The Role of mprF Mutations in Seesaw Effect of Daptomycin-Resistant Methicillin-Resistant Staphylococcus aureus Isolates

ABSTRACT

The emergence of daptomycin-resistant (DAP-R) Staphylococcus aureus strains has become a global problem. Point mutations in mprF are the main cause of daptomycin (DAP) treatment failure. However, the impact of these specific point mutations in methicillin-resistant S. aureus (MRSA) strains associated with DAP resistance and the “seesaw effect” of distinct beta-lactams remains unclear. In this study, we used three series of clinical MRSA strains with three distinct mutated mprF alleles from clone complexes (CC) 5 and 59 to explore the seesaw effect and the combined effect of DAP plus beta-lactams. Through construction of mprF deletion and complementation strains of SA268, we determined that mprF-S295A, mprF-S337L, and one novel mutation of mprF-I348del within the bifunctional domain lead to DAP resistance. Compared with wild-type mprF cloned from a DAP-susceptible (DAP-S) strain, these three mprF mutations conferred the seesaw effect to distinct beta-lactams in the SA268ΔmprF strains, and mutated mprF (I348del and S337L) did not alter the cell surface positive charge (P > 0.05). The susceptibility to beta-lactams increased significantly in DAP-R CC59 strains, and the seesaw effect was found to be associated with distinct mutated mprF alleles and the category of beta-lactams. The synergistic activity of DAP plus oxacillin was detected in all DAP-R MRSA strains. Continued progress in understanding the mechanism of restoring susceptibility to beta-lactam antibiotics mediated by the mprF mutation and its impact on beta-lactam combination therapy will provide fundamental insights into treatment of MRSA infections.

INTRODUCTION

Staphylococcus aureus is one of the major causes of hospital and community-acquired infections. Methicillin-resistant S. aureus (MRSA) accounts for at least 25% to 50% of S. aureus infections in hospital settings, resulting in serious consequences, including central line-associated bloodstream infections, endocarditis, and osteomyelitis (1, 2). Daptomycin (DAP) or vancomycin (VAN) monotherapy is the recommended first line of treatment for MRSA bacteremia (3). Owing to the limited tissue distribution and emergence of isolates with reduced susceptibility to VAN, DAP has become an important alternative therapy targeting MRSA infections (4).

DAP is a cyclic lipopeptide approved by the Food and Drug Administration (FDA) for use in a wide variety of S. aureus infections, such as bloodstream infections (BSI) and endocarditis. However, clinical S. aureus strains with mprF mutations, especially MRSA, have been reported to be associated with DAP treatment failures. These mutations involve altering surface charge, cell wall thickness, and cell membrane synthesis (5–8). Most studies have investigated mprF functions in clinical strains and have not evaluated other mutations in genes influencing mprF, which have been associated with the DAP-resistant (DAP-R) phenotype, such as yycFG, rpoB, rpoC, pgsA, and cls (9). Thus, it is necessary to study mprF mutations in a defined genetic background. Yang et al. showed that the S. aureus (Newman) strain mprF deletion mutant exhibited increased susceptibility to DAP, and mprF-S295L and mprF-T345A increased the DAP MIC and net surface positive charge (10).

The “seesaw effect” describes a phenomenon in which DAP resistance sensitizes MRSA to beta-lactams, which has been observed both in vitro and in vivo (11–14). Some researchers have shown that the mprF mutation could contribute to beta-lactam sensitization, which is involved in the effectiveness of DAP-oxacillin (OXA) combination (15). Furthermore, various studies have revealed that combinations of DAP with nafcillin (NAF) (in vivo), as well as other beta-lactams, displayed strong synergistic interactions against DAP resistance in MRSA isolates (11). However, it has been demonstrated that the mprF mutation does not always correlate with enhanced OXA susceptibility (10, 16). Thus, it remains to be elucidated whether all mprF mutations could restore susceptibility to beta-lactam antibiotics and the mechanism by which mutated mprF influences the effectiveness of the DAP-beta-lactam combination against DAP-R MRSA.

