Synergistic activity of colistin in combination with resveratrol and capsaicin against mcr-1-posotive Escherichia coli

2.1 Bacteria and chemicals

The specific information of strains used in this study could be found in Table S1. All mcr-1-positive Escherichia coli were confirmed by PCR analysis of 16S rRNA and mcr-1. Colistin (99 %, purity), resveratrol (99 %, purity), capsaicin (98 %, purity) and other chemical reagents were purchased from Mackin (Shanghai, China), Aladdin (Shanghai, China) or Solarbio Technology (Beijing, China).

2.2 Antimicrobial susceptibility testing

The minimum inhibitory concentrations (MICs) of colistin, resveratrol and capsaicin against mcr-1-positive Escherichia coli isolates were determined by the micro-dilution method, according to the CLSI 2023 guideline (Rai et al., 2023). The Escherichia coli ATCC 25922 was used as a quality control strain. All tests were performed in triplicate for all strains.

2.3 Checkerboard assays

Synergistic activity of colistin and resveratrol, capsaicin was evaluated by checkerboard assays according to the previous study (Yi et al., 2022). The specific details of experiment could be found in supplementary materials S1.

2.4 Time-killing curves

To further explore the synergistic antimicrobial activity of colistin and antibiotic adjuvants, clinical strain mcr-1-positive Escherichia coli (HP-144) was selected to construct time-killing curves. The specific details of experiment could be found in supplementary materials S2. Compared with the colistin group, synergy was determined if an over 2 log10 reduction of CFU/mL was observed in the drug combination group.

2.5 The determination of biochemical index

The pretreatment of biochemical index assays were conducted as the following protocols. Escherichia coli (HP-144) was incubated to exponential grow phase. Then, bacterial suspensions were centrifuged and re-suspended in buffer with an OD₆₀₀ of 0.5. The fluorescent dye was added according to the instructions of different kits. After 30 min incubation at 37 °C, then the mixed cultures were treated with different groups of drugs. After being incubated at 37 °C for 1 h, 200 µL cultures were added to the 96-well plate and the microplate reader was used to evaluate biochemical indexes. The specific excitation wavelength and emission wavelength used in this study were listed in supplementary materials.

2.6 Scanning electron microscopy (SEM)

Escherichia coli HP-144 was grown to the exponential phase, then bacteria were washed, and suspended in LB with OD₆₀₀ of 0.5. Bacterial suspension incubated with drugs for 24 h at 37 °C. Cultures were fixed with 2.5 % glutaraldehyde at 4 °C for 12 h. The bacteria were dehydrated with gradient ethanol (30 %, 50 %, 70 %, 90 %, and 100 %) and precipitates were suspended with acetone at 4 °C and dried at the critical point as previous described (Koh et al., 2022), followed by the observation under the SEM.

2.7 Transcriptomic analysis

The Escherichia coli HP-144 in early-exponential phase was treated with colistin (2 μg/mL) alone or in combination with resveratrol (128 μg/mL) and capsaicin (256 μg/mL) for 4 h. Total RNA was extracted by TRIzol Reagent (Invitrogen) according to instructions. The transcriptomic sequencing was performed as previous described, the specific details could be found in supplementary materials S9.

2.8 Metabolome analysis

Escherichia coli HP-144 was grown to the exponential phase, then the bacterial cultures were incubated with colistin (2 μg/mL) alone or in a combination of resveratrol (128 μg/mL) and capsaicin (256 μg/mL) for 4 h. A mass spectrometry (MS)-based metabolomics was performed, the specific details could be found in supplementary materials S10.

