 patient was resistant to methicillin, teicoplanin, and vancomycin, while sensitive to azithromycin and ciprofloxacin. The MIC of methicillin was >512 μg/mL, teicoplanin was 16 μg/mL, and vancomycin was 10 μg/mL (Table 2). RNA sequencing results of WT revealed that multiple genes and pathways have been affected, comprising 62 differentially expressed genes. The differential expression of highly affected genes is shown by a heat map (Figure 1A). Among the differential expressed genes, Clusters of Orthologous Genes (COG) involved in cell motility (N), extracellular structures (W), prophages and transposons (V), cell division, and chromosome partitioning (D) were marginally expressed whereas amino acid transport and metabolism (E), inorganic ion transport and metabolism (P), and general function prediction (R) were highly expressed (Figure 1B). The COG group associated with nucleotide transport and metabolism (F), translation (J), transcription (K), secondary metabolites biosynthesis (Q), intracellular trafficking (U), defense mechanism (V), cell wall/membrane/envelope biogenesis (M) were slightly expressed (Figure 1B). Differentially expressed genes such as ClpP, EsxB, EsxA, toxin-like hypothetical protein, glyS, Cro, sgtB, ddl, alr2, and vWbp genes expression was high whereas tcaA, essC, tcaB, FtsL, and ABC transporter permease gene expression was low (Table 3). We screened out the genes responsible for glycopeptide resistance in S. aureus which is reported to be the tcaRAB operon. The tcaRAB is stimulated when S. aureus is exposed to a minimum inhibitory concentration of glycopeptides. In the tcaRAB operon, the tcaA is a 454-residues zinc ribbon domain-containing protein, the tcaB is a 402-residues multidrug efflux MFS transporter, and tcaR is a 151-residues MarR family transcriptional regulator. The WT strain RNA sequencing data revealed a significant decrease in tcaA (p = 0.008) and tcaB (p = 0.04) expression while tcaR was not significantly altered (p = 0.08) (Figure 1A and Table 3). We validated the RNA sequence data by RT-qPCR which confirmed that tcaAB and essC genes were significantly suppressed whereas EsxAB, ClpP, glyS, and sgtB were significantly upregulated (Figure 2 and Table 3). The EsxAB are the secretory protein of type seven secretion system (T7SS), ClpP is involved in proteostasis, glyS is glycine tRNA synthetase, sgtB is mono-functional peptidoglycan glycosyltransferase, ddl codes for D-alanine-D-alanine ligase/synthetase, and alr2 is alanine racemase. From this data, we proposed that EsxAB, ClpP, glyS, and sgtB expression was upregulated when tcaA was suppressed which might be linked with cell wall protection against wall-damaging agents such as glycopeptides and viruses. Besides, the T7SS ATP synthesis machinery gene essC was significantly downregulated (Figure 2) which is associated with S. aureus survival during host infection. Collectively, these results corroborated the RNA sequencing data and confirmed that tcaAB suppression is associated with glycopeptides resistance either to protect the cell wall against wall-piercing agents or to increase survival during host infection.

We performed CRISPRi-mediated tcaA, tcaB, and tcaR knockdown by dCas9 in S. aureus. Gene-specific sgRNAs were designed that can bind to the target site of tcaA, tcaB, and tcaR (Table 1). Media were supplied with 100 ng/mL of ATC to induce the CRISPR-dCas9 to suppress gene expression. Initially, the dCas9 repression efficiency was determined in both the knockdown and WT strain by RT-qPCR. The data revealed that the strains expressing sgRNA1, sgRNA2, and sgRNA3 exhibited a 22-fold, 20-fold, and 26-fold decrease in the tcaA, tcaB, and tcaR mRNA levels, respectively, (Figure 3A). Further, we analyzed the dCas-tcaA, dCas-tcaB, and dCas-tcaR protein expression level by rabbit anti-tcaA, anti-tcaB, and anti-tcaR antibodies through Western blotting. The immunoblots showed brighter tcaA, tcaB, and tcaR bands in S. aureus WT whereas suppression was seen in dCas-tcaA, dCas-tcaB, and dCas-tcaR carrying S. aureus (Figures 3B–D). The dCas9 was used as a negative control to check the inhibitory effects of the dCas-9 vector on tcaRAB proteins. From Figures 3B–D, it is confirmed that CRISPRi-mediated tcaA, tcaB, and tcaR knockdown was successful and can be used for persister assay.

