Target-Mediated Fluoroquinolone Resistance in Neisseria gonorrhoeae: Actions of Ciprofloxacin against Gyrase and Topoisomerase IV

Abstract

Fluoroquinolones make up a critically important class of antibacterials administered worldwide to treat human infections. However, their clinical utility has been curtailed by target-mediated resistance, which is caused by mutations in the fluoroquinolone targets, gyrase and topoisomerase IV. An important pathogen that has been affected by this resistance is Neisseria gonorrhoeae, the causative agent of gonorrhea. Over 82 million new cases of this sexually transmitted infection were reported globally in 2020. Despite the impact of fluoroquinolone resistance on gonorrhea treatment, little is known about the interactions of this drug class with its targets in this bacterium. Therefore, we investigated the effects of the fluoroquinolone ciprofloxacin on the catalytic and DNA cleavage activities of wild-type gyrase and topoisomerase IV and the corresponding enzymes that harbor mutations associated with cellular and clinical resistance to fluoroquinolones. Results indicate that ciprofloxacin interacts with both gyrase (its primary target) and topoisomerase IV (its secondary target) through a water-metal ion bridge that has been described in other species. Moreover, mutations in amino acid residues that anchor this bridge diminish the susceptibility of the enzymes for the drug, leading to fluoroquinolone resistance. Results further suggest that ciprofloxacin primarily induces its cytotoxic effects by enhancing gyrase-mediated DNA cleavage as opposed to inhibiting the DNA supercoiling activity of the enzyme. In conclusion, this work links the effects of ciprofloxacin on wild-type and resistant gyrase to results reported for cellular and clinical studies and provides a mechanistic explanation for the targeting and resistance of fluoroquinolones in N. gonorrhoeae.

Keywords: DNA cleavage; DNA supercoiling/decatenation; ciprofloxacin; fluoroquinolone; gyrase; topoisomerase IV.

Figure 1

Schematic of the water–metal ion bridge that mediates interactions between fluoroquinolones and bacterial type II topoisomerases. Residue numbering reflects that of N. gonorrhoeae gyrase (GyrA) and topoisomerase IV (ParC). For simplicity, only interactions with the protein (and not DNA) are shown. A noncatalytic divalent metal ion (orange, Mg²⁺) forms an octahedral coordination sphere (green dashed lines) between four water molecules (green) and the C3/C4 keto acid of ciprofloxacin (black). Two of the water molecules form hydrogen bonds (blue dashed lines) with the serine side chain hydroxyl group (blue), and one water molecule forms hydrogen bonds (red dashed lines) with the aspartic acid (GyrA) or glutamic acid (ParC) side chain carboxyl group (red).

Figure 2

Effects of ciprofloxacin on DNA supercoiling catalyzed by WT and mutant N. gonorrhoeae gyrase. The abilities of WT (black), GyrA^S91F (S91F, blue), GyrA^D95G (D95G, red), and GyrA^S91F/D95G (S91F/D95G, purple) gyrase to supercoil relaxed plasmids in the presence of ciprofloxacin are shown in the left panel. The right-hand panel displays the ability of GyrA^S91F/D95G gyrase to supercoil relaxed DNA at high ciprofloxacin concentrations (up to 500 μM). Error bars represent the standard deviation of at least 3 independent experiments. The table indicates the corresponding IC₅₀ values (the drug concentration at which the enzyme activity is inhibited by 50%), including the standard error of the mean and the fold-change in IC₅₀ from WT.

Figure 3

Effects of ciprofloxacin on DNA cleavage mediated by WT and mutant N. gonorrhoeae gyrase. The ability of ciprofloxacin to induce double-stranded (DS) DNA cleavage mediated by WT (black), GyrA^S91F (S91F, blue), GyrA^D95G (D95G, red), and GyrA^S91F/D95G (S91F/D95G, purple) gyrase is shown in the top panel. Error bars represent the standard deviation of at least 3 independent experiments. The table at the bottom lists the CC₅₀ value (the drug concentration at which 50% maximal DS DNA cleavage is reached) for each enzyme, including the standard error of the mean, the fold-change in CC₅₀ from WT, and the max % DSB value (maximal percentage of DS DNA breaks) induced at 100 μM ciprofloxacin. Values marked as N/A (Not Analyzed) were excluded from additional analyses due to low signal.

