Abstract Carbapenem- and colistin-resistant Gram-negative bacteria have become one of the most severe public health issues worldwide. The development of advanced antibacterial agents that can outpace microbial adaptation is imperative. The thioredoxin (Trx) and glutaredoxin (Grx) systems play important roles in maintaining redox homeostasis within Gram-negative bacterial cell membranes, with thioredoxin reductase (TrxR) and glutathione reductase (GR) being classical antibacterial targets. In this study, we found that Au(III) Schiff base complexes Au10 and Au17 exhibited potent activities against carbapenem- and colistin-resistant Gram-negative bacteria, as demonstrated in evaluations both in vitro and in vivo. Mechanistic studies revealed that Au10 and Au17 exert their antibacterial effects through multiple pathways: irreversibly inhibiting TrxR and GR activities via targeting redox-active motifs, impairing bacterial energy metabolism even at low concentrations, degrading deoxyribonucleic acid (DNA), causing reactive oxygen species (ROS) generation and intracellular redox imbalance. This study provides the first evidence that Au(III) Schiff base complexes possess strong activity against carbapenem- and colistin-resistant Gram-negative bacteria and can simultaneously inhibit the oxidoreductase in carbapenem- and colistin-resistant Gram-negative bacteria, establishing a new paradigm for antibacterial strategies and guiding future innovations in antibacterial therapy. Keywords: Au(III) Schiff base complexes, Glutathione reductase, Thioredoxin reductase, Redox imbalance, Multi-target pathway, Drug-resistant bacteria Graphical abstract [49]Image 1 [50]Open in a new tab 1. Introduction The emergence of multidrug-resistant (MDR) Gram-negative bacteria has posed a significant threat to global public health, leading to increased morbidity and mortality [[51]1]. Among these bacteria, carbapenem-resistant Enterobacterales (CRE) are of significant concern. CRE exhibits high resistance to any carbapenem class antibiotics, such as imipenem, meropenem, ertapenem, and doripenem, specifically speaking, with minimum inhibitory concentrations (MICs) of ≥4 mg/L for imipenem, meropenem, or doripenem, and ≥2 mg/L for ertapenem. CRE can produce carbapenemases to degrade carbapenems. This resistance is mediated through various mechanisms, encompassing the production of carbapenemases such as class A enzymes (e.g., Klebsiella pneumoniae carbapenemases, KPC), class B enzymes (e.g., NewDelhi metallo-β-lactamase-1, NDM), and class D enzymes (e.g., OXA-48), overproduction of AmpC enzymes, production of extended-spectrum beta-lactamases (ESBLs), loss of outer membrane porin proteins, and high expression of efflux pumps [[52]2]. Colistin has been revitalized to combat CRE [[53]3]. The acquisition of the colistin resistance gene mcr-1 in CRE is particularly concerning, as polymyxins are considered a last resort for treating MDR infections [[54]4]. However, the proliferation of CRE that produce ESBLs, along with the acquisition of the colistin resistance gene mcr-1 in these bacteria, could result in the development of untreatable bacterial infections [[55]5]. Thus, developing novel antibiotics has become an especially acute need for carbapenem and colistin resistant bacteria. Traditional antibiotics often induce bacterial cell death by interacting with a drug molecule and its single bacterial-specific target to inhibit the essential cellular functions [[56]6]. However, bacteria can resist the inhibitory or killing effects of such single target-based antibiotics by mediating the enzymes that either degrade the enzymes or modify the antibiotic molecules [[57]7]. Thus, it is essential to design or develop novel drugs that target two or more vital physiological processes throughout the bacterial life cycle to overcome antibiotic resistance. Recently, the antimicrobial effects of metallodrugs against drug-resistant bacteria have successfully captured people's attention [[58]8]. Auranofin, an Au(I) complex comprised of tetraacetylated thioglucose and triethylphosphine ligands, is an Food and Drug Administration (FDA)-approved medication that has been used as an anti-inflammatory treatment for rheumatoid arthritis [[59]9]. It has been suggested that auranofin exerts its antibacterial effects primarily by inhibiting the activity of the thioredoxin reductase (TrxR) [[60]10]. Moreover, recent studies have identified auranofin as a dual inhibitor targeting both metallo-β-lactamases and mobilized colistin resistance, resensitizing carbapenem- and colistin-resistant bacteria to antibiotics [[61]5]. Our previous research revealed that gold complexes being modified based on auranofin can also directly degrade bacterial DNA [[62]11]. These results inspired us to consider that gold complexes can exert antibacterial effects through multiple modes of action, and the development of auranofin and other gold complexes as antibacterial agents is a potential therapeutic strategy. The thioredoxin (Trx) and glutaredoxin (Grx) systems are two primary thiol-containing antioxidant systems typically found in many Gram-negative bacteria. They play crucial roles in maintaining redox balance within the cytosol and are involved in various physiological and biochemical processes, such as deoxyribonucleotide synthesis, protein folding, DNA and protein repair, and sulfur metabolism [[63]12]. The Trx system consists of Trx, TrxR, and nicotinamide-adenine dinucleotide phosphate (NADPH). The Grx system consists of Grx, glutathione reductase (GR), glutathione (GSH), and NADPH. Among them, TrxR and GR are promising antimicrobial targets. Auranofin demonstrates strong antibacterial effectiveness against Gram-positive bacteria but exhibits a diminished effect on Gram-negative strains. This reduced activity against Gram-negative bacteria has been attributed to the Grx system, which mitigates the inhibitory effects of the Trx/TrxR system, thereby lowering oxidative stress levels [[64]10]. In addition to this, the impermeability of the outer membrane or efflux pumps of the Gram-negative bacteria, leading to reduced drug uptake [[65]13,[66]14]. Hereby, the current challenge has shifted to a chemical paradox: Can we modify the structure based on auranofin to disrupt the complementary dual antioxidant system of Gram-negative bacteria or to damage the low permeability of the outer membrane to Au uptake, in order to design metal lead complexes with multi-target antibacterial effects? Auranofin is readily metabolized by biomolecules containing thiol groups, leading to the loss of coordinated ligands before interacting with the intended target enzyme [[67]15]. Therefore, it is crucial to design and synthesize gold complexes with enhanced structural stability. Au(III) exhibits a richer coordination environment than Au(I), and Au(III) complexes are often tetradentate. In earlier work, DNA was considered a potential target for anticancer Au(III) complexes [[68]16,[69]17]. Since Au(III) complexes display redox activity, selecting appropriate ligands to stabilize Au(III) is of paramount importance [[70]17]. Schiff bases are regarded as “privileged ligands” within the field of transition metal chemistry due to their straightforward synthesis and their capacity to stabilize metals across a range of oxidation states [[71]18,[72]19]. They exhibit a diverse array of biological activities, including antibacterial and anticancer effects [[73]20]. Furthermore, when coordinated with metal ions, Schiff bases contribute to the enhancement of both the stability and activity of the complexes [[74]21,[75]22]. Specifically, Schiff base ligands containing NNXX motifs (X = N, O, S, Se, or P) exhibit strong coordination with metal ions and effectively regulate the associated parameters, thereby enhancing the pharmacological properties of the resulting complexes [[76]16]. Numerous Schiff base metal complexes (zinc, copper, nickel, iron, and palladium) exhibiting antimicrobial properties have been reported [[77]23]. Consequently, we developed and synthesized a series of novel Au(III) Schiff base complexes with NNOO. Naphthalene groups with known antibacterial activity were incorporated into the structure [[78]24,[79]25]. Additionally, elements such as fluorine, methyl, and methoxy groups recognized for their ability to enhance the antibacterial efficacy of complexes were also introduced [[80]26,[81]27]. The presence of biphenyl and ethylpiperidine structures may strengthen the complexes’ DNA binding affinity and improve solubility, thereby potentially increasing their antibacterial activity [[82]28]. Furthermore, considering the simplicity and precision of synthesizing asymmetric complexes, we chose the R, S-configured diaminocyclohexane (DACH) ligand. To the best of our knowledge, this is the first report of Au(III) Schiff base complexes with antimicrobial activity. 2. Materials and methods 2.1. Chemicals and reagents Auranofin (Yuanye, Shanghai, China) and other gold complexes were dissolved in N, N-dimethylformamide (DMF) (Sigma Aldrich, USA). Antibiotics (Yuanye, Shanghai, China), including imipenem, meropenem, gentamicin, colistin B, gatifloxacin, cefoxitin, and sulbactam, were freshly prepared in sterilized Milli-Q ultrapure water (Millipore, USA), while chloramphenicol is dissolved in anhydrous ethanol (Yuanye, Shanghai, China). Tryptone, yeast extract, and NaCl were purchased from Solarbio Corp (Beijing, China). 2.2. Bacterial strains Klebsiella pneumonia ATCC BAA-1705 (carbapenem- and colistin-resistant Klebsiella pneumonia), Pseudomonas aeruginosa ATCC BAA-2108, Escherichia coli (E. coli) ATCC BAA-2140, and Staphylococcus aureus ATCC 700699 were obtained from the American Type Culture Collection (Manassas, USA). Additionally, four bacterial strains were isolated from hospital, including carbapenem- and colistin-resistant Klebsiella pneumonia (CRKP) (NCBI accession code WH2023-10), CRAB (NCBI accession code [83]JAKLSN000000000), carbapenem-resistant Pseudomonas aeruginosa and carbapenem-resistant E. coli. All bacterial strains were cultured in standard Luria-Bertani (LB) broth containing 10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl (Solarbio, Beijing, China). 