In a previous study, we collected 20 sequential clinical DAP-S and DAP-R MRSA isolates from three infective endocarditis patients (5). In this study, we investigated the impact of the specific mutations of mprF genes on susceptibility to DAP and beta-lactams in vitro. We aimed to evaluate the synergistic effects of DAP in combination with OXA against MRSA strains with different mprF mutations using the time-kill assay. This research provides new insights for the development of effective treatments for MRSA infection.

RESULTS

Distinct point mutations in mprF lead to DAP resistance.

In our previous studies, different mutations (S295A, I348del, and S337L) on mprF genes were detected in DAP-R MRSA isolates. All three mutations are located at the junction of the flippase domain and the synthase domain of MprF (5). To elucidate the contribution of distinct mprF mutations to DAP resistance in a defined genetic background, each mutation was cloned into a plasmid and then transferred to the S. aureus strain SA268ΔmprF. Phenotypic analysis comparing SA268 and its isogenic SA268ΔmprF showed that the inactivation of mprF led to increased susceptibility to DAP; the MIC decreased from 0.5 μg/ml to 0.0625 μg/ml (Table 1). Furthermore, complementation of SA268ΔmprF with wild-type mprF (Δ-pmprF_SA268) and respective parental mprF (Δ-pmprF_SAZ1, Δ-pmprF_SAW1, and Δ-pmprF_SAC1) restored the susceptibility to DAP (MIC, 0.5 μg/ml). Importantly, three mutants complemented with mprF containing S295A, I348del, or S337L mutations (Δ-pmprF_S295A, Δ-pmprF_I348del, and Δ-pmprF_S337L) revealed higher DAP resistance, with MICs of 2, 4, and 4 μg/ml, respectively, compared to their isogenic mutants complemented with mprF from DAP-susceptible (DAP-S) strains. Thus, these results indicated that these three mprF mutations confer DAP resistance in S. aureus.

TABLE 1.

MICs of DAP and beta-lactams for SA268 and mprF derivatives

Strain	MIC (μg/ml) of:
	DAP	FOX	OXA	CZ	AMO	PEN
SA268	0.5	64	32	16	8	4
SA268-p	0.25	64	64	16	8	4
ΔmprF	0.0625	16	8	4	4	2
ΔmprF-p	0.125	16	16	8	8	4
Δ-pmprFSA268	0.5	128	64	16	8	4
Δ-pmprFSAZ1	0.5	128	32	64	16	4
Δ-pmprFS295A	2	64	32	16	8	2
Δ-pmprFSAW1	0.5	64	32	32	16	8
Δ-pmprFI348del	4	32	4	2	8	4
Δ-pmprFSAC1	0.5	128	32	32	8	4
Δ-pmprFS337L	4	16	8	4	8	2

There was no significant impact of the mprF deletion and point mutations on the MICs of other non-beta-lactam antibiotics, such as quinolones, macrolides, and glycopeptides (see Table S2 in the supplemental material).

Contribution of mprF mutations to the seesaw effect.

The beta-lactam MICs were measured to investigate the association between the seesaw effect and the mprF mutations from DAP-R strains. All 20 clinical DAP-S and DAP-R MRSA isolates were resistant to cefoxitin (FOX), OXA, and penicillin (PEN), except for SAC4A and SAC4B, which exhibited susceptibility to OXA at an MIC of 0.5 μg/ml (see Table 2). Although there were no breakpoints for amoxicillin (AMO) and cefazolin (CZ) in the Clinical and Laboratory Standards Institute (CLSI) guidelines, we observed that the beta-lactam MICs of all SAZ strains that belonged to clone complex (CC) 5 were higher than that of the other three CC59 MRSA strains. Furthermore, compared with their isogenic DAP-S strains, the susceptibility to beta-lactams increased significantly among SAW5 with mprF-I348del and SAC4A and SAC4B with mprF-S337L, whereas SAZ7 (DAP = 2 μg/ml) with mprF-S295A did not influence the phenotypic variability of beta-lactams. In particular, the beta-lactam MICs of SAC4A/B declined by ≥4-fold. For example, the MIC of OXA declined from 32 μg/ml to 0.5 μg/ml, and the AMO MIC declined from 16 μg/ml to 2 μg/ml, suggesting that CC59 strains with the mprF mutation displayed the seesaw effect, whereas the SAZ7 (CC5) strain did not.

