Back

Backward collateral sensitivity can restore antibiotic susceptibility

Chowdhury, F. R.; Findlay, B.

2024-11-08 evolutionary biology
10.1101/2024.11.06.622341 bioRxiv
Show abstract

The prevalence of antibiotic resistance continues to rise, rendering many valuable drugs ineffective. Antibiotic cycling regimens that incorporate collateral sensitivity (CS), the phenomenon where resistance to one antibiotic leads to hypersensitivity to another, are hypothesized to slow the evolution of antibiotic resistance. However, the repeatability of CS interactions and their ability to drive bacterial extinction and resensitizations remain unclear. In this study, we thoroughly investigate four drug pairs proposed for cycling regimens with experimental evolution. We find that reported pairwise CS interactions are not always robust, and even when they are, forward CS (where resistance to drug A leads to hypersensitivity to drug B) does not reliably reduce resistance or promote bacterial extinction. Instead, we find that if evolution of resistance to drug B in naive cells is associated with CS to drug A, a phenomenon we term backward CS, drug A-resistant cells can be rendered more sensitive to A again when resistance to B develops. We describe the mechanism of resistance disruption via backward CS in an aminoglycoside-{beta}-lactam pair, where perturbation of the electron transport chain to inhibit aminoglycoside entry impairs {beta}-lactam efflux. Overall, we highlight the importance of applying antibiotics in the correct order in cycling regimens and identify robust CS interactions that may be used to design treatment regimens less likely to lead to resistance evolution. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=129 SRC="FIGDIR/small/622341v1_ufig1.gif" ALT="Figure 1"> View larger version (18K): org.highwire.dtl.DTLVardef@1745531org.highwire.dtl.DTLVardef@85172corg.highwire.dtl.DTLVardef@1b48784org.highwire.dtl.DTLVardef@12d2153_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOTOC GraphicC_FLOATNO C_FIG

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.