2.3. Au(III) complexes synthesis We synthesized complexes Au1 – Au17, following the synthetic route as shown in [84]Fig. 1. Subsequently, Au1 – Au17 were characterized by nuclear magnetic resonance (NMR) spectroscopy and low-resolution mass spectrometry. High-performance liquid chromatography (HPLC) were used to determine their purity. In addition, the superior complexes were tested by liquid chromatograph mass spectrometry (LC-MS), for details please see the Supplementary Material. Fig. 1. [85]Fig. 1 [86]Open in a new tab The structures and synthetic route of Au(III) Schiff base complexes Au1 – Au17. (a) MgCl[2], (CH[2]O)[n], TEA, CH[3]CN, reflux, 22 h; (b) (1) EtOH, 0 °C, 2 h, (2) room temperature, 12 h; (c) NaAuCl[4]·2H[2]O, NH[4]PF[6], DCM, EtOH, Ar, room temperature, 24 h. 2.4. Stability analysis Au10 or Au17 (10 mM) were dissolved in 90 % DMSO-d[6]/10 % D[2]O, and ^1H NMR spectra were taken at 298 K over 72 h (time ranging from 30 min, 3 h, 6 h, 12 h, 24 h, 48 h, to 72 h). Au10 or Au17 (20 mM, DMF) was diluted to 0.1 μM with phosphate-buffered saline (PBS, pH = 7.4), and then the UV wavelength was scanned by using the UV–Vis spectrophotometer at 298 K over 72 h (time ranging from 0 h, 24 h, 48 h, to 72 h; Baseline correction by PBS). 2.5. Solubility analysis Au10 or Au17 (20 mM, DMF) was diluted to 4 mM, 8 mM, 12 mM, and 16 mM using DMF, Then, PBS (pH = 7.4) was added at 1000 times (v[1]/v[2]) and the mixture was mixed thoroughly, resulting in solutions of various concentrations (4 μM, 8 μM, 12 μM, 16 μM, 20 μM) with PBS: DMF = 1000 : 1 (v[1]/v[2]). The UV absorbance at 238 nm were measured sequentially (Baseline Correction by PBS: DMF = 1000 : 1). A standard curve was created ([87]Fig. S6) and the UV absorbance of the samples were measured at 238 nm sequentially. A saturated solution of Au10 or Au17 was made in the same way, and the upper clear solution was taken after centrifugation and diluted 10 times. The absorbance at 238 nm was then measured to determine the solubility of Au10 and Au17. 2.6. Screening of antibacterial activity of Au(III) complexes and ligands The broth microdilution method, in accordance with the guidelines from the Clinical and Laboratory Standards Institute (CLSI), was used to evaluate the antibacterial effects of Au(III) complexes (Au1 – Au17) and ligands (L1 – L17). The complexes were introduced at final concentrations of 40 μM. After incubation at 37 °C for 24 h, the 96-well plates were analyzed using a microplate reader (PerkinElmer, USA). The MICs were determined as the lowest antimicrobial concentrations that showed no visible bacterial growth. 2.7. Time-killing assay After overnight culturing, the bacteria were diluted 2000-fold and incubated until the optical density at 600 nm reached between 0.4 and 0.6. This standardized bacterial culture was then further diluted to an OD[600nm] of 0.1 and used to inoculate a 96-well plate. Different concentrations of gold complexes (Au10, Au17, and auranofin) at 0, 10, 20, 40, and 80 μM were added to the 96-well plate. The bacterial growth dynamics were measured every hour using a microplate reader for a total duration of 12 h. 2.8. Cytotoxicity assay Cell viability was evaluated by methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay. Briefly, human liver cancer cell lines (HepG2, and Huh7) and human normal cells (293T) that were in the logarithmic growth phase were digested with trypsin and subsequently resuspended in culture medium to create a uniform cell suspension. A volume of 100 μL was transferred to a 96-well plate, adjusting the cell density to 5 × 10^3 cells/well or 2 × 10^3 cells/well. The plate was incubated at 37 °C for 24 h. Subsequently, 1 μL of the complex diluted in DMF (0.625 mM, 1.25 mM, 2.5 mM, 5 mM, 10 mM, 20 mM) was added and further diluted to a final volume of 500 μL with culture medium. Then 100 μL of this solution was added to each well resulting in a final volume of 200 μL per well. The final concentrations of the complexes were 0.625 μM, 1.25 μM, 2.5 μM, 5 μM, 10 μM, 20 μM. After incubating for an additional 72 h, 20 μL of MTT solution (5 mg/mL in PBS) was added to each well and then incubated for 4 h. The supernatant was then discarded, and 200 μL of DMSO was added to dissolve the formazan crystals. The absorbance was measured at 570 nm using a microplate reader. Each sample was tested in triplicate wells. Cell viability (%) was calculated using the following formula: Cell viability (%) = (OD[t] - OD[0])/ (OD[c] - OD[0]) × 100 % In this context, OD[t] refers to the optical density of the treatment group, OD[0] denotes the optical density of the blank control, and OD[c]indicates the optical density of the blank culture medium. 2.9. Checkerboard microdilution assay The checkerboard method was employed to screen the synergistic effects of antibiotics in combination with gold complexes (Au10, Au17, and auranofin). Briefly, overnight bacterial cultures were diluted 1000-fold and regrew to the mid-log phase (OD[600nm]–0.6), then further diluted according to the CLSI guidelines and seeded into 96-well plates. The bacterial suspensions were treated with varying concentrations of gold complexes alongside different concentrations of antibiotics. The plates were incubated at 37 °C for 24 h and the absorbance at OD[600nm] was measured using a microplate reader. The fractional inhibitory concentration index (FICI) was calculated according to the method previously described [[88]29]. 2.10. Determination of Au concentration in CRKP An overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6). The bacterial suspensions were treated with 20 μM of gold complexes (Au10, Au17, and auranofin) for 2 h. After treatment, the samples were pelleted and washed three times with ice-cold PBS, then adjusted to an OD[600nm] of 0.7 ± 0.02. A 1 mL sample of the bacterial culture was collected from each group, pelleted, digested with 0.5 mL of hydrochloric acid (HCl), and then 4.5 mL of double-distilled water was added. The samples were analyzed by Thermo X series 2 Inductively coupled plasma mass spectrometry (ICP-MS) (Thermo, USA) to determine the final Au content. The bacteria-associated Au was quantified by normalizing the concentration of Au to the bacterial densitymeasured at OD[600nm][.] 2.11. Cellular TrxR activity assay Cellular TrxR activity was measured using the TrxR activity detection kit (Solarbio, Beijing, China). Briefly, an overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6). The bacterial suspensions were exposed to varying concentrations (0, 2.5, 5, 10, 20 μM) of gold complexes (Au10, Au17, and auranofin) for 4 h. Afterward, the cell pellets were collected and washed three times with ice-cold PBS, and normalized to OD[600nm] of 0.7 ± 0.02, after which, the cellular TrxR was extracted, purified, and measured according to the manufacturer's instructions. 2.12. Expression of CRKP TrxR in E. coli The DNA sequence encoding TrxR (331 amino acids, 993 bp) from CRKP was synthesized by Sangon Biotech Corp. (Shanghai, China). This sequence was subsequently cloned into the pET-28a vector, incorporating an N-terminal 6 × His tag via Nco I/BamH I restriction sites. The recombinant protein was expressed in E. coli. Specifically, E. coli BL21 cells were transformed with the TrxR-containing vector and cultured overnight at 37 °C with shaking at 200 rpm. The following day, a portion of the overnight culture was transferred into fresh LB medium and incubated for an additional 2 h until reaching an optical density of 0.4- at 600 nm. To induce expression, IPTG was added to a final concentration of 0.5 mM, and incubation continued at 24 °C for 24 h with shaking. The cells were then harvested by centrifugation, and the resulting pellets were collected for purification procedures. 2.13. Purification of recombinant CRKP TrxR The bacterial pellets were resuspended in a buffer consisting of 50 mM Tris, 150 mM NaCl, and 20 mM imidazole. The cells were lysed through three freeze-thaw cycles, combined with 1 mg/mL lysozyme and ultrasonication on ice. Following centrifugation at 12,000 rpm for 40 min, the supernatant was collected and passed through a nickel affinity column (HisTrap FF crude, Cytiva, Sweden). The eluted fractions were concentrated using a 3-kDa cut-off ultrafiltration device (15 mL, Millipore, USA) before being loaded onto a HiPrep Sepharcryl S-300 gel filtration column (16/60 HR, Cytiva, Sweden). The concentration of CRKP TrxR was assessed by measuring the absorbance of FAD at 463 nm, employing an extinction coefficient of 11,300 M^−1 cm^−1. 2.14. Inhibition of CRKP TrxR by gold complexes For the inhibition study, a concentration of 2 μM CRKP TrxR was pre-reduced with 100 μM NADPH at room temperature for 10 min to reduce the disulfide bond between Cys^145 and Cys^148. The reduced TrxR was subsequently incubated with Au10, Au17, and auranofin, respectively. The remaining enzymatic activity of TrxR was assessed by using the 5, 5'-Dithiobis-(2-nitrobenzoic acid) (DTNB) reduction assay. The Ki for the gold complexes were determined by varying the DTNB concentrations from 0 to 2 mM and the gold complex concentrations from 0 to 40 μM, using a competitive inhibition model analyzed with Prism 8.0 software (GraphPad, USA) [[89]30]. After pre-reduction with NADPH (100 μM for 10 min), TrxR was treated with 30 μM of each complex for 30 min and subsequently passed through a NAP-5 desalting column (Cytiva, Sweden). The activities of both the desalted and untreated enzymes were evaluated using the DTNB reduction assay. The TrxR enzyme was incubated with the gold complexes (Au10, Au17) at room temperature for 30 min. After incubation, the mixture was transferred into a dialysis bag with a 15,000 Da molecular weight cutoff (Solarbio, China) and dialyzed at 4 °C for 4 h. Following dialysis, enzyme activity was assessed by adding the enzyme to a reaction mixture containing NADPH and DTNB. The change in absorbance at 412 nm was measured before and after incubation at 37 °C for 5 min to evaluate whether the inhibition was reversible or irreversible. Preincubation with GSH reduces the inhibitory effects of Au10 and Au17 on TrxR. Each gold complex was preincubated with GSH at specified molar ratios for 10 min. Following this, NADPH-reduced TrxR was incubated with 25 μM GSH conjugated Au10 and Au17 for an additional 30 min, and small aliquots were taken from the incubation systems and the residual TrxR activity were assessed by using the classic DTNB reduction assay. Iodine oxidation was carried out by adding a 30 μM iodine solution to the TrxR and incubating for 30 min at room temperature. After this oxidation step, the TrxR enzyme was incubated with Au10 and Au17 at room temperature for 30 min, the residual TrxR activity was determined using the classic DTNB reduction assay. The reduction in activity relative to control samples reflects the extent of iodine-induced oxidation of TrxR. 2.15. Measurement of cellular GR activity Cellular GR activity was measured using the GR activity detection kit (Solarbio, Beijing, China). Briefly, an overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6). The bacterial suspensions were subjected to treatment with varying concentrations of gold complexes (Au10, Au17, and auranofin) at 0, 2.5, 5, 10, and 20 μM for a duration of 4 h. Then, the cell pellets were collected and washed with ice-cold PBS for three time, and normalized to OD[600nm] of 0.7 ± 0.02, after which, the cellular GR was extracted, purified, and measured according to the manufacturer's instructions. 