TABLE 2.

MICs of DAP and beta-lactams for clinical MRSA isolates

Strain	CC	ST/SCC meca	MIC (μg/ml) of:
			DAP	FOX	OXA	CZ	AMO	PEN
SAZ1	5	5/II	0.25	1024	2048	512	64	32
SAZ2	5	5/II	0.25	1024	2048	512	64	32
SAZ3	5	5/II	0.25	1024	2048	512	64	32
SAZ4	5	5/II	0.5	1024	2048	512	32	32
SAZ5	5	5/II	0.5	1024	2048	512	32	32
SAZ6	5	5/II	0.5	1024	2048	512	64	32
SAZ7	5	5/II	2	1024	2048	512	32	32
SAW1	59	59/IVa	0.25	32	128	128	32	16
SAW2	59	59/IVa	0.25	32	64	32	16	8
SAW3	59	59/IVa	0.5	32	64	32	16	8
SAW4	59	59/IVa	0.5	64	128	128	32	16
SAW5	59	59/IVa	1b	16	32	32	8	4
SAC1	59	4513/IVa	0.125	32	32	16	16	4
SAC2	59	4513/IVa	0.25	32	32	8	8	4
SAC3	59	4513/IVa	0.5	64	128	128	32	16
SAC4A	59	4513/IVa	4	8	0.5	2	2	0.5
SAC4B	59	4513/IVa	4	16	0.5	1	2	0.5
SAC5A	59	4513/IVa	0.25	64	64	128	16	8
SAC5B	59	4513/IVa	0.25	64	64	128	16	8
SAC5C	59	4513/IVa	0.25	64	64	32	32	16

^aST, sequence type; SCC mec, staphylococcal cassette chromosome mec.

^bThe MIC of SAW5 was 2 μg/ml irregularly.

To determine whether the mprF point mutations contributed to the seesaw effect, we measured the susceptibility of different beta-lactams in the mprF deletion and complementation MRSA strains. The SA268ΔmprF resulted in increased susceptibility to beta-lactams such as FOX, OXA, and CZ with a ≥4-fold decrease in MICs (Table 1). The strains of Δ-pmprF_SA268 restored beta-lactam susceptibility compared to SA268, indicating that the mprF gene impacted beta-lactam susceptibility in SA268. Compared with strains carrying parental mprF (Δ-pmprF_SAW1 and Δ-pmprF_SAC1), the mutants complemented with two mutated mprF (Δ-pmprF_I348del and Δ-pmprF_S337L) genes produced significant increases in susceptibility to OXA and CZ, as evidenced by a ≥4-fold decline in MICs. The MIC of Δ-pmprF_S295A decreased 4-fold in CZ. Moreover, the susceptibility of PEN and AMO was not significantly changed (≤2-fold) among these strains, indicating that the seesaw effect was variable among distinct beta-lactams.

In vitro activity of DAP-OXA combinations against DAP-S/DAP-R MRSA strains.