2.9 Mouse peritonitis-sepsis infection model

Four-week-old ICR female mice (approximately 20 g) were randomly divided into seven groups (n = 12): Control, CAP (2 mg/kg), RES (50 mg/kg), COL (10 mg/kg), COL+RES (10 mg/kg + 50 mg/kg), COL+CAP (10 mg/kg + 2 mg/kg), COL+RES+CAP (10 mg/kg + 50 mg/kg + 2 mg/kg). ICR female mice were treated with 5.0 × 10⁶ CFUs HP-144 via intraperitoneal injection. Mice were treated with different groups of drugs 2 h after injection. After 48 h of infection, mice were euthanized and the spleen, liver, and kidney were isolated, weighed, homogenized, and serially diluted in PBS. Samples were ten-fold serially diluted and incubated on LB agar plates at 37 °C for 18 h. Bacterial colonies were counted and the primary CFUs/mL were calculated. Moreover, part of the tissues (spleen, liver, and kidney) were subjected to histological analysis.

2.10 Molecular docking

To predict the possible interaction between resveratrol, capsaicin, and MCR-1, molecular docking was performed as previously describled (Lee et al., 2022), the specific details could be found in supplementary materials S11.

2.11 Statistical analyses

Statistical analysis was performed using GraphPad Prism 8 8. All datas were presented as mean ± SD. Statistical assessments were performed using unpaired 2-tailed t-tests, one-way ANOVA among multiple groups. Significance levels are indicated with asterisks: *P<0.05, **P<0.01, ***P<0.001.

3.1 Synergistic activity of colistin in combination with resveratrol and capsaicin against mcr-1-positive Escherichia coli

Resveratrol or capsaicin was tested against mcr-1-positive Escherichia coli. As shown in Table 1, the MIC values of 1024 or 2048 μg/mL for resveratrol or capsaicin; sub-inhibitory concentration showed that capsaicin or resveratrol alone had no significant antibacterial activity against mcr-1-positive Escherichia coli. Resveratrol at a subinhibitory concentration (1/4 MIC, 1/8 MIC, 1/16 MIC) combined with colistin enhanced the activity of colistin, and the MIC fold change of colistin ranges from 4 to 512. Capsaicin at sub-inhibitory concentration (1/4 MIC, 1/8 MIC, 1/16 MIC) combined with colistin enhanced the activity of colistin, and the MIC fold change of colistin ranges from 64 to 2048.

The synergistic activity of colistin in combination with antibiotic adjuvants was evaluated by checkboard assays. As shown in Table 2, the FICI values for resveratrol combined with colistin were from 0.09375 to 0.5. The FICI values for capsaicin combined with colistin were from 0.078125 to 0.15625. This demonstrated that resveratrol and capsaicin had synergistic effects on mcr-1-positive Escherichia coli when combined with colistin. To further assess the reversal ability of resveratrol and capsaicin in combination with colistin. Antibacterial tests were carried out with 1/16 MIC RES+1/4MIC CAP, 1/16 MIC RES+1/8 MIC CAP, and 1/16 MIC RES+1/4 MIC CAP. The MICs of colistin decreased to ≤ 0.00003 μg/mL, and the colistin MIC fold change ≥ 65536. The combination therapy completely reversed the colistin resistance of all mcr-1-positive Escherichia coil.

3.2 Time–kill results for the colistin in combination with resveratrol and capsaicin against mcr-1-positive Escherichia coli

Subsequently, the time-kill experiment was carried out on the selected clinical mcr-1-positive Escherichia coli HP-144 isolated from fecal sample during the exponential growth phase with drugs. As shown in Fig. 1, resveratrol, capsaicin, and colistin alone did not kill exponentially growing mcr-1-positive Escherichia coli HP-144. Colistin in combination with resveratrol and capsaicin completely killed the bacteria throughout 12–24 h. At 24 h, compared with the colistin alone the bacterial concentration reduced by 11.519 log₁₀ CFU/mL. These results illustrated that the combination of colistin, resveratrol and capsaicin exhibited synergistically to mcr-1-positive Escherichia coli.

Close

Fig. 1

Time-kill curves of the Escherichia coli HP-144 in the presence of resveratrol (128 µg/mL), capsaicin (256 µg/mL) or colistin (2 µg/mL) alone or in combination for 24 h.