2.16. Recombination and purification of GR The DNA sequence (1399 bp) containing the GR from CRKP was synthesized by Sangon Biotech. Co. (Shanghai, China), and subsequently cloned into the pET-28a vector using NcoI/BamH I restriction enzymes. The recombinant CRKP GR was expressed in E. coli. In brief, the pET28a-GR plasmid was introduced into E. coli BL21 cells, from which single colonies were selected and grown overnight in LB medium at 37 °C with shaking at 200 rpm. The following day, the cultured cells were diluted 1 : 50 and incubated until the optical density at 600 nm reached between 0.6 and 0.8-. Subsequently, 1 mL of 0.5 M IPTG was added to the culture medium, and expression was induced at 18 °C with 200 rpm for 16 h. The bacterial cells were collected by centrifugation at 5000 rpm for 15 min at 4 °C and then resuspended in Buffer C (20 mM Tris, pH = 7.9, 200 mM NaCl, 5 % Glycerol, ddH[2]O). The cells were lysed using a homogenizer until the suspension transitioned from turbid to clear, and the lysate was collected. The lysate was then centrifuged at 12000 rpm for 15 min at 4 °C to collect the supernatant. Nickel resin was added to the supernatant at a 1 : 1 volumetric ratio and mixed at 4 °C for 45–60 min for adsorption. The mixture was filtered through gravity, followed by washing with Buffer C. A gradient elution was performed using 25 mM, 40 mM, 50 mM, and 200 mM imidazole. The elution fractions at 50 mM and 200 mM imidazole were concentrated to 5 mL and subjected to size exclusion chromatography using a S75 molecular sieve. 200 mM imidazole was used to elute the fraction containing the target protein, and determine its protein concentration using A260/A280, and stored in aliquots at −80 °C. 2.17. Inhibition of CRKP GR by gold complexes The reaction system was prepared using a 96-well plate. The buffer solution was composed of 50 mM Tris-HCl (pH = 7.5) and 2 mM EDTA. The purified GR enzyme was diluted to a working concentration of 40 nM, and 10 μL was added to achieve a final concentration of 2 nM in the reaction system. Gold complexes (Au10, Au17, and auranofin) were then added and the mixtures incubated at 37 °C for 30 min, after which oxidized glutathione (GSSG) and NADPH were rapidly added. The reagents were mixed thoroughly, and the absorbance at 340 nm was recorded after 10 s. The mixture reagents were then quickly incubated at 37 °C for 5 min, after which the absorbance at 340 nm was measured and recorded again. The Ki for the gold complexes were determined by varying the GSSG concentrations from 0 to 5 mM and the gold complex concentrations from 0 to 20 μM, using a competitive inhibition model analyzed with Prism 8.0 software (GraphPad, USA). 2.18. Determination of the reversibility of GR activity The GR enzyme was incubated with the gold complexes (Au10, Au17) at room temperature for 30 min, after which the mixture was passed through a NAP-5 desalting column to collect the filtrate. GSSG solution was added into the desalted fraction to achieve a final concentration of 1 mM, followed by the addition of 300 μM NADPH solution. The reagents were mixed thoroughly, and the absorbance at 340 nm was measured and recorded after 10 s. The mixture was incubated at 37 °C for 5 min, and the absorbance at 340 nm was measured and recorded again. The GR enzyme was incubated with the gold complexes (Au10, Au17) separately at room temperature for 30 min. After incubation, the mixture was transferred into a dialysis bag with a 15,000 Da molecular weight cutoff and dialyzed at 4 °C for 4 h. After dialysis, the small aliquots of enzyme were added to the enzyme to a reaction mixture containing NADPH and GSSG. The absorbances of samples at 340 nm were measured before and after incubation at 37 °C for 5 min. 2.19. The effect of GSH-reduced complexes on GR enzyme activity The gold complexes (Au10, Au17) were pre-incubated with different ratios of GSH at room temperature for 10 min, and 2 nM pure GR enzyme was added. Following an additional incubation at room temperature for 30 min, GSSG solution was rapidly added to achieve a final concentration of 2.5 mM, followed by the addition of NADPH solution to a final concentration of 300 μM. The reagents were then mixed thoroughly, and the absorbance at 340 nm was measured and recorded after 10 s. The mixture reagents were then quickly incubated at 37 °C for 5 min, after which the absorbance at 340 nm was measured and recorded again. 2.20. Determination of the NADPH-reduced GR activity The reaction system was prepared using a 96-well plate. A total of 160 μL of buffer solution was added, along with NADPH solution to achieve a final concentration of 300 μM and purified GR enzyme to a final concentration of 2 nM. After incubating NADPH with GR at room temperature for 10 min, Au10 and Au17 were then added to reach final concentrations of 20 μM. Following an additional incubation at room temperature for 30 min, GSSG solution was rapidly added to achieve a final concentration of 2.5 mM, followed by the addition of NADPH solution to a final concentration of 300 μM. The reagents were mixed thoroughly, and the absorbance at 340 nm was measured and recorded after 10 s. The mixture reagents were then quickly incubated at 37 °C for 5 min, after which the absorbance at 340 nm was measured and recorded again. Iodine oxidation was carried out by adding a 30 μM iodine solution to the GR and incubating for 30 min at room temperature. After this oxidation step, the GR enzyme was incubated with Au10 and Au17 at room temperature for 30 min, after which GSSG and NADPH were rapidly added. The reagents were mixed thoroughly, and the absorbance at 340 nm was recorded after 10 s. The mixture reagents were then quickly incubated at 37 °C for 5 min, after which the absorbance at 340 nm was measured and recorded again. 2.21. Intracellular superoxide radical detection Overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6), and then incubated with the dihydroethidium (DHE) probe at 37 °C while shaking for 20 min. Then, the bacteria suspensions were treated with different concentrations (0, 10, 20, 40 μM) of gold complexes (Au10, Au17, and auranofin) for 4 h, after which the samples were pelleted and washed with ice-cold PBS for three time. Fluorescence intensity was measured using the BD Accuri™ C6 Plus flow cytometer, with 100,000 cells collected for the analysis. FlowJo V10 was utilized to process the flow cytometric data. 2.22. Synthesis of 2-hydroxyethidium (2-OH E^+) 2-OH E^+ was synthesized following the procedures described in the relevant literature [[90]31,[91]32]. In a glass container, 1 mL of phosphate buffer (0.5 M, pH = 7.4), 1 mL of diethylenetriaminepentaacetic acid (DTPA, 1.0 mM in water), 6 mL of water, and 2 mL of nitrosodisulfonate (NDS, 1 mM in 50 mM phosphate buffer, pH = 7.4, containing 100 μM DTPA) were added. The solution was then mixed thoroughly, followed by the addition of 50 μL of DHE (20 mM in DMSO). The resulting mixture was incubated at room temperature for 1 h. It is important to note that the concentrations of DHE and NDS should be confirmed using a UV–Vis spectrophotometer. The identity of the synthesized 2-OH E^+ was confirmed by mass spectrometry, as provided in the Supporting Information. 2.23. Preparation of the sample for HPLC analysis Overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.3), and then incubated with the DHE probe (20 μM) at 37 °C while shaking for 20 min. Then, the bacteria suspensions were treated with different concentrations (0, 80 μM) of gold complexes (Au10, Au17, and auranofin) for 4 h, after which the samples were pelleted and washed with ice-cold PBS for three time. Then, the cell pellets were resuspended in 150 μL of TieChui™ E. coli Lysis Buffer. After 30 min of incubation at room temperature, the samples were centrifuged at 20,000 ×g for 10 min. The supernatants were then extracted with 100 μL of ice-cold acetonitrile (CH[3]CN) containing 0.1 % formic acid, vortexed, and incubated on ice for 30 min. After centrifugation at 20,000 ×g for 30 min, 100 μL of the supernatant was mixed with 100 μL of water containing 0.1 % formic acid, vortexed, and centrifuged again. Finally, 150 μL of the supernatant was transferred into HPLC vials and stored at 4 °C for analysis. 2.24. HPLC analyses of 2-OH E^+ The analysis of 2-OH E^+ was performed using HPLC, following the method described in the literature [[92]31]. As detailed in the Supporting Information ([93]Table S2), a Kromasil C18 column (250 mm × 4.6 mm, 5 μm, 12 nm) was used for the analysis, with an injection volume of 50 μL per sample. The mobile phase consisted of 0.1 % trifluoroacetic acid (TFA) in water (A) and 0.1 % TFA in CH[3]CN (B). 2.25. Determination of intracellular K^+, Mg^2+, MDA, SOD and GSH Overnight culture of CRKP was diluted and re-cultured to OD[600nm]–0.3, after which the bacteria suspensions were incubated with different concentrations (0, 10, 20, 40, 80 μM) of gold complexes (Au10, Au17) for 4 h. Then, the cell pellets were collected and washed with ice-cold PBS for three time, and normalized to OD[600nm] of 0.7 ± 0.02. The cellular levels of K^+, Mg^2+, malondialdehyde (MDA), and superoxide dismutase (SOD) were assessed following the protocols of the K^+, Mg^2+, MDA, and SOD activity assay kits (Jiancheng, Nanjing, China). Cellular GSH was measured according to the protocol of the Reduced Glutathione Content Assay Kit (Solarbio, Shanghai, China), utilizing a microplate reader (PerkinElmer, USA). 