Assessing the threat of Yersinia pestis harboring a multi-resistant IncC plasmid and the efficacy of an antibiotic targeting LpxC

Abstract

Self-transmissible IncC plasmids rapidly spread multidrug resistance in many medically important pathogens worldwide. A large plasmid of this type (pIP1202, ~80 Kb) has been isolated in a clinical isolate of Yersinia pestis, the agent of plague. Here, we report that pIP1202 was highly stable in Y. pestis-infected mice and fleas and did not reduce Y. pestis virulence in these animals. Although pIP1202 inflicted a fitness cost in fleas (but not in mice) when the insects fed on blood containing a mixture of plasmid-free and plasmid-bearing strains, such a co-infection scenario has never been reported in nature, indicating that pIP1202 could persist in Y. pestis strains. Despite being resistant to commonly used antibiotic treatments, we show that plague caused by Y. pestis harboring the pIP1202 plasmid is effectively cured by LPC-233-a potent inhibitor of the essential LpxC enzyme in the lipid A biosynthetic pathway. Taken as a whole, our data highlight the alarming threat posed by Y. pestis harboring multidrug-resistant IncC plasmids that may persist in wild animals as a reservoir for long periods without antibiotic pressure and illuminate the impact of antibiotics with a novel mode of action against such a biothreat.

Keywords: Flea; LpxC; Rodents; Yersinia pestis; drug targets; multidrug resistance; plague; plasmid-mediated resistance; virulence.

Fig 1

Virulence of Y. pestis harboring pIP1202 and its parental (WT) strain in fleas and mice, and the plasmid’s stability in vivo. (A) Survival curves of mice (n = 8 for each group) inoculated intradermally with ~80 CFU of the CO92 strain harboring pIP1202 (gray) or not (white). The curves did not differ significantly (P = 0.228 in a log-rank test). (B) Overlay of a Tukey style box-and-whisker plot and scatter plot showing the bacterial load in the flea on days (D) 0, 6, and 27 after a blood meal contaminated with 5 × 10⁸ Y. pestis KIM6+/mL harboring pIP1202 (gray, n = 17 for each time point) or not (white, n = 19 for each time point). The value of the data point below the X-axis (gray, D27) is 1. The curves did not differ significantly (P > 0.08 in a two-way analysis of variance). However, Sidak’s test for multiple comparisons indicated that the bacterial load in fleas infected with a multidrug-resistant strain harboring pIP1202 was significantly lower (*P = 0.0423). The inset shows the blockage rate determined using fleas (n = 100; equal numbers of males and females) infected with the strain harboring pIP1202 or not (WT) over a period of 4 weeks. (C) Overlay of a Tukey style box-and-whisker plot and scatter plot showing the ratio between the number of antibiotic-sensitive bacteria (ATB^S) and antibiotic-resistant bacteria (ATB^R) and isolated from (i) the spleen of mice (n = 8) showing the terminal stage of bubonic plague (T) following intradermal inoculation of ~80 CFUs of the strain harboring pIP1202 and (ii) from fleas infected with Y. pestis harboring pIP1202 and collected on day (D) 6 (n = 15) and D27 (n = 20) post-infection. Panel B shows the total number of bacteria in fleas infected with the MDR strain, whereas panel C illustrates the proportion of these bacteria that retained the MDR plasmid using the ATB^S/ATB^R ratio. Additionally, panel C provides the ATB^S/ATB^R ratio based on bacterial counts from the mice used to generate panel A.

Fig 2

Competition experiments in fleas and mice infected with equal numbers of Y. pestis pIP1202 and its WT strain. (A) Overlay of a Tukey style box-and-whisker plot and scatter plot showing the ratio between the number of antibiotic-sensitive bacteria (ATB^S) and antibiotic-resistant bacteria (ATB^R) Y. pestis isolated from fleas (n = 13–19) sampled immediately (D0) or on days (D) 6, 13, and 27 after feeding on blood contaminated with a bacterial suspension comprising equal numbers (2.5 × 10⁸ CFUs) of Y. pestis harboring pIP1202 (multidrug-resistant) or not (WT). (B) The percentage of Y. pestis growing on an agar plate supplemented with antibiotic (black) or not (gray) and isolated from the spleen and blood of mice inoculated intradermally with an inoculum comprising a 1:1 mix (~80 CFUs) of Y. pestis harboring pIP1202 (multidrug-resistant) or not (WT).

Fig 3

Efficacy of LPC-233 against bubonic plague caused by multidrug-resistant and non-multidrug-resistant Y. pestis. Eight- to 9-week-old OF-1 female mice were inoculated intradermally with ~80 CFU of Y. pestis CO92 harboring pIP1202 (squares) or not (circles). Eighteen hours later, mice were given a 5-day course of intraperitoneal LPC-233 (40 mg/kg q12h, gray symbols) or not (white symbols). The two curves differed significantly (P < 0.0001 in a log-rank test). Bacterial loads in the survivors' draining lymph nodes, spleen, and blood were determined 18 h after the end of the course of treatment. A group of 12 and 11 for untreated mice infected with the CO92 strain and the CO92 harboring pIP1202 were used, respectively. A group of 11 and 13 mice for treated mice infected with the CO92 strain and the CO92 harboring pIP1202 were used, respectively. For the enumeration of the bacteria in the survivors, 10 and 12 mice were used.