Export to PPT

3.3 Morphological changes of mcr-1-positive Escherichia coli after treatment of colistin in combination with resveratrol and capsaicin

Colistin in combination with resveratrol and capsaicin exhibited significant synergistic antibacterial effect. Therefore, we speculate that the combination of colistin, resveratrol and capsaicin treatment could promote the destruction of the OM. In order to confirm our hypothesis, SEM was utilized to observe the morphological alterations of mcr-1-positive Escherichia coli HP-144 after different treatment. As shown in Fig. 2A, the membrane damage and morphological shrinkage caused by combined therapy were more severe compared with mono-treatment. These findings suggested that the colistin combined with resveratrol and capsaicin combination might enhance ability to damage membrane structures.

Close

Fig. 2

Effects of combined therapy on bacterial membrane. A, Morphological changes of Escherichia coli HP-144 treated with colistin (2 µg/mL), resveratrol (128 µg/mL), capsaicin (256 µg/mL) alone or their combination. Scar bar, 2.00 µm. Destroyed outer membrane was marked by red arrows. The experiment is grouped as follows: (a) Control, (b) CAP, (c) RES, (d) COL, (e) CAP+COL, (f) RES+COL, (g) RES+CAP+COL. B, Outer membrane permeability was evaluated by measuring the fluorescence intensity of NPN. C, Inner membrane permeability was evaluated by measuring the fluorescence intensity of PI.

Export to PPT

3.4 Colistin in combination with resveratrol and capsaicin disrupts the bacterial membrane

To visually show the effect of the combined therapy on cell membrane permeability, the OM and IM permeability were measured by NPN and PI, respectively. After the colistin in combination with resveratrol and capsaicin treatment, the fluorescence intensity of NPN increased significantly (Fig. 2B), indicating that the combination observably improved the permeability of the OM which was consistent with the electron microscopy results. In addition, there was no significant change in the permeability of the IM, indicating the integrity of the IM structure was maintained (Fig. 2C). Collectively, the findings demonstrated that capsaicin and resveratrol improved colistin's capacity to damage the OM.

3.5 Colistin in combination with resveratrol and capsaicin dissipates the proton motive force

PMF (proton motive force), comprises two components: electric potential (ΔΨ) and transmembrane proton gradient (ΔpH). Due to the two components compensating for each other, the PMF remains stable in the absence of excessive external disturbances. Specifically, to equilibrate the homeostasis of PMF, the ΔpH could increase when the ΔΨ dissipated and vice versa. In fact, the increase of fluorescence intensity of DiSC3(5) suggests the dissipation of ΔΨ, while a decrease otherwise implies the disruption of ΔpH. The results showed that the fluorescence intensity of DiSC3(5) was reduced after treatment (Fig. 3A), which suggested that the drug effect may have affected the change in ΔpH. Therefore, the fluorescence intensity of the fluorescent probe BCECF AM were measured, and we found that the group with the addition of resveratrol had a significant decrease in intracellular pH (Fig. 3B), decreasing ΔpH. Resveratrol was shown to be responsible for the dissipation of ΔpH. The disturbance of PMF by colistin in combination with resveratrol and capsaicin mainly caused the changes of intracellular pH, which disrupted PMF and might affect the related ATP synthesis.

Close

Fig. 3

Effects of combined therapy on bacterial energy metabolism and efflux pump activity in Escherichia coli HP-144. A, Intracellular ΔΨ, B, Intracellular pH changes, C, Intracellular ATP levels, D, The activity of efflux pump.

Export to PPT

3.6 Colistin in combination with resveratrol and capsaicin inhibits intracellular ATP synthesis of mcr-1-positive Escherichia coli

PMF, as an energy pathway on the cell membrane, can regulate ATP synthesis. For this reason, we further examined the changes of intracellular ATP. As expected, ATP was reduced after drug treatment, while the intracellular ATP level was markedly reduced in mcr-1-positive Escherichia coli HP-144 treated with the combination (Fig. 3C). Thus, we speculated that the inhibition of ATP production affects the life-related processes of bacteria, which may be related with the enhanced antimicrobial ability of colistin.

