Isoalantolactone restores the sensitivity of gram‐negative Enterobacteriaceae carrying MCR‐1 to carbapenems

Abstract Polymyxin B has been re‐applied to the clinic as the final choice for the treatment of multidrug‐resistant gram‐negative pathogenic infections, but the use of polymyxin B has been re‐assessed because of the emergence and spread of the plasmid‐mediated mcr‐1 gene. The purpose of this study was to search for an MCR inhibitor synergistically acting with polymyxin to treat the infection caused by this pathogen. In this study, we used the broth microdilution checkerboard method to evaluate the synergistic effect of isoalantolactone (IAL) and polymyxin B on mcr‐1‐positive Enterobacteriaceae. Growth curve analysis, time‐killing assays and a combined disc test were used to further verify the efficacy of the combined drug. Colonization of the thigh muscle in mice, survival experiments and lung tissue section observations was used to determine the effect of synergy in vivo after Klebsiella pneumoniae and Escherichia coli infection. We screened a natural compound, IAL, which can enhance the sensitivity of polymyxin B to mcr‐1‐positive Enterobacteriaceae. The results showed that the combined use of polymyxin B and IAL has a synergistic effect on mcr‐1‐positive Enterobacteriaceae, such as K pneumoniae and E coli, not only in vitro but also in vivo. Our results indicate that IAL is a natural compound with broad application prospects that can prolong the service life of polymyxin B and make outstanding contributions to the treatment of gram‐negative Enterobacteriaceae infections resistant to polymyxin B.

gram-negative pathogens and are regarded as the last line of defence for the treatment of such bacterial pathogens. 3,4 Additionally, among these antibiotics, polymyxin B and colistin are cationic peptide antibiotics with lipophilic acyl side chains 5 that were developed in the 1940s but fell into disfavour because of their high toxicity rates. 6 The mechanism by which polymyxin kills gram-negative pathogens relies on disrupting membrane permeability through polar and hydrophobic interactions. In these interactions, there is an electrostatic interaction between the positively charged residues of polymyxin and the negatively charged lipid A moiety of the lipopolysaccharide. 7,8 With the emergence of multidrug-and pandrug-resistant gram-negative pathogens in the 1990s, especially carbapenem-resistant Enterobacter, we have to reactivate polymyxin in the absence of new antibacterials. [9][10][11] Since the first discovery of the transferable colistin resistance gene mcr-1 in China, more than 40 countries and regions around the world have detected such resistance genes. 12 MCR-1, a protein product of the mcr-1 gene, is predicted to be an integral membrane protein with the catalytic activity of phosphoethanolamine transferases. The MCR-1 enzyme modifies the chemical structure of the lipid A moiety on bacterial LPS by the addition of phosphoethanolamine, which in turn reduces the binding affinity of LPS to colistin. 13 The mcr-1 gene can be spread between different bacteria in various regions by plasmids or transposition, which not only poses a great threat to public health but also poses a great challenge to clinical anti-infective treatment. 14 The emergence of the transferable polymyxin resistance genes mcr-1 represents a new mechanism of bacterial resistance that greatly challenges the last line of defence for the treatment of multidrug-resistant gram-negative pathogens. 15 There is a need for an MCR inhibitor that works synergistically with polymyxin to treat infections caused by polymyxin B-resistant mcr-1-positive Enterobacteriaceae. Isoalantolactone (IAL) has been shown to possess various pharmacological activities, including antitrypanosomal, anti-apoptosis and antimicrobial activities. [16][17][18] In addition, it has been reported that IAL has anticancer activity against several cancer cells, such as cervical squamous cell carcinoma, prostate cancer and gastric adenocarcinoma. 19,20 However, as a promising natural compound, the inhibitory effect of IAL against bacterial resistance enzymes has not been reported.
Here, we screened IAL as a new MCR-1 inhibitor with the ability to enhance the sensitivity of mcr-1-positive Enterobacteriaceae to polymyxin B. Furthermore, the synergistic effect of IAL and polymyxin B was determined both in vivo and in vitro.

| Bacterial strains and reagents
The strains used in this experiment were described in our preceding studies [21][22][23] and are listed in Table 1

| FIC values determination
The microbroth checkerboard dilution method recommended by the Nation Committee for Clinical Laboratory Standards (NCCLS) 24,25 was employed to evaluate the combined bactericidal effect of IAL and polymyxin B. The strains listed in Table 1

| Growth curve determination
The tested strains K pneumoniae ZJ02, E coli DZ2-12R or E coli BL21(DE3) (pET28a-mcr-1) were inoculated into a 2 mL of LB overnight and expanded into 100 mL of LB medium. When the OD value of the bacterial solution reached 0.3, the LB medium was quantitatively dispensed and supplemented with IAL at final concentrations of 0, 16, 32, 64 or 128 μg/mL. The cultures were grown at 37°C with shaking at 200 rpm, and a growth curve was plotted by monitoring the OD 600nm value at 0.5-to 1-hour intervals.

