DNA Polymerase IV dinB Favors the Adaptive Fitness of mcr-carrying Bacteria Through a Negative Feedback Regulatory Mechanism

Abstract

The plasmid-borne resistance gene mcr drastically undermines the effectiveness of colistin, posing a substantial threat to public health. Although several key plasmid elements that balance mcr-1 persistence and bacterial growth are identified, the regulatory interactions between mcr-1 and host bacteria remain poorly understood. Using a genome-wide CRISPRi crRNA library, it is identified that DNA polymerase IV, dinB, is essential for controlling the fitness cost associated with mcr-1 in Escherichia coli. The absence of dinB operon enhances mcr-1-mediated colistin resistance but simultaneously compromises bacterial growth and competitiveness. Meanwhile, dinB deficiency mitigates inflammatory response in RAW267.4 cells and enhances bacterial colonization in murine tissues. Further investigation reveals that mcr-1 actively upregulates dinB expression, with the increased reactive oxygen species induced by mcr-1 being crucial for this activation. These findings suggest that dinB modulates mcr expression and bacterial fitness via a negative feedback regulatory mechanism. Leveraging this regulatory relationship, a Toxin-Intein is engineered under the control of dinB promoter to selectively target and kill mcr-positive E. coli both in vitro and in vivo. Overall, the work uncovers a novel adaptive mechanism underlying mcr persistence and provides a precise antimicrobial strategy to combat antibiotic-resistant pathogens.

Keywords: bacteria; colistin resistance; dinB; fitness cost; mcr expression.

Figure 1

Construction of a genome‐wide CRISPRi crRNA library. a) Construction protocol of CRISPRi crRNA library based on CRISPR‐Cas adaptation system. The CRISPR‐Cas adaptation system comprises p15A replicon, four cas genes (cas1, cas2, csn2 under the L‐arabinose‐inducible promoter, and cas9 under the constitutive promoter J23110 modified with three mutations: D10A, H840A, and 473F), a single CRISPR repeat, and the minimal tracrRNA (89 bp). “‘R”’ denotes as the location of new spacers integrated into the empty CRISPR array. b) The frequency and location of crRNAs in the library generated by electroporating sheared MG1655 genomic DNA. c) The distance between adjacent crRNAs of a CRISPRi library. The distance was calculated by subtracting the position of the latter crRNA from that of the former one.

Figure 2

dinB reduces mcr‐1‐mediated colistin resistance by downregulating its expression. a) Schematic illustration of screening the genes that involved in the regulation of mcr‐1 expression. b and c) Transcriptional b) and translational levels c) of MG1655‐mcr‐1, MG1655‐∆dinB‐mcr‐1, and MG1655‐∆dinB::dinB‐mcr‐1. d) Relative bacterial surface negative charge measured by FITC‐PLL. Surface charges were calculated relative to MG1655‐mcr‐1 (set as 1). e) Absorbance of OD_{620 nm} of MG1655‐mcr‐1, MG1655‐∆dinB‐mcr‐1, and MG1655‐∆dinB::dinB‐mcr‐1 with increasing concentrations of colistin. f) Survival rates of MG1655‐mcr‐1, MG1655‐∆dinB‐mcr‐1 and MG1655‐∆dinB::dinB‐mcr‐1 upon 1 µg mL⁻¹ colistin treatment. p‐values were determined using an unpaired, two‐tailed Student's t‐test.

Figure 3

dinB improves the fitness advantages of mcr‐1‐haboring bacteria. a) Growth curves of MG1655‐mcr‐1, MG1655‐∆dinB‐mcr‐1, and MG1655‐∆dinB::dinB‐mcr‐1. b) The percentage of MG1655, MG1655‐∆dinB, and MG1655‐∆dinB::dinB bacteria, with or without mcr‐1. c) Death rate measured by SYTO9/PI live/dead bacterial double staining method. d) Relative fitness analysis of MG1655‐∆dinB‐mcr‐1, and MG1655‐∆dinB::dinB‐mcr‐1 in competition with MG1655‐pUC20 (kanamycin resistance). e) Protocol for in vivo competition model. g) The ratio changes of mcr‐1‐positive E. coli with or without dinB. p‐values were determined using an unpaired, two‐tailed Student's t‐test.

Figure 4

dinB diminishes proinflammatory response elicited by mcr‐1‐positive bacteria in both RAW267.4 cells and mice. RAW267.4 cells were infected at an MOI of 100 with APEC‐mcr‐1, APEC‐∆dinB‐mcr‐1, and APEC‐∆dinB::dinB‐mcr‐1. a–c) Levels of IL‐1α a), IL‐1β b), and IL‐6 c) in RAW267.4 cells were measured by ELISA. d) Survival rate of G. mellonella larvae over 72 h post‐infection with APEC‐mcr‐1 and APEC‐∆dinB‐mcr‐1. e) Protocols for in vivo infection model. f) APEC‐mcr‐1 and APEC‐∆dinB‐mcr‐1 loads in the heart, lung, liver, spleen, and kidney after intraperitoneal injection. p‐values were determined using an unpaired, two‐tailed Student's t‐test.

Figure 5

Mechanisms of the negative feedback regulation between mcr‐1 and dinB. a) Differential expression analysis of MG1655‐pUC19 and MG1655‐pUC19‐mcr‐1 by RNA‐seq. The x‐ and y‐axes represent the Log₂ fold change and the adjusted p value, respectively. Genes were considered upregulated (red dots) and downregulated expressed (blue dots) when they displayed a Log₂ fold change of more than 2 or less than ‐2, as well as an adjusted p‐value of less than 0.05. b and c) GO Functional enrichment of differentially upregulated b) and downregulated c) genes in MG1655‐pUC19‐mcr‐1 compared with MG1655‐pUC19. d) Transcriptional levels of dinB, yafN, and yafO in MG1655‐mcr‐1 relative to MG1655‐pUC19. e and f) Intracellular ROS level e) and transcriptional level f) of catalytic domain of mcr‐1 (mcr‐1‐C). g) Relative fitness analysis of MG1655‐mcr‐1, MG1655‐ΔdinB‐mcr‐1, MG1655‐mcr‐1‐C, and MG1655‐ΔdinB‐mcr‐1‐C competing with MG1655‐pUC20 (kanamycin resistance). h) Schematic illustration of negative feedback regulatory mechanism between mcr‐1 and dinB to maintain a balance between bacterial fitness and colistin resistance mediated by mcr‐1. p‐values were determined using an unpaired, two‐tailed Student's t‐test.

Figure 6

Construction of mcr‐1‐positive E. coli targeted killing system. a–c) Targeted killing efficacy of TK‐mcr‐1 system with different mating times a), temperatures b), and pH values c). Targeted killing efficacy is measured as the ratio of survival clones between MG1655‐mcr‐1 and MG1655‐pUC19. d) Targeted killing efficacy of TK‐mcr‐1 system in different species including E. coli EC600 and L73, S. enteritidis ATCC13076, and K. pneumoniae ATCC700603. e and f) Survival transformants e) and transconjugants f) of eighteen clinical mcr‐1‐positive and ‐negative E. coli isolates treated with TK‐mcr‐1 system with or without dinB promoter, respectively. g) Protocol for measuring targeted killing efficacy in vivo. h) Targeted killing efficacy measured by S17‐2 harboring the pTK‐mcr‐1 system with or without dinB promoter as the donor strain, and E. coli B2 as the recipient strain. p‐values were determined using an unpaired, two‐tailed Student's t‐test.