Inhibition of Drug Resistance of Staphylococcus aureus by Efflux Pump Inhibitor and Autolysis Inducer to Strengthen the Antibacterial Activity of β-lactam Drugs

Staphylococcus aureus infection and drug resistance problems have caused increasing public health problems. The increase in antimicrobial resistance coupled with intracellular infection makes this bacteria the third-largest threat to human health according to the WHO (Lowy 1998; Demon et al. 2012). MRSA is of particular concern because of its ability to spread extensively and rapidly, along with its multi-drug resistance to β-lactam and aminoglycoside antibiotics (Boucher et al. 2009; Kolendi 2010). MRSA infection is always associated with chronic or recurrent infections, including osteomyelitis, pulmonary infection, and endocardial inflammation (Que et al. 2005). In China, almost 10% of S. aureus clinical isolates were considered resistant to penicillin in recent years (Hu et al. 2016; Chen et al. 2017). Several new targets have been discovered and addressed in recent drugs, including ClpP protease and FtsZ of the cell division machinery. Resistance can be modified and inactivated by enzymatic drugs, enzymatic modification of drug-binding sites, drug efflux, and the others. Studies on the resistance of the current antibiotics have been reported also using drug combinations (Foster et al. 2017). Several synergistic combinations of small molecules and antibiotics have also been provided to treat S. aureus infection by reversing the resistance mechanisms, attenuating S. aureus virulence and/or interfering with quorum sensing (Vermote et al. 2017). Thus, the urgency is required in the development of the new strategies for antimicrobial drug combinations against MRSA.

One novel strategy is to utilize helper compounds in combination with traditional antibiotics. Helper compounds are drugs approved for other therapeutic purposes that also possess antibacterial activity (Dickey et al. 2017). Thioridazine (TZ) is primarily an antipsychotic drug and also functions as an efflux pump inhibitor, which can be used as a helper compound (Klitgaard et al. 2008; Pule et al. 2016). Several in vitro studies have shown that TZ significantly increases the susceptibility of MRSA to β-lactam antibiotics (Poulsen et al. 2013). It has been shown that cytoderm synthesis and autolysis are linked since the inhibition of the former activates the latter. Thus, the destruction of the cytoderm is an important step in the bactericidal process of penicillin and other antibiotics (Zore et al. 2011). Miconazole (MCZ) is an antifungal drug that is considered as an autolysis inducer, which causes a release of cellular K⁺ at low concentrations, and MCZ at the minimum inhibitory concentration (MIC) showed a certain antibacterial effect on clinically isolated MRSA (Falk et al. 2010). In our study, we aimed to suppress the multiple drug resistance of MRSA by combining cloxacillin (CXN) with an autolysis inducer MCZ and an efflux pump inhibitor TZ.

The previous studies have shown that the innate immune response, including pattern recognition receptors (PRR), was activated upon infection with S. aureus (Elazar et al. 2010). The release of cytokines is an important indicator for the evaluation of antibiotics. When inoculated with S. aureus, immune cells produce the inflammatory cytokines such as tumor necrosis factor-α (TNF-α), interleukin 1β (IL-1β), and interleukin 6 (IL-6) in high concentrations (Chen et al. 2017). S. aureus may also stimulate nuclear factor-ĸB inhibitor (IKB), nuclear factor-ĸB (NF-KB), and mitogen-activated protein kinase phosphorylation (Gao et al. 2015). The previous studies have shown that TNF-α is the earliest and primary endogenous mediator and plays crucial role in both inflammatory and neuropathic hyperalgesia (Zhang et al. 2007). In these studies, TNF-α and IL-6 after infection were released at high concentrations for 24 h, and there was no time gradient for IFN-γ detection after infection, which was consistent with previous reports (Trigo et al. 2009; Hu et al. 2010). In the studies on the mechanism of action of antibiotics, the level of cytokines was measured at 12–24 h (Wei W et al. 2009; Fu Y et al. 2014). The inflammatory cells including macrophages regulate inflammatory responses by the induction of significant inflammation and release of inflammatory cytokines (IL-1β, IL-6, and TNF-α) and chemokines (Kim et al. 2015) at the high concentrations.