Previous observations showed that DAP, in combination with beta-lactams, exhibits significant synergistic effects (17–19). Thus, the time-killing test was performed to investigate whether our clinical DAP-S/DAP-R MRSA strains have a similar phenotype in the presence of 1/2× MIC of DAP and OXA. As shown in Fig. 1A to F, after 24 h of incubation with DAP and OXA, synergistic activity was observed in three DAP-R strains (SAZ7, SAW5, and SAC4A), as demonstrated by strain killing of ≥2-log₁₀ CFU/ml at 24 h compared with the initial inoculum and the most active single agents. Furthermore, the curves demonstrated a dramatic killing starting in the first 8 h. In contrast, all DAP-S MRSA strains were unaffected by the combination of DAP and OXA, with change in killing being <2-log₁₀ CFU/ml. Thus, the combination of DAP and OXA may have a major impact on DAP-R MRSA strains. In addition, our results revealed this combination is equally effective for strains without the seesaw effect.

FIG 1.

In vitro time-kill curves for 7 clinical MRSA strains. Time-kill analyses were performed using MH broth with a 10⁶ CFU/ml inoculum at 0, 1, 4, 8, and 24 h in 1/2× MIC of DAP and OXA. (A, C, and E) DAP-S MRSA strains. (B, D, F, and G) DAP-R MRSA strains.

To further investigate the effect of this combination, in vitro time-kill curves for its respective parental mprF- and mutated mprF-complemented strains were performed (Fig. 2). Similar to the previous results, neither DAP nor OXA used alone showed any bactericidal effects. After 24 h, the combination of DAP and OXA demonstrated indifference to all wild-type mprF-complemented strains, with the differences in counts < 2-log₁₀ CFU/ml change in killing. However, the mprF-mutated strains showed a downward growth, and count decreased ≥2-log₁₀ CFU/ml compared with those of the single agents and the initial inoculum, suggesting that the combination of DAP and OXA is highly synergistic in killing mprF mutant strains.

FIG 2.

In vitro time-kill curves for mprF-mutated and -complemented strains. Time-kill analyses were performed using MH broth with a 10⁶ CFU/ml inoculum at 0, 1, 4, 8, and 24 h in 1/2× MIC of DAP and OXA. (A, C, and E) Wild-type mprF-complemented strains derived from three clinical DAP-S MRSA strains. (B, D, and F) Mutated mprF-complemented strains derived from three clinical DAP-R MRSA strains.

Effect of mprF mutations on cell surface positive charge and cell wall thickness.

To analyze whether the point mutations in mprF could affect the overall S. aureus surface charge, we compared the cytochrome c-binding capacities of strains expressing mprF to those of the wild-type or mutated strains. As shown in Fig. 3, the SA268ΔmprF had a significantly reduced positive surface charge compared to that of the SA268 strain (P ≤ 0.01). Of note, the percentage of unbound cytochrome c of the strain carrying an mprF mutation (S295A) was higher than the strain complemented with its wild-type mprF (P ≤ 0.05). However, the other two DAP resistance-associated mutations (I348del and S337L) did not impact the positive surface charges (P > 0.05) compared with the parental mprF gene. Thus, these results indicate that not all the DAP resistance-associated mprF mutations lead to alteration of the cell surface positive charge.

FIG 3.

Binding of cytochrome c to SA268 and its mprF derivates. Data represent the means (±SDs) from three independent experiments. **, P ≤ 0.01; *, P ≤ 0.05; NS, P > 0.05.

Furthermore, there were no significant increases in cell wall thickness in clinical DAP-R strains compared to the respective DAP-S strain (P > 0.05) (see Fig. S1).

DISCUSSION

MRSA is responsible for the increased disease burden, and in recent years, it has rapidly disseminated among the general population in most areas, including China (20, 21). Resistance to DAP is an emerging clinical problem in the treatment of MRSA infections such as bacteremia and endocarditis. Recently, it has been demonstrated that mprF mutations located at a central bifunctional domain bridging the synthase and translocase domains play a crucial role in MprF function (6, 16, 22). In this study, we determined that mutated mprF genes at S295A and S337L and one novel mutation, I348del, which is located at the junction of the flippase and synthase domains, derived from clinical strains, could mediate DAP resistance (Fig. 4). Notably, our results indicated that change of cell surface charge and cell wall thickness did not necessarily occur in mprF-mutated strains. Some studies have reported the association of increased cell positive charge with mprF mutations (10, 23). However, increasing evidence suggests that DAP resistance-conferring point mutations in mprF do not lead to a general alteration of the cell surface charge and cell wall thickness (24). The membrane lipids need to be evaluated in the future to better understand the phenotypic membrane adaptations in these mprF-mutated strains. Mechanistically, the T345A mutation in mprF may enable the flippase to accommodate DAP, conferring DAP resistance (16).

