Bacterial strains, plasmids, and antimicrobial susceptibility testing

The clinical colistin-resistant strains TSAREC02 and TSAREC03 were obtained from a previously described nationwide surveillance program and their relevant antimicrobial susceptibility data were collected as previously reported [15]. To determine the colistin minimum inhibitory concentrations (MICs) and breakpoints in the current study, we adhered to the broth microdilution method according to the guidelines provided by the international susceptibility testing committee, specifically the European Committee on Antimicrobial Susceptibility Testing 2021. For further information on the E. coli strains, plasmids, and primers used in this study, please refer to Supplementary Table S1, S2, and S3, respectively.

PCR (polymerase chain reaction amplification) and sequencing of genes involved in colistin resistance

In addition to pmrA and pmrB, the genes phoP, phoQ, and mgrB from TSAREC02 and TSAREC03 were amplified and sequenced using the appropriate primers (Supplementary Table S3) [16]. As described previously, the sequences of phoP, phoQ, and mgrB from TSAREC02 and TSAREC03 were compared to those of the wild type (WT) E. coli reference strain MG1655 and eight colistin-sensitive clinical E. coli strains (ECS01, ECS02, ECS03, ECS04, ECS05, ECS06, ECS07, and ECS08) to exclude possible synonymous polymorphisms. This enables the identification of any unique amino acid substitutions that may contribute to colistin resistance in TSAREC02 and TSAREC03. Sequence comparisons was analyzed on the National Center for Biotechnology Information website (www.ncbi.nlm.nih.gov). The Basic Local Alignment Search Tool was used for this analysis [15]. Protein Variation Effect Analyzer (https://www.jcvi.org/research/provean) software was used to predict whether the unique amino acid substitutions found in PhoP, PhoQ, and MgrB would affect the function of these proteins.

Construction of the mgrB or pmrB deletion mutant of TSAREC02, as well as the phoP or pmrB deletion mutants of TSAREC03

The generation of the mgrB deletion mutant of TSAREC02 and the phoP deletion mutant of TSAREC03 were carried out utilizing plasmid-based gene knockout methods with a suicide vector called pUT-KB plasmid [17]. The protocol for generating these mutants followed a similar approach as previously described but with minor modifications [15]. For the creation of the mgrB deletion mutant from TSAREC02, a two-step PCR approach was employed. Two PCR products were initially generated using primer sets KOmgrB_F-1/KOmgrB_R-1 and KOmgrB_F-2/KOmgrB_R-2. These PCR products contained complementary ends that would later be used for overlap PCR. To generate the final deletion fragment, the two PCR products with complementary ends were combined as templates and amplified using KOmgrB_F-1 and KOmgrB_R-2 in an overlap PCR reaction. This resulted in the creation of a 1,667-bp fragment lacking the mgrB gene with 15 nucleotides complementary to a PfoI-digested linear pUT-KB plasmid on both sides. The resulting gel-purified PCR then were cloned into a PfoI-digested linear pUT-KB plasmid with the In-Fusion HD cloning kit (TaKaRa Bio, Japan) according to the manufacturer’ s instructions. Thereafter, the final plasmid was constructed named pUT-KB-KOmgrB. The plasmid pUT-KB-KOmgrB then was transformed into E. coli S17-1 λpir through the process of electroporation. Subsequently, this plasmid was mobilized into TSAREC02 through conjugation. The transconjugants were screened on Luria-Bertani (LB) agar supplemented with kanamycin (50 µg/mL) and colistin (4 µg/mL). The selected transconjugants then were further incubated in 20 mL of brain-heart infusion broth for 6 h without kanamycin. Subsequently, they were spread onto LB agar containing 10% w/v sucrose. After the double-crossover events had occurred, sucrose-resistant and kanamycin sensitive colonies were selected (indicating loss of the plasmid). The mgrB deletion mutants of TSAREC02, hereafter, was referred to as TSAREC02_ΔmgrBΔ43–47. As for phoP deletion mutant construction from TSAREC03, the process is slimier to the previous steps described for the mgrB deletion mutant, with the main difference being the primer sets used. The phoP deletion mutant therefore was referred to as TSAREC03_ΔphoPΔ84–224. The pmrB deletion mutants from TSAREC02 and TSAREC03 were constructed as our reported study with some modification [4]. As for the pmrB deletion mutant of TSAREC02/TSAREC03, the strain TSAREC02/TSAREC03 was used as the template. Two PCR products were created using the primer sets KOpmrB_XbaI-F/ KOpmrB-R and KOpmrB-F/ KOpmrB_BcuI-R. The two PCR products contained complementary ends and were mixed as a template for the subsequent overlap PCR step. Using the primers KOpmrB_XbaI-F and KOpmrB_BcuI-R, the mixed PCR products were amplified again, resulting in a fragment with a 1088 bp deletion in the pmrB gene through overlap PCR. The resulting PCR fragment was subjected to digestion with XbaI and BcuI enzymes. Simultaneously, the pUT-KB plasmid was also digested with XbaI and BcuI. The digested PCR fragment was then inserted into the digested pUT-KB plasmid, resulting in the construction of the pUT-KB-KOpmrB plasmid. Further steps involving the homologous recombination method to construct the pmrB deletion mutant of TSAREC02/TSAREC03 were the same as TSAREC02_ΔmgrBΔ43–47 and TSAREC03_ΔphoPΔ84–224. The pmrB deletion mutants from TSAREC02 and TSAREC03 were referred to as TSAREC02 _ΔpmrBg616a, t618g, t664c and TSAREC03_ΔpmrBg3c, t41c, c532t, c704a, respectively.