Our study is the first to explore the combination of TZ, MCZ, and CXN in the treatment of mastitis using a mouse model to provide a basis for subsequent combined antibacterial therapy.

Mouse mastitis treatment results. Two S. aureus mastitis mouse models were constructed, each injected with a single strain of S. aureus, either ATCC 29213 or the MRSA135 strain (Fig. 1A). Clinical observations showed that none of the mice in either the ATCC 29213 or MRSA135 groups died during the experiment, and their mental states were normal. The areolas of both bacterial infection groups were swollen and red. The areolas in the MCZ or TZ monotherapy group were pale red and swollen, but there were no significant changes in the other groups. In the MRSA-infected groups, the mammary glands varied in color; those of the blank control group were milky white, those of the infection control group were purple, those of the MCZ or TZ treated group were red, those of a portion of the CXN treated group were red but most were milky white, and most were also milky white in the group treated with the combination of the three drugs (Fig. 1B). In the ATCC 29213-infected groups, the mammary glands varied in color; the blank control group had a normal mammary gland color, the infection control group had purple mammary glands, most of the glands in the MCZ- and TZ-treated groups were red, and those of the CXN-treated and combination-treated groups were mostly milky white (Fig. 1C).

Fig. 1A i 1B.

Fig. 1C.

Histopathology of mammary tissue by the hematoxylin and eosin stain (H&E) was observed (Fig. 2A, and 2B). No obvious pathological changes were found in the blank control group, which displayed normal acinar mammary glands and neat rows of acinar epithelial cells with prolactin. A large number of bacterial lumps and detached cells were seen in the infection control group, and both had red blood cell infiltrate observed in the interstitial space. The agglomerated cell mass was seen in the acinus, and large numbers of lymphocyte infiltrate were observed in the interstitium. In the MCZ or TZ monotherapy groups, epithelial cells were swollen, obvious bacterial masses were visible, and lymphocytes had infiltrated the stroma. Few lymphocytes and certain bacterial masses were observed in the CXN group. In the combination group, few lymphocytes and very few bacteria were visible, in addition to a lack of cell shedding. The results showed that, in addition to the known effect of CXN monotherapy, the combination of CXN, TZ, and MCZ has obvious therapeutic effects against infection by both strains of S. aureus tested. In the group treated with the combination of the three drugs, most of the pathological changes were milder than the CXN treatment group.

Fig. 2.

S. aureus count results. Colony counts from mice of the clinical strain MRSA135-infected group showed that the bacterial concentration was 6.96 × 10⁷ CFU/ml without therapy. Following treatment, the bacterial concentration in mice of the MCZ-treatment group was 2.23 × 10⁷ CFU/ml; in the TZ-treatment group it was 1.73 × 10⁷ CFU/ml; after CXN treatment it was 3.20 × 10⁶ CFU/ml (p = 0.0447 < 0.05), and after the combination therapy with CXN, TZ, and MCZ it was 1.10 × 10⁶ CFU/ml (p = 0.0427 < 0.05) (Fig. 3B). There was no significant difference between the CXN monotherapy- and the combination therapy groups (Fig. 3C). The colony counts after the mice were infected with the reference strain ATCC 29213 showed the following: the bacterial concentration without therapy was 2.34 × 10⁵ CFU/ml; in the MCZ-treatment group it was 1.56 × 10⁵ CFU/ml; in the TZ-treatment group it was 7.91 × 10⁴ CFU/ml; after CXN treatment it was 1.17 × 10⁴ CFU/ml (p = 0.0212 < 0.05); and after the combination therapy with CXN, TZ, and MCZ was 4.43 × 10³ CFU/ml (p = 0.0191 < 0.05) (Fig. 3A). The difference between the CXN monotherapy and the three-drug treatment groups was significant (p = 0.0040 < 0.01) (Fig. 3D).

Fig. 3A i 3B

Fig. 3.