3.7 Colistin in combination with resveratrol and capsaicin inhibits efflux pumps activity

In general, the efflux pump plays a vital factor in antimicrobial resistance. As an inhibitor of efflux pump, capsaicin can inhibit the efflux pump activity and ensure the retention of the drug in the cell. In this experiment, it was envisaged that capsaicin could inhibit the efflux pump and increase the intracellular concentration of resveratrol, thereby strengthening the antimicrobial effect of colistin. To prove our assumptions, we measured the activity of the efflux pump, the increase in fluorescence indicates the accumulation of Etbr in the cell, and we found that capsaicin supplementation could increase intracellular accumulation of Etbr (Fig. 3D), indicating that the efflux pump activity was inhibited. These findings demonstrated that colistin in combination with resveratrol and capsaicin could effectively inhibit the efflux pumps activity in mcr-1-positive Escherichia coli.

3.8 Changes in intracellular oxidation levels after the different treatments

Reactive oxygen species are widely considered to be an essential part of antimicrobial activity. However, controversy remains as to whether reactive oxygen species are the decisive factor for sterilization. It has been reported that bacterial death is not associated with the production of reactive oxygen species. Resveratrol and capsaicin act as exogenous reducing agents that can remove oxidizing substances. Based on the results, we found that the adding resveratrol and capsaicin could reduce the level of ROS. Subsequently, in order to detect the changes in intracellular antioxidant levels, we then measured the main intracellular oxidoreductase SOD enzyme, and the activity of SOD enzyme markedly decreased in the mcr-1-positive Escherichia coli treated with combination of colistin, resveratrol and capsaicin (Fig. 4AB). The results showed that due to the addition of exogenous reducing agents, which were sufficient to remove reactive oxygen species, excessive production of endogenous oxidoreductase was not required. Combined with the results of in vitro synergistic activity, it was suggested that the reason for excellent antibacterial effect of resveratrol and capsaicin combined with colistin may not be related to oxidative stress. It may be related to the inhibition of cellular respiration as well as the destruction of cell membranes, but the exact reason remains to be explored.

Close

Fig. 4

Oxidative stress of Escherichia coli HP-144 after different treatments. A, ROS levels, B, SOD activity.

Export to PPT

3.9 Molecular docking

The results of the molecular docking of the MCR-1 protein with bioactive groups of resveratrol and capsaicin are displayed in Fig. 5. The binding free energy between MCR-1 and resveratrol was −5.45 kcal/mol. Resveratrol has the ability to bind to MCR-1 both hydrophobically and through hydrogen bonding. According to the binding model of resveratrol to MCR-1, resveratrol formed strong contacts through hydrogen bonding with VAL413, GLU423, and GLU418 (Fig. 5ABC). Moreover, capsaicin could also bind to MCR-1, the binding free energy between MCR-1 and capsaicin was −3.91 kcal/mol, with hydrogen bonds THR283, ASN482, and TYR287 serving as the primary means of attachment (Fig. 5DEF). In summary, the molecular docking data demonstrated that resveratrol and capsaicin have good bonding stability with the MCR-1 protein.

Close

Fig. 5

Putative pattern of interaction between resveratrol, capsaicin and MCR-1 protein. A, Stable 3D structure of resveratrol with MCR-1 obtained from the AutoDock simulations. B, The interactions formed between the amino acid residues (stick) and the docked resveratrol molecule (ball and stick) in the MCR-1 binding site. C, Interaction of planar amino acids between resveratrol and MCR-1 molecule. D, Stable 3D structure of capsaicin with MCR-1 obtained from the AutoDock simulations. E, The interactions formed between the amino acid residues (stick) and the docked capsaicin molecule (ball and stick) in the MCR-1 binding site. F, Interaction of planar amino acids between capsaicin and MCR-1 molecule.