2.26. Determination of cell membrane integrity Propidium iodide (PI) is a derivative of ethidium bromide (EB) that emits red fluorescence upon intercalation with DNA. PI is unable to penetrate intact cell membranes but can cross damaged cellular membranes, making it useful for nuclear staining. Therefore, it is commonly employed to assess cell membrane integrity. Overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6), and incubated with PI probe at 37 °C and shaken for 20 min. Then, the bacteria suspensions were treated with different concentrations (0, 10, 20, 40 μM) of gold complexes (Au10, Au17, and auranofin) for 4 h, after which the samples were pelleted and washed with ice-cold PBS for three time. Fluorescence intensity was measured using the BD Accuri™ C6 Plus flow cytometer, with 100,000 cells collected for the analysis. FlowJo V10 was utilized to process the flow cytometric data. 2.27. DNA degradation assay DNA extraction from CRKP was performed according to the MiniBEST Bacteria Genomic DNA Extraction Kit ver.3.0 (TaKaRa, Japan). The quality of the extracted DNA was assessed using a micro-spectrophotometer and agarose gel electrophoresis. Then, the qualified DNA solution was mixed with different concentrations of gold complexes (Au10, Au17, and auranofin) and incubated on ice for 30 min. DNA degradation was assessed using EB staining followed by polyacrylamide gel electrophoresis, and the results were documented using a Gel imaging system. Concurrently, the DNA solution was co-incubated with Au10 and Au17 along with 10 mM H[2]O[2] for 30 min, followed by gel electrophoresis to evaluate the combined effects of simulated intracellular oxidative stress conditions and gold complexes on DNA damage. The extracted DNA concentration was diluted to 30 ng/μL, and 3 mL of this DNA solution was incubated on ice with 40 μM concentrations of Au10 and Au17 for 2 h. DNA integrity was assessed using circular dichroism spectroscopy under the following scanning conditions: speed: 100 nm/min, reaction time: 0.5 s, and wavelength range: 220–320 nm. The TUNEL assay (Beyotime, China) was used to assess DNA fragmentation in bacteria by labeling the 3′-OH ends of fragmented DNA. Overnight cultures of CRKP were diluted and re-cultured to the mid-log phase (OD[600nm]–0.6). The bacterial suspension was then treated with various concentrations (0, 10, 20, 40 μM) of gold complexes (Au10, Au17) for 4 h. After treatment, the samples were centrifuged, washed three times with ice-cold PBS, and fixed with paraformaldehyde. The cells were then permeabilized and incubated with terminal deoxynucleotidyl transferase (TdT) and dUTP-fluorescein (or biotin). DNA degradation was subsequently measured using a BD Accuri™ C6 Plus flow cytometer, with 100,000 cells collected for analysis. 2.28. Label-free proteomics Briefly, an overnight culture of CRKP was diluted and re-cultured to the mid-log phase (OD[600nm]–0.6). The bacteria suspensions were treated with 10 μM of gold complexes (Au10, Au17) for 4 h. Then, the cell pellets were collected and washed with ice-cold PBS for three time. Approximately 100 mg of cell pellets were lysed using SDT lysis buffer combined with 1/100 volume of DTT, followed by 5 min of ultrasonication on ice. The mixture was then heated at 95 °C for 8–15 min, cooled in an ice bath for 2 min, and centrifuged at 12,000 g for 15 min at 4 °C. The supernatant was then alkylated with a sufficient amount of IAM for 1 h at room temperature in the dark. Following this, the samples were thoroughly mixed with four times their volume of pre-cooled acetone by vortex and incubated at −20 °C for at least 2 h. Afterward, the samples were centrifuged at 12,000 g for 15 min at 4 °C, and the precipitate was collected. Following a wash with 1 mL of cold acetone, the cell pellet was dissolved in Dissolution Buffer. Subsequently, the supernatants were collected and subjected to protein concentrations determination using Bradford protein quantitative kit. Then, the protein samples were further subjected to proteolytic digestion using the methods described in previously reported method [[94]33]. A gradient elution was performed using mobile phase A (2 % CH[3]CN, pH adjusted to 10.0 with ammonium hydroxide) and mobile phase B (98 % CH[3]CN, pH adjusted to 10.0 with ammonium hydroxide). The lyophilized powder was reconstituted in solution A and subsequently centrifuged at 12,000 g for 10 min at room temperature. The resulting sample was then analyzed using a C18 column (Waters BEH C18, 4.6 × 250 mm, 5 μm) on a Rigol L3000 HPLC system, with the column oven maintained at 45 °C. Peptides were then eluted using the following gradient: 3 %–5 % B in 10 min; 5 %–20 % B in 20 min; 20 %–40 % B in 18 min; 40 %–50 % B in 2 min; 50 %–70 % B in 3 min; 70 %–100 % B in 1 min. The eluates were monitored at UV 214 nm, collected at a rate of one tube per minute, and combined into a total of 10 fractions. All fractions were then dried under vacuum and subsequently reconstituted in 0.1 % (v[1]/v[2]) formic acid (FA) in water. 2.29. UHPLC-MS/MS analyses and data analysis Analyses using UHPLC-MS/MS were conducted at Novogene Co., Ltd. (Beijing, China) with a nanoElute UHPLC system (Bruker, Germany) paired with a tims TOF pro2 mass spectrometer (Bruker, Germany). First, mobile phase A (100 % water with 0.1 % FA) and solution B (100 % CH[3]CN with 0.1 % FA) were prepared. The lyophilized powder was then dissolved in 10 μL of solution A, centrifuged at 14,000 g for 20 min at 4 °C, and 200 ng of the supernatant was then transferred into the LC-MS spectrometry system for detection. The UHPLC used was a nano Elute with a nano-upgraded configuration, and the analytical column was a homemade column (15 cm × 100 μm, 1.9 μm). The liquid chromatography elution was conducted using the following gradient: 2 %–22 % B in 45 min; 22 %–35 % B in 5 min; 35 %–80 % B in 5 min; maintained at 80 % for 5 min. The tims TOF pro2 mass spectrometer was fitted with a Captive Spray ion source, set to a spray voltage of 2.1 kV. The mass full scan range spanned from m/z 100 to 1700, with a ramp time of 100 ms. The Lock Duty Cycle was configured at 100 %. The PASEF settings included 10 MS/MS scans with a total cycle time of 1.17 s, an ionic strength threshold of 2500, and a scheduling target intensity of 20,000. MaxQuant search parameters were established as follows: the mass tolerance for precursor ions was set to 20 ppm, while the mass tolerance for product ions was 0.05 Da. Carbamidomethyl was designated as a fixed modification, oxidation of methionine (M) was noted as a dynamic modification, and acetylation was identified as an N-terminal modification. A maximum of 2 missed cleavage sites was permitted. The protein quantification results were statistically analyzed using a T-test. Proteins that exhibited significant differences in quantification between the experimental and control groups (p < 0.05 and |log[2]FC| > ∗ [where FC > ∗ or FC < ∗ represents fold change]) were classified as differentially expressed proteins (DEP). Gene Ontology (GO) and InterPro (IPR) functional analyses were performed using the interproscan program against a non-redundant protein database (including Pfam, PRINTS, ProDom, SMART, ProSite, PANTHER), while the COG (Clusters of Orthologous Groups) and KEGG (Kyoto Encyclopedia of Genes and Genomes) databases were utilized for analyzing protein families and pathways, as previously described [[95]34]. 2.30. Determination of intracellular ATP Overnight culture of CRKP was diluted and re-cultured to OD[600nm]–0.3, after which the bacteria suspensions were incubated with different concentrations (0, 10, 20, 40, 80 μM) of gold complexes (Au10, Au17) for 4 h. Then, the cell pellets were collected and washed with ice-cold PBS for three time. Cellular adenosine triphosphate (ATP) was determined according to the protocols of the ATP assay kit (Beyotime, China) as previously described [[96]35]. And the BCA Protein Assay kit (Tiangen, China) was used to correct the different samples for comparison. 2.31. Modeling of wound infection Female BALB/c mice, aged 6–8 weeks and weighing 18–20 g, were purchased from GemPharmatech Co., Ltd (No. SCXK2023-0009) and domesticated in the SPF Animal Center of Nanjing University of Chinese Medicine (No. SYSK2023-0077). All procedures received approval from the Experimental Animal Ethics Committee at Nanjing University of Chinese Medicine. The wound infection model was performed as follows: After anesthetizing female BALB/c mice, a circular wound approximately 0.6 cm in radius was created on the dorsal region. A total of 100 μL of CRKP suspension (10^9 CFU/mL) was injected into the wound using a syringe (The CRKP was kindly supplied by Director Ma Ling from the Union Hospital, Tongji Medical College, Huazhong University of Science and Technology; the high-throughput DNA sequencing results have been uploaded to the NCBI database). After 24 h infection, the mouse was divided into six groups, with three mice per group, and treated with 20 μL of either 0.9 % saline, 1 % H[2]O[2], auranofin, colistin B, Au10, or Au17 (all at a concentration of 5 mM) daily for 14 days. The skin around the wound was excised 24 h after the final treatment for the detection of inflammatory cytokines, pathological analysis, and immunohistochemistry. During this period, daily photographs of the wounds were taken, and the wound sizes were measured and recorded. The formula used to calculate wound healing is as follows: F = (A - B)/A × 100 %, where F represents the wound healing rate, A denotes the original wound area, and B indicates the wound area after treatment. 2.32. Modeling of murine peritonitis infection The female BALB/c mice (18–20 g) were randomly divided into the control, colistin B, Au10 and Au17 groups (n = 6). Then, each mouse received an intraperitoneal injection of 200 μL of CRKP suspension (10^9 CFU/mL). After 1 h infection, the mice were treated with PBS, colistin B, Au10 and Au17 respectively. The survival rate was monitored and recorded over 48 h period. To determine Au content in mouse serum, each mouse received an intraperitoneal injection of 200 μL of CRKP suspension (10^9 CFU/mL), followed by an intraperitoneal injection of Au-containing complexes Au10 and Au17 at a dose of 5 mg/kg, respectively. Blood samples were then collected from the eye socket at time points 0 h, 2 h, 4 h, 8 h, 12 h, and 24 h post-injection. Serum was separated by centrifugation at 3000 ×g for 15 min. A 50 μL aliquot of serum was added to 500 μL of concentrated HCl and then diluted with deionized water to a final volume of 3 mL. The Au concentration in the serum samples was determined using ICP-MS (Thermo, USA). 