| Time-killing assays
The bactericidal curve was used to evaluate the potential bactericidal effect of IAL combined with polymyxin B. For the tested strains K pneumoniae ZJ02, E coli BL21(DE3)(PET28a-mcr-1) and E coli DZ2-12R, the concentrations of polymyxin B used in this study were 4, 2 and 2 μg/mL, respectively. The concentration of IAL was 32 μg/mL. The tested strains were treated with IAL, polymyxin B, a combination of IAL and polymyxin B, or were not treated (positive control). The bacterial solution of each sample was diluted with PBS, plated on TSB solid medium and grown for 0, 1, 3, 5 and 7 hours. Following cultivation in a 37°C incubator for 24 hours, the bacteria on the plate were counted to construct the timekilling curve.

| Combined disc test
The bacterial solutions (MCR-1-positive E coli BL21(DE3) (pET28amcr-1), E coli DZ2-12R and K pneumoniae ZJ02 or negative control bacteria, E coli ATCC 25922) were diluted 1:4000 and uniformly coated on the surface of LB solid medium containing various concentrations of IAL (0, 4 or 32 μg/mL). Then, a tablet containing polymyxin B was placed in the middle of the plate. The diameter of the inhibition zones was measured to evaluate the combination effect after being placed in a 37°C incubator for 24 hours.

| Western blot assay
The tested strains K pneumoniae ZJ02, E coli DZ2-12R or E coli BL21 (DE3) (pET28a-mcr-1) were cocultured with various concentrations of IAL (0, 4 and 32 μg/mL) as described in the growth curve determination for 4 hours. Following centrifugation at 13 800 g for 5 min, the precipitate of each sample was resuspended in loading buffer, mixed, boiled at 100°C for 7 minutes and separated by SDS-PAGE. 27 Then, the protein was transferred to a PVDF membrane. After blocking, the mouse monoclonal antibody against MCR-1 (1:2000, Laboratory preservation), goat antimouse HRP-conjugated secondary antibodies (1:2000, Proteintech) were applied for the determination of MCR-1 production as described in our previous study. 23 Isocitrate dehydrogenase (ICDH) was used as an internal control according to a previous study. Note: The concentration of ISO was 32 μg/mL for all tested strains. All the MIC results were counted as no less than 3 times. The FICs lower than 0.5 was displayed in bold and recognized as synergistic effect.

| Ethics statement
Six-to eight-week-old female Balb/c mice with an average bodyweight of 20 ± 2 g were purchased from the Experimental Animal Center of Jilin University (Changchun, Jilin, China). Animal experiments were approved by and conducted in accordance with the guidelines of the Animal Care and Use Committee of Jilin University.

| Mouse thigh infection with K pneumoniae ZJ02 and E coli DZ2-12R
Establishment of a model of mouse thigh muscle infection by experimental methods in the literature. 28 In this experiment, K pneumoniae

| Survival analysis
Overnight culture of E coli DZ2-12R was injected intraperitoneally into mice with 2 × 10 8 CFUs per mouse. After inoculation, the mice were divided randomly into four groups (n

| Lung tissue section experiment
Overnight culture of K pneumoniae ZJ02 was injected transnasally into mice with 5 × 10 7 CFUs per mouse. After inoculation, the mice were

| Statistical analysis
The data are presented as the average plus the relative standard deviation and analysed with Student's t test. P-value of <.05 was considered significant. *, P < .05; **, P < .01.

| IAL enhances mcr-1-positive Enterobacteriaceae sensitivity to carbapenems without affecting bacterial growth
The checkerboard method is commonly used to screen inhibitors of resistance enzymes. 29 Here, we found that 32 μg/mL IAL, combined with polymyxin B or colistin, had a significantly synergistic effect against the engineering strain E coli BL21(DE3) (pET28a-mcr-1) ( However, the treatment with IAL alone at concentrations varying from 16 to 128 μg/mL had no effect on the growth of the tested strains (E coli BL21(DE3) (pET28a-mcr-1), E coli DZ2-12R and K pneumoniae ZJ02, shown in Figure 1A-C, respectively), suggesting that no antibacterial activity existed at IAL concentrations lower than 128 μg/mL. Furthermore, the synergistic effect of IAL and carbapenems was also observed in the time-killing analysis ( Figure 1D

| IAL had no inhibitory effect on MCR-1 production
A direct inhibition of MCR-1 activity or MCR-1 production may contribute to the synergistic effect of IAL with carbapenems.
Western blot analysis was further employed to evaluate whether IAL treatment inhibits MCR-1 production. 30 As shown in Figure