Construction of the TSAREC02 mutant with a WT-mgrB replacement and the TSAREC03 mutant with a WT-phoP replacement

The mutated mgrB in TSAEC02 and the mutated phoP in TSAREC03 were replaced with WT-pmrB and phoP from E.coli MG1655, respectively. DNA fragments of the mgrB and phoP genes from E. coli MG1655 were amplified via PCR using the primer sets KOmgrB_F-1/KOmgrB_R-2 for mgrB and KOphoP_F-1/KOphoP_R-2 for phoP. The generated PCR fragments were cloned into pUT-KB, resulting in pUT-KB-WT-mgrB and pUT-KB-WT-phoP, respectively. Using a previously described allelic exchange method with a plasmid pUT-KB-WT-mgrB and a pUT-KB-WT-phoP, a TSAREC02 mutant with WT-mgrB replacement (TSAREC02 WT-mgrB revertant) and TSAREC03 mutant with WT-phoP replacement (TSAREC03 WT-phoP revertant) were created from TSAREC02_ΔmgrBΔ43–47 and TSAREC03_ΔphoPΔ84–224, respectively [4].

Complementation of different TSAREC02 and TSAREC03 mutants

The pmrB genes from TSAREC02 and TSAREC03 were amplified using PCR.

The amplified pmrB genes were then directly inserted into the pCRII-TOPO TA vector (Invitrogen, U.S.A). The insertion was done following the protocol provided by the manufacturer. The ligation reaction between the amplified pmrB genes and the pCRII-TOPO vector resulted in the creation of different pCRII-TOPO-derived plasmids. Each generated plasmid containing a specific variant of the pmrB gene were transformed into different TSAREC02 and TSAREC03 mutants via electroporation method. In the current study, to confirm accuracy the target gene with its franking regions in deletion, revertant and complemented mutants, several experiments were performed. First, we employed PCR method using specific primers (see Table S3) to amplify the region of interest. Subsequently, agarose gel electrophoresis was conducted to visualize the amplicons and confirm their sizes as expected. The expected amplicon sizes using primer pairs mgrB-F and mgrB-R from TSAREC02, TSAREC02_ΔmgrBΔ43–47,and TSAREC02 WT-mgrB revertant were 297 bp, 115 and 297 bp, respectively. The expected amplicon size using primer pairs phoP-F and phoP-R from TSAREC03, TSAREC03_ΔphoPΔ84–224 and TSAREC03 WT-phoP revertant were 920 bp, 250 bp, 920 bp, respectively. The expected amplicon sizes using the primer pairs pmrB-F and pmrB-R from TSAREC02 and TSAREC02 _ΔpmrBg616a, t618g, t664c as well as TSAREC03 and TSAREC03_ΔpmrBg3c, t41c, c532t, c704a were 1397 bp and 302 bp, respectively. Additionally, two amplicons with sizes of 302 and 1397 bp from complemented strains were anticipated. These amplicons serve as indicators: the 302 bp fragment suggests the presence of the deleted pmrB gene in the chromosome, while the 1397 bp fragment indicates the intact cloned pmrB gene in the plasmid transformed. Finally, to ensure the accuracy of the results, the amplicons were subjected to sequencing using sequencing primer for validation (see Table S3). The primer pairs M13-F and M13-R designed based on the sequence of the pCRII-TOPO TA vector (Invitrogen, U.S.A) were used to sequence the cloned pmrB from complemented strains.

Real-time quantitative PCR

An increased expression of the pmrHFIJKLM operon in E. coli is associated with colistin resistance [1]. The expression of the pmrK gene, representative of pmrHFIJKLM operon expression, was measured via real-time quantitative PCR method as described previously [15]. Other genes regulating PmrAB and PhoPQ systems which may be related to colistin resistance were also measured. Briefly, Bacterial RNA and cDNA were obtained using an RNeasy mini kit (Qiagen, U.S.A) and Prime Script RT Master Mix (TaKaRa Bio, Japan) according to the manufacturer’s instructions. Expression levels of pmrK, phoP, pmrD, pmrA and pmrB were estimated using SYBR Green PCR Master Mix (Thermo Fisher Scientific, U.S.A). The gapA gene served as an endogenous reference for normalizing expression levels. Primer pairs used were according to previous studies and were listed in Supplementary Table S3 [8, 14, 18, 19]. Expression levels were calibrated against the baseline expression level of E. coli MG1655, and fold change in expression was calculated using the comparative threshold cycle method [20]. Data are expressed as the mean ± standard deviation of four independent experiments (performed in triplicate). A Student’s t-test was used for statistical analysis. The p values < 0.05 were considered statistically significant.