2.33. Statistical analysis All statistical analyses were conducted based on three independent experiments. The experimental data were analyzed using GraphPad Prism 8.0 software. Differences between groups were assessed using the t-test, with p < 0.05 considered statistically significant. 3. Results and discussion 3.1. Synthesis and characterization of novel Au(III) Schiff base complexes S (1 eq) was dissolved in CH[3]CN, followed by the addition of magnesium chloride (MgCl[2]) (2 eq). The reaction mixture was then adjusted to pH ≈ 10 with triethylamine (TEA). Finally, (CH[2]O)[n] (11 eq) was added. After refluxing for 22 h, the mixture was acidified to pH ≈ 2 with HCl and then extracted with ethyl acetate (EA). N was purified on silica gel (PE/EA). M1 – M15 were commercially available. M16 was obtained in the same way as N, while M17 was prepared according to the procedures described in the corresponding literature [[97]36]. N (1 eq) and R, S-1, 2-DACH (1 eq) were dissolved in ethanol (EtOH) at 0 °C and stirred for 2 h. Then M1 – M17 (1 eq) was added, and the mixture was stirred at room temperature for 12 h. L1 – L17 were obtained by recrystallizing from EtOH or purified on silica gel. A dichloromethane (DCM) solution of L1 – L17 (1 eq) was added dropwise to an EtOH solution of NaAuCl[4]·2H[2]O (1 eq). Subsequently, NH[4]PF[6] (5 eq) was added, and the mixture was stirred at room temperature under Ar atmosphere for 24 hto obtain Au1 – Au17 ([98]Fig. 1). Then we characterized ligands L1 – L17 and complexes Au1 – Au17 using ^1H NMR and ^13C NMR Spectra. Additionally, we used mass spectrometry for further characterization of complexes Au1 – Au17 and assessed their purity by HPLC. The purity of Au10 and Au17 was further verified using LC-MS Spectrometry. We are able to observe that L1 – L17 has OH peaks at δ = 13–15 ppm, but the OH peaks disappear for Au1 – Au17. Positive mode mass spectra revealed base peaks corresponding to the [M − PF[6]]^+ fragment for Au(III) complexes Au1 – Au17. The results of stability analysis showed that the ^1H NMR spectra of Au10 and Au17 remained unchanged in 90 % DMSO-d[6]/10 % D[2]O for 72 h, proving that both Au10 and Au17 could remain stable in D[2]O ([99]Figs. S1–S2). Additionally, the stability of Au10 and Au17 in PBS was investigated using a UV–Vis spectrophotometer. The UV absorption spectra of Au10 and Au17 ([100]Fig. S3) did not change considerably during 72 h, demonstrating that complexes Au10 and Au17 were relatively stable in PBS. 3.2. Au10 and Au17 exhibited antibacterial advantages against CRKP First, we investigated the antibacterial effects of the synthesized ligands and metal complexes on MDR bacteria, including CRAB, CRKP, carbapenem-resistant Pseudomonas aeruginosa (CRPA), carbapenem-resistant Escherichia coli (CREC), and intermediate vancomycin-resistant Staphylococcus aureus (VISA). Primary screening showed that none of the 17 synthetic ligands exhibited antibacterial activity at a concentration of 40 μM ([101]Table S1). In our initial screening at a concentration of 20 μM, the gold complexes displayed varied antibacterial efficacy. Notably, the 3OCH[3]-substituted complex Au10 and the 4-O(CH[2])[2]N(CH[2])[5]-substituted complex Au17 exhibited significant activity against Gram-negative bacteria compared to auranofin ([102]Table 1). The F-substituted complexes Au2–Au4 demonstrated good antimicrobial effects, with Au3 (4F-substitution) showing the most prominent activity, although all F-substituted complexes were ineffective against VISA. The CH[3]-substituted complexes Au6, Au7, and Au9 also showed antimicrobial effects, albeit slightly weaker than the F-substituted ones. Additionally, the 5-(tert-butyl)-substituted complex Au14 exhibited better activity than the 3, 5-(tert-butyl)-substituted complex Au15. Overall, complexes with substitutions at the 3- or 4-positions on the benzene ring are more likely to demonstrate significant antimicrobial activity and warrant further investigation. Most complexes showed the highest antibacterial activity against CRKP and CRPA. Carbapenems have traditionally been viewed as the most effective and powerful agents against MDR Gram-negative pathogens. Consequently, the emergence of carbapenem resistance in these bacteria presents a significant clinical challenge and is classified as a critical priority for the research and development of new antibiotics [[103]37]. Colistin has been revived as a last-line therapeutic option for the treatment of infections caused by MDR Gram negative bacteria, therefore resistance to this antibiotic is extremely hazardous [[104]38]. The CRKP strains studied in this research are not only resistant to carbapenem antibiotics but also resistant to colistin ([105]Table 1). Thus, the search for a class of complexes that can overcome resistance to imipenem and colistin, and identify novel antibacterial targets, is expected to provide new insights for drug development. In this study, we used the CRKP as the research subject to further investigate the mechanisms of antibacterial action of the gold complexes (Au10, Au17). Table 1. The minimum inhibitory concentration (MIC) of gold complexes against MDR bacteria. Complexes (μM) CRKP CRKP (Clinical) CRAB (Clinical) CRPA CRPA (Clinical) CREC CREC (Clinical) VISA Au1 >20 >20 >20 20 20 >20 >20 20 Au2 10 10 >20 >20 >20 20 20 >20 Au3 10 10 20 10 10 20 20 >20 Au4 10 10 >20 >20 >20 20 20 >20 Au5 >20 >20 >20 >20 >20 >20 >20 >20 Au6 20 20 >20 >20 >20 20 20 10 Au7 >20 >20 >20 10 10 >20 >20 20 Au8 >20 >20 >20 >20 >20 >20 >20 >20 Au9 10 10 >20 >20 >20 20 20 20 Au10 10 10 20 10 10 20 20 20 Au11 >20 >20 >20 >20 >20 >20 >20 >20 Au12 >20 >20 >20 >20 >20 >20 >20 >20 Au13 >20 >20 >20 20 20 >20 >20 >20 Au14 10 10 >20 10 10 >20 >20 10 Au15 >20 >20 >20 >20 >20 >20 >20 10 Au16 >20 >20 >20 >20 >20 >20 >20 >20 Au17 10 10 20 10 10 20 20 20 Auranofin >20 >20 20 >20 >20 >20 >20 0.5 Imipenem 200 100 200 200 100 50 100 – Colistin B 50 100 3 3 3 1 1 – [106]Open in a new tab Next, we conducted antimicrobial activity assays on gold complexes (Au10, Au17, and auranofin), to further verify the antibacterial advantages of the complexes we designed. The results were shown in [107]Fig. 2A. As expected, Au10 and Au17 shown excellent antibacterial effects against CRKP, which exhibited at least a 16-fold increase in MICs compared to auranofin. In addition, dynamic monitoring experiments over 12 h confirmed that both Au10 and Au17 significantly inhibited bacterial growth as the concentration increased. In contrast, auranofin exhibited only a marginal inhibitory effect on bacteria at the same drug concentrations ([108]Fig. 2B). Additionally, we evaluated the antibacterial activity of metal complexes in conjunction with 11 antibiotics against CRKP using the standard checkerboard microdilution method. The results indicated that, except for cefepime and meropenem, which demonstrated synergistic antibacterial activity, the other 9 antibiotics did not show any synergistic or additive effects with Au10 or Au17 ([109]Figs. S4–S5). In the presence of Au10 (2 μM), the MIC of cefepime was reduced from 40 μM to 5 μM, with a FICI value of 0.45, indicating a synergistic effect between Au10 and cefepime. Similar synergistic effect was observed for Au10 and meropenem against CRKP with a FICI of 0.35. Au17 exhibited a somewhat weaker synergistic effect, demonstrating merely an additive interaction with cefepime, while showing a synergistic effect with meropenem (FICI = 0.45). In contrast, auranofin did not exhibit any synergistic interaction with either of these two antibiotics ([110]Fig. 2C – D and [111]Table 2). These results confirm that the complexes we designed based on auranofin exhibited effectiveness in both single and combined antibacterial activities against carbapenem- and colistin-resistant Gram-negative bacteria. There are several potential mechanisms that may explain this observed synergy, such as enhancing antibiotic penetration, inhibiting β-lactamases [[112]5], disrupting metal ion homeostasis [[113]39], and targeting various bacterial pathways [[114]40], which deserved a further investigation. Fig. 2. [115]Fig. 2 [116]Open in a new tab Antimicrobial activity assays. (A) MIC determination. The MICs of Au10, Au17, and auranofin against CRKP were determined using the broth microdilution method. (B) Time-dependent antimicrobial activity. The antimicrobial effects of Au10, Au17, and auranofin against CRKP were monitored over a 12 h period at concentrations ranging from 0 to 80 μM. (C–D) Synergy evaluation via checkerboard assays. Heatmaps show the results of microdilution checkerboard assays assessing the combinatory effects of cefotaxime or meropenem with Au10, Au17, and auranofin against CRKP. Table 2. Potency of gold complexes in combination with different classes of antibiotics against CRKP. Antibiotics MIC (μM) FIC index (Au10) FIC index (Au17) FIC index (Auranofin) Cefepime 20 0.45 0.70 0.525 Meropenem 40 0.35 0.45 0.625 [117]Open in a new tab 3.3. Au10 and Au17 displayed lower cytotoxicity and higher permeability than auranofin The cytotoxicity assays are usually applied for safety evaluation in early drug discovery. Next, we assessed the cytotoxicity of the Au(III) complexes (Au10, Au17) relative to auranofin in liver cancer cell lines (Huh7, HepG2) and normal cell line (293T). The results showed that Au10 and Au17 exhibited toxicity that were 1–8 times lower than that of auranofin ([118]Fig. 3A–C). These results indicated that Schiff base complexes can reduce the cytotoxicity, as Schiff bases generally exhibit better biocompatibility and lower cytotoxicity. The solubility assays showed that Au10 is soluble up to 54.74 μM, and Au17 up to 127.05 μM in PBS-DMF (1000 : 1, v[1]/v[2]), indicating that these complexes are adequately soluble under experimental conditions ([119]Fig. S6). Furthermore, Schiff bases possess strong cell membrane penetration capabilities, which may enable their derivatives to potentially overcome the barriers posed by the double membrane of Gram-negative bacteria in the development of antimicrobial agents [[120]41]. Subsequently, we conducted further verification using ICP-MS, which accurately determines the content and distribution of metal elements. After a 4 h treatment with 20 μM gold complexes (Au10, Au17, and auranofin), whole-cell uptake was measured. The uptake of Au10 (mean = 734.643 ng/OD[600nm]) and Au17 (mean = 1319.64 ng/OD[600nm]) was higher than that of auranofin (mean = 290.524 ng/OD[600nm]), as demonstrated by the results ([121]Fig. 3D). We simultaneously monitored the dynamic intracellular whole-cell uptake of Au in CRKP over a period of 1–4 h following 20 μM gold complexes (Au10, Au17) treatment ([122]Fig. 3E). Fig. 3. [123]Fig. 3 [124]Open in a new tab Cell cytotoxicity assay and cellular Au content determination with Au10, Au17, and auranofin. (A–C) Cell viability of Au10, Au17, and auranofin in three cell lines (Huh7, HepG2, and 293T) was assessed. (D) Au uptake by CRKP cells following 4 h of treatment with 20 μM Au10, Au17, or auranofin. Whole-cell Au content was measured. (E) Time-dependent Au uptake in CRKP cells treated with 20 μM Au10 or Au17 for 1–4 h. Data are presented as mean ± SD. Differences between the auranofin and the Au10 or Au17 groups were assessed with t-tests. ****p < 0.0001 indicates highly significant differences. 