Export to PPT

3.10 Colistin in combination with resveratrol and capsaicin down-regulates TCA cycle and oxidative phosphorylation, inhibited efflux pump, reduced nucleotide metabolism

In order to further investigate the molecular mechanisms of resveratrol and capsaicin enhancing bactericidal activity of colistin against mcr-1-positive Escherichia coli, we performed transcriptomic and metabonomic analysis of HP-144 isolate after exposure to colistin or colistin combined with resveratrol and capsaicin. The comparison of treatment with the combination and that with colistin alone revealed 1791 differentially expressed genes (DEGs), including 1080 up-regulated genes, and 711 down-regulated genes. KEGG enrichment analysis demonstrated that DEGs are mainly involved in pathways such as TCA cycle, oxidative phosphorylation, and carbon metabolism (Fig. 6). Notably, up-regulated genes involved in oxidation–reduction process, TCA cycle, oxidative phosphorylation and efflux pump system. Similarly, metabolite enrichment analysis demonstrated that differential metabolites mainly including pathways such as nucleotide metabolism, purine metabolism, pyrimidine metabolism, TCA cycle, oxidative phosphorylation, these pathways are closely related to biological life-sustaining processes (Fig. 7). The results showed that the combined therapy mainly inhibited TCA cycle, oxidative phosphorylation, nucleotide metabolism, and efflux pump. The inhibition of the TCA cycle, and oxidative phosphorylation reduced electron transfer and ATP synthesis, and the inhibition of efflux pumps reduced the efflux of drugs, the inhibition of nucleotide metabolism, likewise affecting subsequent intracellular translation and transcription. Redox-related genes indicate a reduced ability of cells to cope with oxidative damage. In summary, we concluded that combined therapy rendered the bacteria imminent death, inhibiting both energy metabolism and essential functions.

Close

Fig. 6

Transcriptomic analysis of Escherichia coli HP-144 treated by colistin, and colistin in combination with resveratrol and capsaicin. A, Volcano-plots of DEGs distribution. B, KEGG enrichment analysis of the DEGs. The x and y axis in A represent the expression changes and corresponding statistically significant degree, respectively. C, Selected DEGs involved in TCA cycle, Oxidative phosphorylation, multidrug efflux pump and oxdation-reduction process. COL, colistin alone, COL+RES+CAP, colistin in combination with resveratrol and capsaicin.

Export to PPT

Close

Fig. 7

Non-targeted metabolomics analysis of Escherichia coli HP-144 treated by colistin, and colistin in combination with resveratrol and capsaicin. A, KEGG enrichment analysis of the differential metabolites. The x and y axis in A represent the expression changes and corresponding statistically significant degree, respectively. B, Selected differential metabolites involved in TCA cycle, Oxidative phosphorylation, nucleotide metabolites. COL, colistin alone; COL+RES+CAP, colistin in combination with resveratrol and capsaicin.

Export to PPT

3.11 Resveratrol and capsaicin restores colistin activity against mcr-1-positive Escherichia coli HP-144 in vivo

As colistin, resveratrol, and capsaicin showed excellent synergistic antibacterial capacity on mcr-1-positive Escherichia coli in vitro, we speculated that colistin's therapeutic effectiveness would be regained in vivo by reversing MCR-mediated colistin resistance. The efficacy of colistin combined with resveratrol and capsaicin was examined in a mouse model of intraperitoneal infection infected with mcr-1-positive Escherichia coli. Organ histological analysis of mice showed that the pathological changes caused by bacterial infection in the three-drug group were obviously alleviated relative to the other groups (Fig. 8A). The results showed that the bacterial loads of the liver, spleen, and kidney significantly reduced in the combination of resveratrol, capsaicin and colistin treatment group relative to other groups (Fig. 8B). These results verified that capsaicin and resveratrol significantly restore colistin function in vivo.

Close

Fig. 8

Resveratrol and capsaicin restore colistin activity in mice infection models. A, Organ histological analysis of mice; B, Bacterial loads were determined in liver; C, Bacterial loads were determined in kidney; D, Bacterial loads were determined in spleen.

Export to PPT