FIG 4.

MprF topology analysis. Specific mutations within the bifunctional domain bridging the flippase and the synthase domains of MprF mediate daptomycin resistance. The MprF synthase and flippase domains are shown in gray and black, respectively. The predicted MprF topology was modified from previous publications (6, 16).

Novel combinations of last-resort antibiotics like DAP or VAN with beta-lactams have been recommended for the treatment of MRSA infections because beta-lactam susceptibility may improve when lipopeptide or glycopeptide susceptibility is decreased (25, 26). Mutated mprF is a key component of the seesaw effect that results in impairment of PrsA location and chaperone functions, which are essential for penicillin-binding protein 2a (PBP2a) maturation (15). Consistent with previous reports, we determined the contribution of S295A, I348del, and S337L mutations in mprF to the seesaw effect in a defined genetic background of ST59 MRSA. We also observed that this effect was associated with distinct mprF alleles, showing that mprF-I348del and mprF-S337L had a higher beta-lactam susceptibility than mprF-S295A. We also noticed that only some beta-lactams, such as OXA and CZ, showed resensitization in mprF-mutated strains, suggesting the seesaw effect is mainly achieved by beta-lactams targeting the different PBPs. On the other hand, we found the seesaw effect was relatively significant in CC59 strains since two CC59 strains exhibited restored susceptibility to beta-lactams. This might be explained by the diversity in the genetic background of the strain. Some researchers have observed that the strains which express a high level of beta-lactam resistance do not display the seesaw effect (11, 15). Thus, we emphasized that the seesaw effect was limited to specific mutations in mprF, distinct beta-lactams, and genetic background.

Many studies suggest that DAP plus beta-lactam in addition to standard monotherapy against MRSA strains demonstrates a synergistic effect both in vitro and in vivo (27, 28). Several mechanisms have been reported explaining synergistic activity of beta-lactams combined with DAP. Mechanistically, beta-lactams enhanced the activity of DAP against DAP-R MRSA strains by inducing a reduction in the cell surface positive charge, increasing the frequency of septation and cell wall abnormalities, boosting the immune system, and reducing virulence (11, 29, 30). Our study showed that DAP combined with OXA also had synergistic activity, not only in CC59 MRSA strains but also in CC5 strains without the seesaw effect. However, the concentration of OXA used in a synergistic assay for this CC5 MRSA was high, and the clinical efficacy of this combination should be further studied in vivo, especially for MRSA lineages with high beta-lactam MICs.

There are some limitations to our study. First, only mutations in the mprF gene were investigated in the present study. Previous studies showed that multiple genes and pathways were involved in the DAP resistance mechanism and seesaw effect. Here, we constructed mprF mutants in an ST59 strain and confirmed the important role of mprF in the seesaw effect. Second, only three series of clinical MRSA were included in this study, which may not be representative of all the phenotypes for these lineages. Further study, including more diverse CCs and strains, should be conducted. However, the data on the seesaw effect in CC5 and CC59 MRSA are important since they both belong to the predominant clones in mainland China. Finally, only cell surface charge and cell wall thickness were analyzed, which may not fully help us understand the gain-in-function mutations in the mprF gene. More cell membrane phenotypic assays need to be investigated in the future.