Figure 4

Effects of 8-methyl-2,4-quinazolinedione on the DNA supercoiling activities of GyrA^S91F/D95GN. gonorrhoeae gyrase. The ability of 8-methyl-2,4-quinazolinedione (8-Me-2,4-QD, structure at the right) to inhibit DNA supercoiling catalyzed by GyrA^S91F/D95G gyrase (purple) is shown in the left panel (IC₅₀ = 19.7 ± 1.2 μM). Error bars represent the standard deviation of at least 3 independent experiments. The ability of 0–500 μM ciprofloxacin to compete with 50 μM 8-methyl-2,4-quinazolinedione (QD) for GyrA^S91F/D95G gyrase during DNA supercoiling is shown in the gel in the right panel. Reaction mixtures contained DNA in the absence of the enzyme (DNA) or GyrA^S91F/D95G gyrase in the absence of the compound (ND, no drug), in the presence of either 50 μM 8-methyl-2,4-quinazolinedione (QD) or 500 μM ciprofloxacin (FQ) alone or in the presence of 50 μM QD and increasing concentrations of ciprofloxacin (cipro, 50–500 μM). Both drugs were added to reaction mixtures simultaneously. The positions of the relaxed (Relax) and negatively supercoiled [(−)SC] plasmids are indicated. The gel is representative of 3 independent experiments.

Figure 5

Effects of 8-methyl-2,4-quinazolinedione on the DNA cleavage activities of the WT and GyrA^S91F/D95G gyrase. The ability of 8-methyl-2,4-quinazolinedione (8-me-2,4-QD) to induce double-stranded (DS) DNA cleavage mediated by WT (black) and GyrA^S91F/D95G gyrase (S91F/D95G, purple) is shown in the left panel. The inset table shows the corresponding CC₅₀, including the standard error of the mean and max % DSB values. The ability of 0–500 μM ciprofloxacin (Cipro) to compete with 25 μM 8-methyl-2,4-quinazolinedione (QD) for GyrA^S91F/D95G gyrase-mediated DNA cleavage is shown in the right panel, including the IC₅₀ value. Both drugs were added to reaction mixtures simultaneously. The relative contribution of the quinazolinedione to the total level of DNA cleavage was calculated as follows: (DS DNA cleavage in the presence of quinazolinedione and fluoroquinolone—DS DNA cleavage in the absence of either compound)/(DS DNA cleavage in the presence of 25 μM quinazolinedione only). Error bars represent the standard deviation of at least 3 independent experiments. A gel image displaying the competition data quantified in the right panel is shown at the top. Reaction mixtures contained DNA in the absence of the enzyme (DNA) or GyrA^S91F/D95G gyrase in the absence of the compound (ND, no drug), in the presence of either 25 μM 8-methyl-2,4-quinazolinedione (QD) or 500 μM ciprofloxacin (FQ) alone, or in the presence of 25 μM 8-methyl-2,4-quinazolinedione (QD) and increasing concentrations of ciprofloxacin (Cipro, 50–500 μM). The positions of nicked (Nick), linear (Lin), and negatively supercoiled [(−)SC] plasmids are indicated. The gel is representative of 3 independent experiments.

Figure 6

Effects of ciprofloxacin on the DNA decatenation activities of WT and mutant N. gonorrhoeae topoisomerase IV. The ability of ciprofloxacin to inhibit decatenation catalyzed by WT (black), ParC^S87N (S87N, blue), ParC^E91G (E91G, red), and ParC^S87N/E91G (S87N/E91G, purple) topoisomerase IV is shown in the top panel. Error bars represent the standard deviations of at least 3 independent experiments. Corresponding IC₅₀ values, including the standard error of the mean and the fold-change in IC₅₀ from WT, are indicated in the table at the bottom.

Figure 7

Effects of ciprofloxacin on the DNA cleavage activities of WT and mutant N. gonorrhoeae topoisomerase IV. The ability of ciprofloxacin to induce double-stranded (DS) DNA cleavage mediated by WT (black), ParC^S87N (S87N, blue), ParC^E91G (E91G, red), and ParC^S87N/E91G (S87N/E91G, purple) topoisomerase IV is displayed in the top panel. Error bars represent the standard deviations of at least 3 independent experiments. The corresponding CC₅₀ values, including the standard error of the mean, fold-change in CC₅₀ from WT, and max % DSB values, are indicated in the table at the bottom.