Cytokine detection in a mouse model of mastitis. Serum supernatants were assayed for TNF-α, IL-6 and IFN-γ levels using an ELISA kit (Fig. 4). The cytokines measured in the sera of mice infected with ATCC 29213 (the ATCC 29213 infected group) were as follows: the infected group had significantly increased levels of TNF-α, IL-6, and IFN-γ when compared to those in the control group (p = 0.0260, p = 0.0348, p < 0.0001, respectively) (Fig. 4A, 4C, and 4E). There was no significant difference in TNF-α, IL-6, and IFN-γ levels between mice treated with MCZ when compared with those of the control group. The levels of IL-6 and IFN-γ in the TZ-treated group were significantly lower than those in the ATCC 29213-infected group (p = 0.0176, p = 0.0046), but there was no significant difference in TNF-α levels. The levels of TNF-α, IL-6, and IFN-γ in the CXN-treated group were significantly lower than those in the infected group (p = 0.0016, p = 0.0245, p < 0.0001, respectively). The levels of TNF-α, IL-6, and IFN-γ in the group of mice treated with three drugs together were significantly lower than those in the ATCC 29213-infected group (p = 0.0004, p = 0.0136, p < 0.0001). The levels of TNF-α and IFN-γ in the group of mice treated with three drugs together were significantly lower than those in the ATCC 29213-infected group (p = 0.0084, p = 0.0280). The cytokine results in the MRSA135-infected mice (the MRSA135-infected group) were as follows: the MRSA135-infected group had significantly higher levels of IL-6 and IFN-γ than those in the control group (p = 0.0185, p = 0.0148, respectively) (Fig. 4B, 4D, and 4F). The levels of TNF-α and IL-6 in the MCZ-treated group were not significantly different from those the MRSA135-infected group, but the IFN-γ levels were significantly decreased (p = 0.0434). There were no significant differences in TNF-α, IL-6, and IFN-γ levels in the TZ-treated group compared with those of the MRSA135-infected group. The levels of IL-6 and IFN-γ in the CXN-treated group were significantly lower than those the MRSA135-infected group (p = 0.0191, p = 0.0262, respectively), but there was no significant difference in TNF-α levels.

Fig. 4.

The levels of TNF-α, IL-6, and IFN-γ in mice treated with the three drugs together were significantly lower than those of the MRSA135-infected group (p = 0.0230, p = 0.0051, p = 0.0060, respectively). The TNF-α levels in mice treated with the three drugs together were significantly lower than those of the MRSA135-infected group (p = 0.0010). In summary, treatment with CXN alone or with the combination of three drugs was capable of inhibiting the expression of TNF-α and IL-6.

A recent report indicated that developing of new natural compounds or combination therapies should be a focus on the fight against S. aureus (Celenza et al. 2012; Dickey et al. 2017) Combinations of antibiotics have been previously used to treat bacterial infections, including pathogens that cannot be suppressed or killed by a single antibiotic or infection with a multiplicity of microbial species (Navon-Venezia et al. 2005). A single antibiotic is hardly capable of killing bacteria that possess multiple drug resistance mechanisms against broad-spectrum β-lactams and aminoglycosides antibiotics (Wax 2008). However, bacteria can be inactivated through the joint use of a synergistically active antibacterial agent along with the antibiotic (Mascaretti 2003; Tegos and Mylonakis 2012). Combination therapy can improve the antibacterial effect and reduce the risk of drug resistance during treatment, thereby reducing drug toxicity (Tegos and Mylonakis 2012; Breser et al. 2018). In addition, it has been found that when synthetic peptides of host defense bind to conventional antibiotics, synergistic effects can reduce the concentration of antibiotics required to eradicate certain bacterial strains of interest (Rudilla et al. 2016). Alternatively, the two antimicrobial agents combination can also neutralize the biofilm development (Hwang et al. 2013). However, the antibiotic enhancement remains a challenge, and clinical treatment of bovine mastitis also lacks preclinical animal and clinical data to validate its utility (Tse et al. 2017). This study demonstrates that the drug combination provides good effects in the in vitro assays and in the in vivo treatment of mouse mastitis, and can provide a basis for clinical development. Following the CLSI recommendations, supplementary tests should be performed, even when the penicillin MIC (≤ 0.12 mg/l) is within the drug-sensitive range in vitro (CLSI 2015). The combination of TZ and β-lactam antibiotics may enhance efficacy against S. aureus as a synergistic effect. In this study, the results of in vivo experiments indicated that the concentration of cloxacillin alone (20 mg/ kg/d) was higher than in the combination with the two other drugs (0.75 mg/kg/d), and the treatment effect was better for the combination of drugs. Thus, this study indicates the concentration of drugs that reduce drug resistance when used synergistically, providing some new ideas for drug resistance research.