3.4. Molecular insights into Au10 and Au17: dual targeting of TrxR and GR TrxR participates in the reduction of the disulfide bonds in the active site of Trx in the presence of NADPH, thereby facilitating the reduction of intracellular thiols to maintain cellular redox homeostasis [[125]10,[126]42]. Auranofin and other gold complexes are known to inhibit TrxR activity through interactions with the S–Se active site [[127][43], [128][44], [129][45]]. This inhibition can lead to uncontrolled oxidative stress, ultimately resulting in cell death [[130][46], [131][47], [132][48]]. To further assess whether the disruption of intracellular redox balance may serve as an antimicrobial mechanism for Au10 and Au17, we conducted inhibitory experiments using the Trx/TrxR system. Firstly, we assessed the inhibitory effects of metal complexes (Au10, Au17, and auranofin) at different concentrations on TrxR enzyme using the TrxR assay kit. The results indicated that all complexes exhibited inhibitory effects on TrxR, with Au10 (IC[50] = 7.943 μM) and Au17 (IC[50] = 1.082 μM) demonstrating stronger inhibition of TrxR compared to auranofin (IC[50] = 11.62 μM) ([133]Fig. S7). To further investigated the inhibitory effects of metal complexes on CRKP TrxR and identified its binding sites, we expressed CRKP TrxR recombinantly in E. coli and then purified the enzyme through nickel affinity chromatography, followed by gel filtration chromatography. We conducted an analysis of the resulting fractions with reducing SDS-PAGE, which showed the presence of CRKP TrxR at a low molecular weight of 38 kDa in the gel ([134]Fig. 4A). The in vitro inhibition assays were conducted using the DTNB-reducing activity assay, which generates colored 2-nitro-5-thiobenzoate (λ[max]: 412 nm) upon reduction by TrxR. The inhibition of TrxR was assessed by monitoring the rate of DTNB reduction in the presence of auranofin, Au10, or Au17. The results showed that auranofin (IC[50] = 27.18 μM), Au10 (IC[50] = 22.6 μM) and Au17 (IC[50] = 16.5 μM) exhibited inhibitory activity against TrxR ([135]Fig. 4B). Next, we determined the inhibition constants (Ki) for the gold complexes (Au10 and Au17) using a competitive inhibition model. Both complexes demonstrated competitive inhibition against TrxR, with Au10 (Ki = 13 μM) showing greater potency than Au17 (Ki = 15.44 μM) ([136]Fig. S8A–B). Notably, Au10 and Au17 showed slightly higher inhibitory activity towards TrxR compared to auranofin. We then validated the irreversible inhibitory effects of Au10 and Au17 on the recombinant TrxR by assessing the activities of both the desalted and unsalted enzymes ([137]Fig. 4C). Additionally, a dialysis experiment was conducted to confirm the irreversibility of inhibition, showing that the enzyme activity did not recover after dialysis, which indicates that the inhibition by these complexes is indeed irreversible ([138]Fig. S8C). Fig. 4. [139]Fig. 4 [140]Open in a new tab Inhibition of recombinant CRKP TrxR enzyme by Au10 and Au17. (A). SDS-PAGE analysis of recombinant CRKP TrxR expressed in E. coli. (B). Inhibition of recombinant CRKP TrxR by Au10, Au17, and auranofin, respectively. (C). Assessment of irreversible inhibition of TrxR by Au10 and Au17. (D). NADPH-reduced CRKP TrxR is susceptible to Au10 and Au17 treatments. (E). Preincubation of GSH separately with Au10 and Au17 abolishes their inhibitory effects on CRKP TrxR. “Con”, control without gold complexes. Data were presented as Mean ± SD (n = 3, unless otherwise stated). Differences between the control group and the Au10 or Au17 groups were assessed with t-tests. The statistical significance of the differences in mean values is indicated as follows: n.s. > 0.05, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. To further explore the covalent binding sites of Au10 and Au17 with TrxR, NADPH was used to pre-reduce TrxR, leading to the reduction of the redox-active disulfide bond, after which the enzyme was incubated with Au10 and Au17 to assess its activity. The results indicated that NADPH-reduced TrxR is sensitive to treatment with Au10 and Au17, suggesting that these complexes may bind to the redox-active motif ([141]Fig. 4D). In contrast, pre-oxidation of TrxR with low concentrations of iodine did not affect the binding affinity of Au10 and Au17 ([142]Fig. S8D), implying that the interaction is likely dependent on the enzyme's redox state. To further validate this hypothesis, GSH was preincubated separately with Au10 and Au17, followed by the incubation of TrxR with GSH conjugated Au10 and Au17 to measure TrxR activity. The results demonstrated that preincubation of GSH with Au10 and Au17 eliminated their inhibitory effects on recombinant TrxR ([143]Fig. 4E). These findings indicated that Au10 and Au17 possess a capacity for sulfhydryl modification, which leads to their irreversible inhibition of bacterial TrxR activity through targeting the redox-active motif. Based on the aforementioned results, we found that auranofin, as well as the structure modified complexes (Au10, Au17), are not effective inhibitors of TrxR. The Grx system and the Trx and Grx system, two disulfide reductase system, working in coordination to perform antioxidant functions and playing critical roles in various cell activities [[144]49,[145]50]. Importantly, the presence of the GSH-GR system in Gram-negative bacteria provides a strong backup for the Trx system [[146]47]. Therefore, it is worth further investigating whether gold complexes (Au10, Au17, and auranofin) have inhibitory effects on GR activity. Firstly, we assessed the inhibitory effects of metal complexes (Au10, Au17, and auranofin) at different concentrations on GR enzyme using the GR assay kit. The results indicated that both Au10 (IC[50] = 6.401 μM) and Au17 (IC[50] = 2.258 μM) exhibited strong inhibitory effects on GR, whereas auranofin did not show any inhibitory effects at the same concentration ([147]Fig. S9). Then, we expressed CRKP GR recombinantly in E. coli and then purified the enzyme through nickel affinity chromatography, followed by gel filtration chromatography. Trx system, two disulfide reductase system, working in coordination to perform antioxidant functions and playing critical roles in various cell activities [[148]49,[149]50]. Importantly, the presence of molecular weight of 45 kDa in the gel ([150]Fig. 5A). GR catalyzes the reduction of GSSG to regenerate GSH using NADPH, while simultaneously oxidizing NADPH to produce NADP^+. The dehydrogenation rate of NADPH can be determined by measuring the rate of decrease in absorbance at 340 nm, allowing for the calculation of GR activity. The GR inhibition was estimated by following the rate of NADPH reduction in the presence of auranofin, Au10 or Au17. The results showed that Au10 and Au17 can rapidly inhibit the GR in a dose-dependent manner, with IC[50] values of 11.62 μM of Au10 and IC[50]values of 5.387 μM of Au17. In contrast, auranofin at a concentration of 40 μM did not exhibit inhibitory effects on GR ([151]Fig. 5B). Next, we determined the Ki for the gold complexes (Au10 and Au17) using a competitive inhibition model. Both complexes demonstrated competitive inhibition against GR, with Au17 (Ki = 0.746 μM) showing greater potency than Au10 (Ki = 1.503 μM) ([152]Fig. S10A–B). Subsequently, we further confirmed the irreversible inhibitory effects of Au10 and Au17 on the recombinant GR by measuring the activities of the desalted and undesalted enzymes ([153]Fig. 5C). Furthermore, a dialysis experiment was performed to verify the irreversibility of the inhibition. The results demonstrated that enzyme activity did not recover post-dialysis, confirming that the inhibition by these complexes is truly irreversible ([154]Fig. S10C). Fig. 5. [155]Fig. 5 [156]Open in a new tab Inhibition of recombinant CRKP GR enzyme by Au10 and Au17. (A). SDS-PAGE analysis of recombinant CRKP GR expressed in E. coli. (B). Inhibition of recombinant CRKP GR by Au10, Au17, and auranofin, respectively. (C). Irreversible inhibition of recombinant CRKP GR by Au10 and Au17, separately. (D). NADPH-reduced CRKP GR is susceptible to Au10 and Au17 treatments. (E). Preincubation of GSH separately with Au10 and Au17 abolishes their inhibitory effects on CRKP GR. “Con”, control without gold complexes. Data were presented as Mean ± SD (n = 3, unless otherwise stated). Differences between the control group and the Au10 or Au17 groups were assessed with t-tests. The statistical significance of the differences in mean values is indicated as follows: n.s. > 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. To further investigate the covalent binding sites between gold complexes and GR, NADPH was used to pre-reduce GR, leading to the reduction of the redox-active disulfide bond, after which the enzyme was incubated with Au10 and Au17 to assess its activity. The results indicated that NADPH-reduced GR is sensitive to treatment with Au10 and Au17, suggesting that these complexes may bind to the redox-active motif ([157]Fig. 5D). In contrast, pre-oxidation of GR with low concentrations of iodine did not affect the binding affinity of Au10 and Au17 ([158]Fig. S10D), implying that the interaction is likely dependent on the enzyme's redox state. To further validate this hypothesis, GSH was preincubated separately with Au10 and Au17, followed by the incubation of GR with GSH conjugated Au10 and Au17 to measure GR activity. The findings revealed that the preincubation of GSH with Au10 and Au17 eliminated