Vanderbilt University
Vanderbilt Institute of Chemical Biology

home research discoveries core facilities training & research opportunities seminars & events news giving contact


Discoveries Featured

Combatting Antibacterial Drug Resistance

By: Carol A. Rouzer, VICB Communications
Published: September 26, 2013

New analogs to important broad-spectrum antibacterials show promising efficacy against drug-resistant targets.

The discovery of antibiotics to treat bacterial infections was one of the seminal accomplishments of 20th century medicine. Yet, the rapid emergence of antibiotic resistant strains of many bacterial pathogens threatens to reverse the major gains in global public health derived from the use of these important drugs. This problem is illustrated by the two fluoroquinolone antibacterials, ciprofloxacin and moxifloxacin (Figure 1), which exhibit broad-spectrum activity against many Gram-negative and Gram-positive bacteria. These drugs are the first-in-line treatment for many infections, but the growing number of resistant bacterial strains threatens to markedly reduce their clinical usefulness. To address this challenge, Vanderbilt Institute of Chemical Biology member Neil Osheroff, his student Katie Aldred, and their collaborators Robert Kerns (University of Iowa) and Charles Turnbough (University of Alabama, Birmingham) explored the mechanism of resistance to the fluoroquinolone antibacterials. Their studies have led to the discovery of new molecules that promise to be effective against the majority of fluoroquinolone-resistant bacterial strains [K. J. Aldred et al. (2013) ACS Chem. Bio., published online September 18, DOI: 10.1021/cb400592n].

Figure 1
. Structures of ciprofloxacin and moxifloxacin.

The molecular targets of the fluoroquinolones are the bacterial enzymes DNA gyrase and topoisomerase IV, which play critical roles in regulating DNA supercoiling and in untangling DNA during replication. Both of these enzymes work by binding a DNA segment, designated the G segment, and then a second segment, the T segment. The G segment is cleaved with the formation of a phosphotyrosine link between each of the newly generated DNA termini and one of two active site tyrosine residues in the protein. The T segment then passes between the cleaved ends of the G segment and is released from the enzyme. Following exit of the T segment, the two ends of the G segment are ligated. Binding and hydrolysis of ATP are required to induce the conformational changes needed to cleave and move the two DNA segments (Figure 2). Ciprofloxacin and moxifloxacin alter the activity of the enzyme by inhibiting its ability to ligate the ends of the G segment after cleavage. The result is an accumulation of DNA double-strand breaks, ultimately leading to cell death.

Figure 2
. Generalized mechanism for the action of a topoisomerase, shown here for human topoisomerase II. The enzyme active site opens and binds the first segment of DNA, designated the G segment. It then binds a second segment, the T segment. After cleavage of the G segment, the T segment is passed through the protein and released on the opposite side. Mg2+ ion is required for cleavage of the G segment, which results in the formation of phosphotyrosine bonds between the cleaved ends of the DNA strand and active site tyrosine residues. ATP binding causes a conformational change that closes the enzyme active site. ATP hydrolysis may assist in passage of the T segment. Reprinted by permission from Macmillan Publishers Ltd. from J. L. Nitiss (2009) Nat. Rev. Can., 9, 327, copyright (2009).

X-ray diffraction analysis of a complex containing Acinetobacter baumannii topoisomerase IV, moxifloxacin, and cleaved DNA indicates that the fluoroquinolones bind to the enzyme through a complexed Mg2+ ion. The ion is coordinated by the C3 keto and C4 carboxylate substituents on the fluoroquinolone skeleton and surrounding water molecules. The water molecules, in turn, hydrogen-bond to a serine and a glutamic acid residue in the protein (Figure 3). Consistent with this model, the most common mechanism by which bacteria gain resistance to the fluoroquinolone antibacterials is through mutation of the serine or the glutamic acid residue involved in Mg2+ complexation. Also consistent is the finding that quinazolinediones (Figure 3) that are structurally similar to ciprofloxacin or moxifloxacin but lack the C3 carboxylate required for the Mg2+ interaction have no topoisomerase IV modulatory activity. However, some quinazolinediones bearing different substituents at C7 from those of ciprofloxacin and moxifloxacin are active against topoisomerase IV, suggesting that other modes of binding to the enzyme are possible.

Figure 3
. The quinolone antibacterials bind to topoisomerase IV through a coordinated Mg2+ ion (orange), which links to a serine and a glutamic acid residue through hydrogen-bonded water molecules (blue). The quinazolinedione skeleton lacks the C3 carboxylate required to form this complex. Reproduced by permission from K. J. Aldred et al. (2013) ACS Chem. Bio., published online September 18, DOI: 10.1021/cb400592n, copyright 2013, American Chemical Society.