We performed a persistence assay with the WT, WT-dCas9, dCas9-tcaA, ΔtcaA, and dCas9-tcaB strains. The WT-dCas9 was the S. aureus WT strain expressing target unspecific sgRNA (control), dCas9-tcaA was the WT strain expressing sgRNA binding to tcaA (sgRNA1), dCas9-tcaB was WT expressing sgRNA binding to tcaB (sgRNA2), and ΔtcaA was tcaA mutant in S. aureus WT. All the strains were challenged with 10-fold MIC of teicoplanin, vancomycin, ciprofloxacin, and azithromycin which revealed that the dCas9-tcaA and ΔtcaA had a higher number of persisters compared to WT-dCas9 and WT strain. After 12 h of teicoplanin treatment, the surviving fraction of cells of the dCas9-tcaA and ΔtcaA strains were 17 and 18 times more than control strains WT-dCas9 and WT, respectively, (Figure 4A). Similarly, upon vancomycin treatment, the dCas9-tcaA and ΔtcaA showed a 15- and 16-fold increase relative to controls, respectively, (Figure 4B). After 48 h, the dCas9-tcaA and ΔtcaA showed a 13- and 14-fold increase in persister cells in response to teicoplanin and a 10- and 11-fold increase toward vancomycin relative to controls, respectively, (Figures 4A,B). The dCas9-tcaB did not induce persister cell formation and was similar to WT-dCas9 and WT strain (Figures 4A,B). When the cells were challenged with azithromycin and ciprofloxacin, the dCas9-tcaA, ΔtcaA, and dCas9-tcaB showed similar results with controls (Figures 4C,D). Overall, neither of the strains showed persister cell formation toward azithromycin and ciprofloxacin that confirmed the emergence of glycopeptides persisters due to tcaA suppression and deletion. Further, we transferred the pALC shuttle vector (with GFP) to S. aureus WT and ΔtcaA competent cells and challenged them with a 20-fold MIC of teicoplanin and vancomycin. After 48 h, the cells were washed and resuspended in a fresh TSB medium without antibiotics, and fluorescence microscopy was performed. The microscopy revealed a high number of cells expressing GFP in ΔtcaA cells compared to WT (Figure 5). This confirmed that persister cells tolerated a high concentration of teicoplanin and vancomycin (20 fold MIC) for 48 h and resuscitated the whole population when the antibiotic was removed. Collectively, the tcaA gene inactivation developed persistence to glycopeptide antibiotics while the tcaB did not influence the persistence phenotype. This data disclosed that S. aureus ΔtcaA formed persister cells and confirmed the RNA-seq and RT-qPCR results where suppression of tcaA was associated with the development of glycopeptides resistance. To date, Brandenberger et al. (2000) and Maki et al. (2004) reported ΔtcaA involvement in glycopeptides resistance, and the present data corroborated their results and reproduced the findings that the ΔtcaA showed resistance to teicoplanin and vancomycin and C-ΔtcaA strain restored the phenotype by expressing the tcaA gene via WT-pRMC-tcaA (Supplementary Figure S1).

We tested cell growth in the TSB medium with 1–2 mg/L of teicoplanin and vancomycin. The results showed that the MRSA MW2 reference strain displayed slow growth rates at 1 mg/L, whereas the WT, ΔtcaA, and C-ΔtcaA strains displayed fast and steady growth (Figures 6A,B). At 2 mg/L, MRSA MW2 did not show any growth whereas WT, ΔtcaA, and ΔtcaA complementary strain displayed slow growth (Figures 6C,D). These results confirmed a decrease in susceptibility of the WT, C-ΔtcaA, and ΔtcaA to teicoplanin and vancomycin compared to MW2. We conclude that S. aureus WT, ΔtcaA, and C-ΔtcaA strains grow at 1–2 mg/L concentration of glycopeptides which supports the fact that tcaA is involved in the emergence of glycopeptides resistance.

The present data indicated that tcaA is involved in cell wall-associated glycopeptide resistance and persistence. From RNA-seq data, the tcaA was significantly suppressed and tcaA deletion increased persister cell formation. During tcaA suppression, the glyS, sgtB, ddl, and alr2 transcript level was high. We hypothesized that evaluating these gene expressions in ΔtcaA would validate the RNA-seq data and disclose the correlation. From RT-qPCR analysis, the glyS and sgtB expression was significant in the tcaA mutant while ddl and alr2 transcript level was not significant (Figure 7). The high expression of glyS and sgtB might help in cell wall biogenesis because glyS is glycine tRNA synthetase which is involved in the supply of glycine for incorporation into nascent polypeptides during bacterial cell wall synthesis (Schneider et al., 2004; Giannouli et al., 2009) whereas sgtB is a mono-functional peptidoglycan glycosyltransferase which is involved in peptidoglycan synthesis and also supports the growth of S. aureus in the absence of the main glycosyltransferase pbp2 (Wang et al., 2001; Reed et al., 2015). Collectively, this data revealed that the tcaA inactivation altered the expression of cell wall-associated genes, probably allowing the cell wall to better withstand external pressures, and it would be interesting to explore glyS and sgtB involvement in cell wall biogenesis during glycopeptide treatment or in the persister cell formation.