their inhibitory effects on recombinant GR ([159]Fig. 5E). These results indicate that Au10 and Au17 possess a strong thiol-modifying effect, and that they irreversibly inhibit GR activity by targeting the redox-active motif. In contrast, auranofin do not exhibit inhibitory activity against GR, which may be a primary reason for their poor antibacterial activity against CRKP. 3.5. Au10 and Au17 compromise the bacterial defense against oxidative stress The Trx and Grx systems regulate cellular redox by scavenging harmful reactive oxygen species (ROS) within the cell [[160]51]. We proposed that inhibiting the functions of TrxR and GR would lead to increased ROS production and disrupt cellular redox homeostasis. To test this notion, we measured the production of intracellular superoxide anions using the DHE probe. The results indicated that Au10 and Au17 at concentrations of 10, 20, and 40 μM significantly elevated intracellular ROS levels compared to the control group (p < 0.001), while auranofin did not induce ROS generation in CRKP ([161]Fig. 6A–D). Given that 2-OH E^+ is a specific product of DHE oxidation by superoxide, we further quantified 2-OH E^+ levels using HPLC to confirm superoxide generation. Consistent with the ROS measurements, both Au10 and Au17 are capable of generating 2-OH E^+, whereas no detectable 2-OH E^+ was observed in the auranofin or control groups ([162]Fig. S11). These results clearly indicate that Au10 and Au17 promote superoxide production. This effect may be explained by the differential inhibition profiles of the gold complexes: while auranofin selectively inhibits TrxR, it does not affect GR activity, allowing the Grx system to compensate and maintain redox homeostasis. In contrast, Au10 and Au17 may interfere with both redox systems, leading to an accumulation of ROS. Fig. 6. [163]Fig. 6 [164]Open in a new tab Effect of Au10 and Au17 on the accumulation of superoxide radicals. The population heterogeneity of CRKP treated with varying concentrations (0, 10, 20, 40 μM) of Au10 (A), Au17 (B) and auranofin (C) was evaluated using single parameter flow cytometry analysis for oxidative stress. Additionally, panel (D) illustrates the total accumulation of superoxide radicals in CRKP treated with the different concentrations of Au10, Au17, and auranofin. (E). Effects of different concentrations of Au10 on intracellular oxidative stress indicators, including MDA, SOD, and GSH. (F). Effects of different concentrations of Au17 on intracellular oxidative stress indicators, including MDA, SOD, and GSH. Data were presented as Mean ± SD. Differences between the control group and the Au10 or Au17 groups were assessed with t-tests. The statistical significance of the differences in mean values is indicated as follows: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Subsequently, we further measured the levels of oxidative stress biomarkers in Au10 and Au17 treated cells, including MDA, SOD and GSH. At concentrations of 10, 20, 40, and 80 μM, Au10 and Au17 significantly raised MDA levels and markedly reduced SOD activity compared to the control group. What's more, both Au10 and Au17 at concentrations of 10, 20, and 40 μM significantly reduced GSH levels in comparison to the control group ([165]Fig. 6E–F). These results suggest that the inhibition of TrxR and GR mediated by Au10 and Au17 disrupts the redox balance in bacteria. The exposure of cells to ROS is likely to interfere with a wide range of cellular components, including lipids, membranes and cytoplasmic protein, leading to alterations in cell membrane permeability and accelerating bacterial death [[166]52,[167]53]. PI probe cannot penetrate the cell membranes of normal viable cells, but it can pass through damaged cell membranes and bind to nucleic acids, resulting in staining. We utilized this property of the probe to assess the integrity of bacterial cell membranes. Flow cytometric analysis revealed that with the increase in Au10 and Au17 concentration, the red fluorescence intensity within bacterial cells significantly increased compare to the control group ([168]Fig. 7A–B, D). In contrast, the red fluorescence intensity of bacterial cells treated with auranofin did not show a significant difference when compared to the control group ([169]Fig. 7C–D). Subsequently, we further assessed the loss of intracellular contents in bacteria. The results showed that as the concentrations of Au10 and Au17 increased, the levels of intracellular Mg^2+ and K^+ gradually decreased ([170]Fig. 7E–F), indicating that increased cell membrane permeability leads to the leakage of intracellular contents. Fig. 7. [171]Fig. 7 [172]Open in a new tab Effect of Au10 and Au17 on membrane integrity. The population heterogeneity of CRKP treated with varying concentrations (0, 10, 20, 40 μM) of Au10 (A), Au17 (B) and auranofin (C) was evaluated using single parameter flow cytometry analysis for membrane integrity. (D) The intensity of red fluorescence in CRAB treated with different concentrations (0, 10, 20, 40 μM) of Au10, Au17, and auranofin. (E). Determination of intracellular Mg^2+ and K^+ under treatment with different concentrations of Au10. (F). Determination of intracellular Mg^2+ and K^+ under treatment with different concentrations of Au17. Data were presented as Mean ± SD. Differences between the control group and the Au10 or Au17 groups were assessed with t-tests. The statistical significance of the differences in mean values is indicated as follows: ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. 3.6. Exploring other targets: degradation of bacterial DNA and inhibition of energy metabolism Beyond disrupting the bacterial redox balance, gold complexes can penetrate bacterial membranes to inhibit other intracellular targets, such as DNA intercalators [[173][54], [174][55], [175][56]]. The results from the Au uptake experiments indicate that Au10 and Au17 can significantly penetrate the cell membrane and enter the cells compared to auranofin. It is plausible that they might internalize within the cells to target intracellular entities, such as DNA. To further validate this hypothesis, we purified the DNA from CRKP and co-incubated it with the gold complexes (Au10, Au17, and auranofin) for 30 min. The gel electrophoresis results showed a clear induction of DNA degradation following treatment with various concentrations of Au10 and Au17, while auranofin did not demonstrate DNA degradation effect ([176]Fig. 8A). Fig. 8. [177]Fig. 8 [178]Open in a new tab DNA-targeting activity assays and proteomics analysis of CRKP treated by Au10 and Au17 at low concentration. (A). Gel electrophoresis showed the degradation effect of Au10 and Au17, and auranofin on DNA in vitro. (B). Circular dichroism analysis of the conformational changes in DNA when preincubated separately with Au10 and Au17 (40 μM) for 30 min. (C). After co-incubation of hydrogen peroxide and metal complexes with DNA for 30 min, gel electrophoresis showed their combined effects on DNA degradation. (D)–(E). KEGG pathway enrichment analysis between the control group and the Au10 or Au17 group. (proteins were considered upregulated when FC > 1.5 and P-value < 0.05, and downregulated when FC < 0.67 and P-value < 0.05). (F)–(G). Measurement of intracellular ATP levels under different concentrations of Au10 and Au17. Data were presented as Mean ± SD (n = 3, unless otherwise stated). Differences between the control group and the Au10 or Au17 groups were assessed with t-tests. The statistical significance of the differences in mean values is indicated as follows: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Subsequently, we performed circular dichroism spectroscopy to assess the changes in absorption wavelength of DNA under Au10 and Au17 treatment. The results indicated that the treated group exhibited significant changes in absorption wavelength compared to the normal control group, which confirm that Au10 and Au17 have a destructive effect on DNA ([179]Fig. 8B). To confirm DNA damage within bacterial cells, we conducted TUNEL staining, and the results revealed a clear correlation between increasing drug concentration and enhanced fluorescence intensity ([180]Fig. S12). This finding indicates a significant increase in DNA damage as the concentration of the gold complexes rises. We also explored whether Au10 and Au17 would accelerate DNA degradation under oxidative stress conditions. We used hydrogen peroxide to simulate oxidative stress condition and assessed the effects of Au10 and Au17 on DNA degradation. The results of gel electrophoresis confirmed that the degradation of DNA by gold complexes was significantly accelerated under conditions of oxidative stress induced by hydrogen peroxide ([181]Fig. 8C). To explore the mechanisms of antimicrobial action of drugs more comprehensively, we subsequently investigated potential additional antimicrobial targets in bacteria under the action of low concentrations of Au10 and Au17 through proteomics. The study was conducted using label-free quantitative proteomics technology. In the Au10 vs. control group, a total of 3240 quantitative proteins were identified, among which 75 proteins were upregulated and 113 proteins were downregulated. In the Au17 vs. control group, a total of 3248 quantitative proteins were identified, with 84 proteins upregulated and 149 proteins downregulated (proteins were considered upregulated when FC > 1.5 and P-value < 0.05, and downregulated when FC < 0.67 and P-value < 0.05). Principal component analysis (PCA) was performed to obtain the protein expression profiles for the normal control group and the treatment groups with Au10 and Au17. A clear separation in intensity was observed between the control and treatment groups along both the X-axis and Y-axis ([182]Fig. S13). Through KEGG pathway enrichment analysis, the Au10 group exhibited a significant inhibition of the Glycolysis/Gluconeogenesis pathway compared to the control group ([183]Fig. 8D). In the case of the Au17 treatment group, a marked inhibition of propanoate metabolism was observed relative to the control group ([184]Fig. 8E). Enrichment results indicated that enriched KEGG terms were significantly associated with the “carbohydrate catabolic process”, which is closely related to bacterial energy metabolism [[185]57]. Subsequently, we further confirmed that low concentrations of gold complexes (Au10 and Au17) affect intracellular energy production in bacteria by measuring