To better understand the interaction of fluoroquinolones and quinazolinediones with topoisomerase IV, the investigators synthesized a series of compounds based on these core structures and a quinazolinedione skeleton lacking the amino group at N3. The compounds contained a piperazinyl, diazabicyclononyl, or 3′-(aminomethyl)pyrrolidinyl [3′(AM)P] substituent at the C7 position, and a hydrogen, methyl, or methoxy substituent at C8 (Figure 4). The investigators found that fluoroquinolones bearing the piperazinyl or diazabicyclononyl substituents as found in ciprofloxacin and moxifloxacin, respectively, were active against wild-type B. anthracis topoisomerase IV, but not against two resistant enzymes bearing mutations of the serine residue required for Mg2+-dependent drug binding. Quinazolinediones and non-amino quinazolinediones bearing these substituents had little to no activity on either the wild-type or mutant enzymes, confirming the requirement for the C3 carboxylate for binding to the enzyme. In contrast, compounds bearing a 3′(AM)P substituent at C7 displayed activity with both the wild-type and mutant enzymes regardless of the core scaffold or the substituent at C8 (Figure 4). These results suggest that the 3′(AM)P substituent at C7 provides a mechanism for enzyme binding that does not require Mg2+ coordination. The findings also imply that fluoroquinolones or quinazolinediones bearing a C7 3′(AM)P substituent may be effective drugs against ciprofloxacin-and moxifloxacin-resistant bacteria.

Figure 4. Structures of the fluoroquinolone, quinazolinedione, and non-amino quinazolinedione cores and the piperazinyl, diazabicyclononyl, and 3′(aminomethyl)pyrrolidine C7 substituents. Compounds bearing various C8 substituents were synthesized as shown. Blue color indicates that the compound modulates the activity of wild-type but not resistant mutant bacterial topoisomerase. Red color indicates that the compound modulates the activity of both the wild-type and mutant enzymes. Black color indicates that the compound has no activity with either wild-type or mutant enzymes.

Both ciprofloxacin and moxifloxacin are useful drugs because they modulate the activity of bacterial type II topoisomerases but not the human enzymes. Their ineffectiveness against human topoisomerase IIα can be explained by the fact that this enzyme lacks the serine and glutamic acid residues that allow Mg2+-mediated binding of the drugs to the bacterial enzyme. The investigators used site-specific mutagenesis to place a serine and glutamic acid residue in the appropriate positions of human topoisomerase IIα. They found that both ciprofloxacin and moxifloxacin could modulate the activity of this mutant enzyme, while the quinazolinedione analogs of the two drugs were ineffective (Figure 5). These findings confirmed, once again, the requirement for Mg2+-dependent binding in the action of the drugs. However, these results also suggested a disconcerting hypothesis. If compounds bearing a 3′(AM)P substituent at C7 could bind to the bacterial enzyme by a mechanism that does not require Mg2+-coordination, they might also bind to and modulate the activity of the human enzyme by this mechanism. If true, these compounds could not be used as antibiotics in humans due to likely toxicity. The investigators tested this hypothesis by evaluating the activity of their compound series on human topoisomerase IIα activity. They found that fluoroquinolone and quinazolinedione compounds bearing a 3′(AM)P group at C7 and a methyl group at C8 did, indeed, modulate the activity of the human enzyme (Figure 5, red). The analogous compounds with a methoxy substituent at C8 were also effective, though somewhat less potent. In contrast, removal of the C8 substituent from these compounds, or the N3 amino group from the quinazolinedione compound eliminated activity against human topoisomerase IIα (Figure 5, blue).

Figure 5. Structures of the drug cores and substituents as shown in Figure 4. In this case, purple color indicates that the compound modulates the activity of human mutant topoisomerase IIα, but not the wild-type enzyme. Green color indicates that the compound has no activity with the mutant or wild-type enzymes. Red color indicates that the compound modulates the activity the wild-type human topoisomerase IIα, while the blue color indicates that the compound has no activity on this enzyme. Black color indicates that results for modulation of the human enzyme were not specified.


These results indicate that both the C7 and C8 substituents are important in the interaction of fluoroquinolones and quinazolinediones with human topoisomerase IIα. To better understand the roles of these substituents, the investigators carried out competitive binding studies with the human enzyme. The results showed that the C8 methyl group has a relatively modest effect on compound binding to the protein, suggesting that its role is likely in execution of the modulatory effect. In contrast, the C7 3′(AM)P group is very important to enzyme binding, confirming that this substituent mediates a key interaction with the protein.

The findings indicate that it is possible to design a molecule that is effective against both wild-type and drug-resistant bacterial topoisomerases while retaining poor efficacy against the human enzyme. Such compounds offer promise as potential antibacterial therapeutics against drug-resistant bacteria. Further work will explore a path for translation of these important findings to the clinic.









Vanderbilt University School of Medicine | Vanderbilt University Medical Center | Vanderbilt University | Eskind Biomedical Library

The Vanderbilt Institute of Chemical Biology 896 Preston Building, Nashville, TN 37232-6304 866.303 VICB (8422) fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2013 by Vanderbilt University Medical Center