To conclude, though the combination of DAP with beta-lactams is recommended to enhance the therapeutic effect for infections such as endocarditis (31), our study illustrated that the efficacy of DAP and beta-lactam combination therapy was strain or clone dependent. Therefore, continued progress in understanding the mechanism of restoring susceptibility to beta-lactam antibiotics mediated by the mprF mutation and its impact on beta-lactam combination therapy will provide fundamental insights into treatment of MRSA infections such as bacteremia and endocarditis.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

The bacterial strains and plasmids used in this study are listed in Table 3. Twenty sequential MRSA strains were isolated from blood cultures of three endocarditis patients as previously described and investigated in this study (5). The wild-type S. aureus SA268 strain, which belongs to ST59, was used for constructing mprF mutants. Strains were grown in tryptic soy broth (TSB; Oxoid, Cambridge, UK) or on tryptic soy agar (TSA; Oxoid, Cambridge, UK) plates unless specified.

TABLE 3.

Strains and plasmids used in this study

Strain or plasmid	Description	Reference or source
Strains
SAZ1	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ2	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ3	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ4	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ5	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ6	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAZ7	Clinically derived MRSA, ST5-CC5, SAZ7 with mutation of S295A in mprF	5
SAW1	Clinically derived MRSA, ST59-CC59, SAW5 with mutation of I348del in mprF	5
SAW2	Clinically derived MRSA, ST59-CC59, SAW5 with mutation of I348del in mprF	5
SAW3	Clinically derived MRSA, ST59-CC59, SAW5 with mutation of I348del in mprF	5
SAW4	Clinically derived MRSA, ST59-CC59, SAW5 with mutation of I348del in mprF	5
SAW5	Clinically derived MRSA, ST59-CC59, SAW5 with mutation of I348del in mprF	5
SAC1	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC2	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC3	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC4A	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC4B	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC5A	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC5B	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SAC5C	Clinically derived MRSA, ST4513-CC59, SAC4A, and SAC4B with mutation of S337L in mprF	5
SA268	CA-MRSA, ST59	39
ΔmprF	SA268, deleted mprF	This study
SA268-p	SA268 strain complementing pTX16	This study
ΔmprF-p	SA268ΔmprF strain complementing pTX16	This study
Δ-pmprFSA268	SA268ΔmprF strain expressing mprF cloned from SA268	This study
Δ-pmprFSAZ1	SA268ΔmprF strain expressing mprF cloned from SAZ1	This study
Δ-pmprFS295A	SA268ΔmprF strain expressing S295A mutation in mprF compared with SAZ1	This study
Δ-pmprFSAW1	SA268ΔmprF strain expressing mprF cloned from SAW1	This study
Δ-pmprFI348del	SA268ΔmprF strain expressing I348del mutation in mprF compared with SAW1	This study
Δ-pmprFSAC1	SA268ΔmprF strain expressing mprF cloned from SAC1	This study
Δ-pmprFS337L	SA268ΔmprF strain expressing S337L mutation in mprF compared with SAC1	This study
Escherichia coli DH5α	Host strain for construction of recombination plasmids
RN4220	A descendant of S. aureus NCTC 8325 employed as a subcloning host	40
Plasmids
pKOR1	E. coli/S. aureus shuttle vector plasmid	32
pTXΔ	S. aureus plasmid derived from pTX15, Tetr	33
pTX16	Control plasmid for pTX expression plasmid series, lipase gene present in pTX15 is deleted	41

Gene deletion and complementation.

Gene deletion mutants were constructed as previously described (32). Briefly, allelic replacement was performed by amplifying approximately 1-kb regions up- and downstream of the gene(s) and cloning them into plasmid pKOR1. The plasmi d was first transformed into S. aureus RN4220 and then into strain SA268. To integrate the plasmid, SA268 (pKOR1-mprF) cultures grown at 30°C in TSB with 10 μg/ml chloramphenicol (TSB_Cm10) were transferred into fresh TSB_Cm10, prewarmed at 43°C, and incubated overnight with rotation at the same temperature. The deletion mutant was verified using PCR and sequencing.