| Combination therapy of IAL and polymyxin B synergistically inhibits the pathogenicity of mcr-1positive bacteria in vivo
To detect the same synergistic bactericidal effect of IAL and polymyxin B for K pneumoniae ZJ02 and E coli DZ2-12R in vivo, a model of mouse thigh muscle infection was established in this study 31 by assessing the bacterial burden in the thigh 36 hours post-infection.
As shown in Figure 5A, treatment with IAL or polymyxin B alone did not result in an evident reduction in the bacterial burden in E coli DZ2-12R-infected mice. However, a combination therapy of IAL and polymyxin B significantly reduced the bacterial colonization in the thigh ( Figure 5A). Although a significant difference between the polymyxin B treatment and solvent control was observed for the K pneumoniae ZJ02-infected mice, the combined therapy resulted in a more than 1.5-log 10 reduction of the bacterial burden in infected mice when compared with the polymyxin B treatment ( Figure 5B).
Together, these results indicated that IAL combined with polymyxin B synergistically inhibited the bacterial burden in mice.

| Protection rate experiment
To further evaluate this synergistic effect, the survival rate of E coli DZ2-12R in intraperitoneally infected mice treated with IAL, polymyxin B or a combination of IAL and polymyxin B was monitored for 72 hours.
As expected, all infected mice were dead after 72 hours. After infection, the mice were monitored for 3 days using the same treatment method, combined with the solvent control group and the IAL group alone, and the survival rate of the mice in the antibiotic group alone was 20% ( Figure 6). The combined polymyxin B and IAL group had a survival rate of 60% (Figure 6), further confirming that the combination treatment has a better effect than the drug or antibiotic alone.
Finally, the pathological damage in the lung was assessed in a mouse model of pneumonia by K pneumoniae ZJ02. As shown in Figure 7, no pathological damage was observed for the uninfected mice under the microscope. In the infected mice, the alveolar cavity was filled with exuded cellulose, alveolar septum thickening and capillaries. Consistent with the above results, the symptoms of the drug-treated group and the antibiotic-treated group were slightly relieved ( Figure 7). However, the cellulose exudation in the alveolar cavity was reduced, and the pulmonary interstitial was only infiltrated by a small amount of inflammatory cells, suggesting that the pathological damage was significantly reduced (Figure 7). Taken together, IAL combined with polymyxin B remarkably inhibited the pathogenicity of mcr-1-positive bacteria in mice.

| D ISCUSS I ON
The emergence and spread of drug-resistant bacteria that have caused serious clinical infections is a matter of great concern to researchers.
The prevalence of mcr-1-positive K pneumoniae and E coli has brought great challenges to clinical treatment, which calls for novel strategies or agents to fight these bacterial pathogens. Among these strategies, the identification of effective MCR-1 inhibitors combined with antibiotics is a promising strategy to address this challenge. Here, we successfully Isoalantolactone has various pharmacologic effects, 31,32 such as inhibiting inflammation, preventing proliferation and inducing apoptosis. 33 Furthermore, this compound has been reported to significantly induce breast cancer cell apoptosis by activating the caspase cascade, cleaving poly (ADP-ribose) polymerase. 34 In addition, IAL was recently identified as selectively toxic to cancer cells. 18,35 These results suggested that IAL is a multiple biological compound, as a drug, used in clinical practice. However, IAL is not soluble in water, which limited its pharmacological activity in vitro/in vivo. Therefore, the molecular structure of IAL requires further modification.
In the present study, we confirmed the synergistic effect of polymyxin B and IAL on plasmid-mediated mcr-1-positive Enterobacteriaceae, both in vitro and in vivo. From checkerboard microdilution assays, we found that IAL only has a synergistic effect on mcr-1-positive Enterobacteriaceae in combination with carbapenem antibiotics, but it has no effect when used in combination with other antibiotics. Moreover, IAL has no synergistic effect on mcr-1-negative isolates (Table 1). Additionally, IAL treatment did not affect the production of MCR-1 in mcr-1-positive Enterobacteriaceae (Figure 4).
This particular synergistic effect suggests that IAL may act directly on MCR-1 and affect its activity. Furthermore, the combination of isoalantolactone and polymyxin B had the best synergistic effect on Klebsiella pneumonia, with a 16-fold change in the MIC values (Table 1).
Upon examination, compared with other MCR-1 inhibitors, IAL has a great advantage with a good synergistic effect on polymyxin-resistant gram-negative bacteria both in vitro and in vivo. All the information Taken together, our results suggest that the combination of IAL and carbapenems can be an option for the treatment of MCR-1positive Enterobacteriaceae. As a promising natural compound, IAL has more pharmacological activities to be discovered.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare. Each mouse was transnasally infected with 5 × 10 7 CFUs K pneumoniae ZJ02 and treated with the solvent control (50 μL of DMSO), IAL (50 mg/kg), polymyxin B (5 mg/kg) or the combination therapy (50 mg/kg IAL + 5 mg/kg polymyxin B). Then, the lungs of infected mice were analysed by histopathological analysis