Previous studies reported that TZ functions as an external pump inhibitor, and MCZ acts as an autolytic inducer against S. aureus (Pule et al. 2016). Therefore, we investigated these two drugs in combination with CXN to inhibit the resistance of S. aureus through the induction of bacterial autolysis, thus enhancing the antibacterial effect of CXN. In this study, the drug susceptibility results showed that S. aureus had particular resistance to CXN. Again, the two drugs showed a synergistic effect in inhibiting S. aureus. The synergistic effect of CXN and MCZ was better that of CXN and TZ. The synergistic effect of the three drugs studied not only reduced the concentration of CXN required but also enhanced the antibacterial effect. Altogether, our results suggested that treatment with a combination of TZ and CXN showed a stronger inhibitory effect against S. aureus in vitro when compared to the effect of CXN monotherapy. When TZ and CXN were used in combination, the MIC value was significantly reduced. In the checkerboard test, the FICI model is commonly used to determine the synergy between anti-staphylococcal drugs. Moreover, the in vivo dosage of the administered combination compound was determined by the ratio of in vitro synergistic combination of the MIC values, the single in vivo doses administered in mouse studies or calculated from the other animal studies already reported.

In S. aureus-induced mouse mastitis, cytokines released by immune cells can aggravate the inflammatory response of mastitis (De and Mukherjee 2009). It has been suggested that TNF-α, IL-1β, and IL-6 may play an important role in the mechanism of milk rupture (Persson et al. 2003). In vivo treatment of S. aureus-induced mouse mastitis showed that CXN monotherapy inhibited inflammation and resolved the redness of mouse mammaries. The inflammatory response was lower in both the MCZ-treated and TZ-treated groups compared with the CXN-treated group. Redness of mammary glands was alleviated in the three-drug treatment group, and there were fewer inflammatory cells in the tissue section than in the other groups. In the pathological changes, it was shown that in both mice groups infected with the bacteria, most of the pathological changes were milder in the group treated with the combination of the three drugs than the CXN-treated group. Our results showed that the combination of three drugs could significantly inhibit the expression of TNF-α, IFN-γ, and IL-6, both in vitro and in vivo. Cytokine assays in mice sera revealed that CXN alone induced IL-6, IFN-γ, and TNF-α in the ATCC 29213-infected mouse groups, and the combination of the three drugs significantly reduced IL-6, IFN-γ, and TNF-α concentrations. The levels of TNF-α and IFN-γ in the sera of mice treated with three drugs simultaneously were significantly lower than those in the CXN-treated group. In the MRSA135-infected mouse group, CXN alone significantly reduced IL-6 and IFN-γ levels but did not significantly reduce the level of TNF-α. The combination of the three drugs significantly reduced IL-6, IFN-γ, and TNF-α concentrations. The level of TNF-α in the sera of mice treated with three drugs simultaneously was significantly lower than those in the CXN-treated group. Therefore, the results demonstrated that the combined use of the three drugs has a significant therapeutic effect on mastitis in mice infected with S. aureus, which may be due to the inhibition of the production of inflammatory cytokines by these three drugs applied together.

In our study, we investigated the effects of TZ, MCZ, and CXN on S. aureus, as well as the antibacterial effect of the combination of the three drugs, both in vitro and in vivo. We found that a bacterial efflux pump inhibitor and an autolysis inducer could be used in combination to inactivate the drug resistance of S. aureus, thus enhancing the efficacy of the antibiotic CXN. To enhance our resources against the bacterial attack, the research on gene expression effects after the combination therapy should be explored in further studies.