Wild-type and mutated derivatives of mprF were cloned in pTX_Δ and transferred into strain SA268ΔmprF (33). Briefly, mprF was amplified using PCR, digested with BamHI and MluI, cloned into BamHI- and MluI-digested pTX_Δ, and transformed into S. aureus RN4220 and then into the SA268ΔmprF strains via electroporation. These plasmids confer resistance to tetracycline, which was added to growth cultures at a concentration of 12.5 mg/ml. Parental SA268 and SA268ΔmprF strains bearing the empty plasmid pTX₁₆ were used as controls. Primers used in this study are listed in Table S1 in the supplemental material. Oligonucleotide sequences were confirmed using Sanger sequencing.

Determination of susceptibility to beta-lactams and other non-beta-lactams.

The MICs of daptomycin (DAP), cefoxitin (FOX), cefazolin (CZ), oxacillin (OXA), gentamicin (GEN), ciprofloxacin (CIP), rifampin (RIP), sulfamethoxazole-trimethoprim (SMT), clindamycin (CLI), linezolid (LZ), tetracycline (TET), levofloxacin (LEV), erythromycin (ERY), vancomycin (VAN), and teicoplanin (TEC) were determined with broth microdilution in Mueller-Hinton broth (MHB) according to the Clinical and Laboratory Standards Institute (34) ([CLSI]) 2019 guidelines. The MICs of penicillin (PEN) and amoxicillin (AMO) were determined by agar dilution and Iso-Sensitest agar (TopBio, China) as previous study described (35). The terms resistant or resistance are used to describe decreased susceptibility, whether or not the MIC reached the clinically relevant breakpoint levels.

Isolate pairs with a positive seesaw effect were defined by a ≥4-fold decrease in beta-lactam MIC in the DAP-R strain compared to its respective parental DAP-S strain (36).

Analysis of synergy combined with DAP and beta-lactams.

Time-kill curves were determined with DAP and OXA at concentrations of 1/2× MIC, using MH broth containing 50 μg/ml Ca²⁺ with an initial inoculum of 1 × 10⁶ CFU/ml according to previously described criteria (11, 37). Viability counts were performed at 0, 1, 4, 8, and 24 h. Bactericidal synergy was defined as a ≥2-log₁₀ CFU/ml decrease between the combination antibiotic and the most active single agent after 24 h. Moreover, the number of surviving organisms in the presence of the combination had to be ≥2-log₁₀ CFU/ml below the starting inoculum. Lack of difference was defined as a <2-log₁₀ change (increase or decrease) in killing. All experiments were performed in duplicate. Data analysis was performed using Prism 7.0 software.

Phenotypic analysis.

To quantify relative cell surface charge in strains, we employed the cytochrome c binding assays as previously characterized (16, 38). Briefly, exponential-phase bacteria were harvested, washed twice with MOPS (morpholinepropanesulfonic acid) buffer (20 mM, pH 7.0), and adjusted to an optical density at 600 nm (OD₆₀₀) of 1. Aliquots of 1 ml were collected, resuspended in 200 μl MOPS buffer and 50 μl of cytochrome c solution (equine heart, 2.5 mg/ml in MOPS buffer; Sigma), and incubated at 37°C with shaking for 15 min. Supernatants were recovered, and the OD₅₃₀ was measured spectrophotometrically. Data shown for assays are the means (±standard deviations [SDs]) of three independent experiments.

To determine the cell wall thickness of the DAP-S and DAP-R strains, S. aureus cells were prepared for transmission electron microscopy (TEM) (TECNAI-10; Philips) analyses as described previously (24). For each strain, 50 cell wall thickness measurements were obtained from a minimum of 25 cells at ×205,000 magnification (Gatan charge-coupled device [CCD]).

Ethics.

This study was approved by the local ethics committees of Sir Run Shaw Hospital with a waiver of informed consent (approval no. 20190426-2).