the intracellular ATP levels ([186]Fig. 8F–G). The results indicated a significant decrease in intracellular ATP as the concentrations of Au10 and Au17 increased. However, the proteomic results also indicated that bacteria can develop resistance mechanisms under the influence of low concentrations of Au10 and Au17. Specifically, compared to the control group, the protein levels associated with the “cationic antimicrobial peptide resistance” pathway were significantly elevated in the Au10 or Au17-treated group ([187]Fig. S14). Bacteria can reduce the affinity of cationic molecules, such as CAMPs, to their surface through various chemical modifications of the cell surface, including changes to the lipid membrane, the peptidoglycan and teichoic acid-containing cell wall, and the capsular polysaccharides [[188]58]. Complexes Au10 and Au17 are cationic complexes. Bacteria increase the expression of proteins to the control group, the protein levels associated with the “cationic antimicrobial peptide Au10 and Au17, thereby providing protection to the bacteria. This also suggests that when designing complexes, we should consider how to incorporate ligands that disrupt such feedback mechanisms to reduce the development of resistance. Collectively, Au10 and Au17 can irreversibly inhibit both TrxR and GR activities in bacteria by targeting redox-active motifs, with a particularly strong effect on GR activity. This inhibition obstructs the roles of TrxR and GR in removing superoxide radicals, leading to the generation of oxidative stress within the bacterial cells. Superoxide radicals affect various functions essential for bacterial survival, including cellular redox balance, membrane integrity, and lipid peroxidation. Additionally, Au10 and Au17 not only directly degrade bacterial DNA but can also accelerate the degradation of DNA under oxidative stress conditions. Moreover, at low concentrations, Au10 and Au17 can inhibit the expression of proteins related to bacterial carbohydrate metabolism pathways, thereby reducing energy production. Au10 and Au17 exert potent antibacterial effects against CRKP through multiple-target pathways. 3.7. In vivo antimicrobial activities of Au10 and Au17 in CRKP-infected models Infectious wounds caused by Gram-negative bacteria pose a significant challenge, and controlling their proliferation and invasiveness is crucial. The primary pathogens associated with diabetic wounds, necrotizing fasciitis, and other chronic wounds are Gram-negative bacteria [[189]59,[190]60]. Gram-negative bacteria particularly resistant strains of E. coli, K. pneumoniae and A. baumanii are observed in several studies [[191]61]. Recognizing that MDR Gram-negative bacteria pose a significant threat to the management of these infections [[192]62], we established a mouse excision wound model infected with clinical isolated CRKP (The MIC of Au10 and Au17 against the clinical isolate is 10 μM ([193]Fig. S15) to assess the efficacy of Au10 and Au17. In this infection model, we artificially created a severe wound infection by making an incision of approximately 1.2 × 1.2 cm on the backs of mice and then injecting 100 μL of a bacterial suspension at a concentration of 10^9 CFU/mL into the wound site. After 24 h of infection, the mice were randomly divided into six groups. The wound area underwent treatment with 0.9 % NaCl; H[2]O[2]; auranofin; colistin B; Au10 and Au17 over 14 days. As shown in [194]Fig. 9A, both the Au10 and Au17 groups showed significant outcomes compared with the control, H[2]O[2], auranofin and colistin B group. The Au10 and Au17 treatment facilitated wound healing with a notably improved recovery of the wound area ([195]Fig. 9B). The absence of toxicity during the Au10 and Au17 treatment was confirmed by the changes in mouse body weight ([196]Fig. 9C). As shown in [197]Fig. 9D, hematoxylin and eosin (HE) staining and immunohistochemical analyses (IL-6 and TNF-α) indicate that the Au10 and Au17 treatment groups exhibited a significant reduction in skin tissue inflammation compared to the control, H[2]O[2], auranofin and colistin B group, along with a marked decrease in the expression of inflammatory factors. Additionally, we confirmed through tissue immunofluorescence experiments that the Au10 and Au17 treatment groups exhibited a significant reduction in the infiltration of inflammatory cells (macrophages and T cells) compared to the control group, leading to a decrease in local tissue inflammation ([198]Fig. 9E–F). Collectively, the in vivo data suggested that the in vivo antibacterial activity of Au10 and Au17 is superior to that of auranofin and colistin B, significantly reducing local tissue infection and promoting wound healing. Fig. 9. [199]Fig. 9 [200]Open in a new tab Au10 and Au17 show antimicrobial activities in a CRKP skin infection model and prolong the survival time of mice with CRKP-induced peritoneal infection. (A). Photography of wound infection status in different groups from Day 1 to Day14. (B).Recording the wound size changes of mice in each group from Day 1 to Day 14. (C). Recording the body weight changes of mice in each group from Day 1 to Day 14. (D). HE staining and immunohistochemistry of inflammatory markers (IL-6 and TNF-α) in infected wounds from each group on Day 14 (scale bar: 100 μm). (E). Immunofluorescence analysis of skin tissue for CD3 (green) and CD4 (red) markers on Day 14 in the control, Au10 and Au17 group (scale bar: 100 μm). (F). Immunofluorescence analysis of skin tissue for CD206 (green) and CD86 (red) markers on Day 14 in the control, Au10 and Au17 group (scale bar: 100 μm). (G). To establish a model of abdominal infection, BALB/c mice were injected intraperitoneally with a bacterial suspension. Treatment was administered 1 h after infection, and survival was monitored over a 48-h period. Data were presented as Mean ± SD. Statistical differences between two groups were analyzed by t-test. The statistical significance of the differences in mean values is indicated as follows: ∗∗∗p < 0.001. Subsequently, we established a peritoneal infection model to evaluate the antibacterial activity of Au10 and Au17. We infected mice by intraperitoneally injecting 200 μL of 10^9 CFU/mL bacterial suspension. After 1 h, the mice were randomly divided into 4 groups, with six mice per group, and treated with saline, colistin B, Au10 and Au17, respectively. The survival of the mice was observed and recorded over a 48-h period. The results showed that no mice in the control group or the colistin B group survived beyond 24 h. In contrast, three mice in both the Au10 and Au17 treatment groups survived for more than 24 h, with one mouse in the Au10 treatment group surviving beyond 48 h, and two mice in the Au17 treatment group surviving beyond 48 h ([201]Fig. 9G). In addition, we measured the Au levels of Au10 and Au17 in the blood at 24 h post-treatment to assess pharmacokinetics and potential systemic exposure. The results showed that the serum Au concentrations for both complexes followed a similar trend, with an initial peak at 2 h (Au10 reaching 2.32 μg/mL and Au17 reaching 1.77 μg/mL). Importantly, both complexes maintained blood concentrations above 1 μg/mL at 24 h ([202]Fig. S16). These results indicate that both Au10 and Au17 can prolong the survival time of mice with severe peritoneal infections and demonstrate therapeutic advantages against carbapenem- and colistin-resistant Gram-negative bacteria. 4. Conclusion We demonstrated for the first time that Au(III) Schiff base complexes exhibit strong antibacterial activity against carbapenem- and colistin-resistant Gram-negative bacteria as oxidoreductase inhibitors. In vitro experiments, Au10 and Au17, representatives of our superior Au(III) Schiff base complexes, possess significantly enhanced antibacterial activity but much decreased cytotoxicity in contrast to auranofin. Moreover, Au10 and Au17 exert their antibacterial advantages through multiple-target pathways, including the irreversible inhibition of TrxR and GR activities in bacteria by targeting redox-active motifs, which abolishes the capacity of TrxR and GR to quench ROS, ultimately leading to oxidative stress, degrading DNA, and inhibiting bacterial energy metabolism under low drug concentration conditions. Importantly, in vivo experiments using mouse models infected with CRKP demonstrated the outstanding bactericidal effects of Au10 and Au17, suggesting a significant advancement of these metal complexes in combating carbapenem- and colistin-resistant bacterial infections. This study provides the first evidence that Au(III) Schiff base complexes can simultaneously target the redox system of carbapenem- and colistin-resistant Gram-negative bacteria, establishing a new paradigm in antibacterial strategies and guiding future innovations in antibacterial therapy. Future work will aim to enhance the antibacterial effects by increasing the affinity of TrxR and GR enzymes in Gram-negative bacteria, thus highlighting the transformative potential of metal complexes in addressing the urgent challenge of antibiotic resistance. CRediT authorship contribution statement Xiuli Chen: Conceptualization, Data curation, Formal analysis, Funding acquisition, Writing – original draft, Writing – review & editing. Lin Lv: Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Jianqiang Xu: Conceptualization, Data curation, Writing – original draft, Writing – review & editing. Jing Shi: Data curation, Formal analysis, Writing – original draft. Xi Chen: Data curation, Formal analysis, Writing – original draft. Guisha Zi: Data curation, Formal analysis, Writing – original draft. Yuxuan Wu: Conceptualization, Data curation, Writing – original draft. Shibo Sun: Data curation, Formal analysis, Writing – original draft. Yufan Pang: Conceptualization, Data curation, Writing – original draft. Qian Song: Data curation, Formal analysis, Writing – original draft. Ling Ma: Data curation, Formal analysis. Shuang Wei: Conceptualization, Data curation, Formal analysis, Funding acquisition, Writing – original draft, Writing – review & editing. Tonghui Ma: Conceptualization, Data curation, Formal analysis, Funding acquisition, Writing – original draft, Writing – review & editing. Wukun Liu: Conceptualization, Data curation, Formal analysis, Funding acquisition, Writing – original draft, Writing – review & editing. Data and materials availability All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments