Download Mechanisms Responsible for Cross

SUPPLEMENT ARTICLE
ANTIMICROBIAL ACTIVITY
Mechanisms Responsible for Cross-Resistance
and Dichotomous Resistance among
the Quinolones
Christine C. Sanders
Center for Research in Anti-Infectives and Biotechnology, Department of Medical Microbiology and Immunology, Creighton University School
of Medicine, Omaha
Resistance to the quinolones almost always arises from the accumulation of mutations in chromosomal genes
responsible for the drug targets, permeability, or active efflux. This resistance can be depicted as a stepwise
process in which each step, represented by separate mutations, diminishes susceptibility on average 4- to 8fold. The precise path followed in this stepwise process differs with the quinolone that selects resistance as
well as the organism involved. At each step, the influence of each mutation on susceptibility to other quinolones
not used in the selection process varies greatly, and a pattern of either cross-resistance or dichotomous resistance
may be seen. From an understanding of the stepwise process by which resistance to the quinolones evolves,
it is possible to use an 8-fold rule to predict which compounds may provide effective therapy for a given
infection and be least likely to select for resistance.
Antimicrobial drug resistance is becoming an increasing
cause for concern. As more and more antibiotics are
being used in both the inpatient and outpatient setting,
more and more resistance is being encountered. It appears that for the first time, antimicrobial drug resistance may be outstripping development of new agents
designed to provide effective therapy for infections
caused by organisms that are resistant to older agents.
One approach to the design of new effective agents
is to understand the mechanisms responsible for resistance to older antibiotics. These mechanisms include
alteration in target, production of a drug-inactivating
enzyme, and modifications that prevent adequate cellular concentrations of drug from being achieved (table
1). An altered target can occur by any one of a variety
of mechanisms and is responsible for the penicillin resistance currently seen among pneumococci and the
glycopeptide resistance occurring among enterococci
Reprints or correspondence: Dr. Christine C. Sanders, Creighton University School
of Medicine, 2500 California Plaza, Omaha, NE 68178 ([email protected]).
Clinical Infectious Diseases 2001; 32(Suppl 1):S1–8
Q 2001 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2001/3206S1-0001$03.00
[1–4]. Drug-inactivating enzymes that add a chemical
group include the diverse aminoglycoside-inactivating
enzymes and chloramphenicol acetyltransferase [5, 6].
The b-lactamases are examples of enzymes that cleave
an essential bond to destroy the drug [7, 8].
“Impermeability” has often been the term applied to
the third general mechanism (table 1), but recent work
shows this to be a great oversimplification. Resistance
via this mechanism not only involves failure of the drug
to gain entry into the cell via passive diffusion through
the outer membrane of gram-negative bacteria, it also
includes failure of the drug to be actively transported
across the cytoplasmic membrane—the latter a problem
with aminoglycosides. Added to these “impermeability”
mechanisms is the recent recognition of active efflux
pumps that, when overexpressed, can reduce cellular
concentrations of drugs below their effective concentration [9]. Some of these pumps can affect not only
compounds in the cytoplasm of the cell, but also those
such as b-lactam antibiotics that stay primarily in the
periplasmic space of gram-negative bacteria.
Among the 3 general mechanisms of resistance outlined above, 2 appear to be very important to resistance
Resistance among the Quinolones • CID 2001:32 (Suppl 1) • S1
Table 1. General mechanisms responsible for resistance to
antimicrobial agents.
Mechanism of resistance
Alteration in target
Mutation in chromosomal gene leading to less susceptible
target
Hyperproduction of susceptible target to saturate drug
Acquisition of genes providing new, less susceptible target
Production of drug-inactivating enzyme
Addition of chemical group to inactivate drug
Cleavage of essential bond in drug
Modifications preventing adequate cellular drug concentrations
from being achieved
Changes in outer membrane porins of gram-negative bacteria
Alterations in active efflux pumps
Alterations in active transport into the cell
NOTE.
From [1–11].
to the quinolones: altered target and modifications affecting
cellular concentrations of the drugs. It is the purpose of this
article to review the mechanisms responsible for resistance to
the quinolones and the process whereby organisms develop
resistance to the drugs. This process, depending on the quinolones and mechanisms involved, can lead to either cross-resistance or dichotomous resistance to other quinolone agents.
An understanding of each is an important first step in determining which quinolones may provide effective therapy for a
given infection and be least likely to select for additional resistance to this important drug class.
MECHANISMS OF RESISTANCE TO THE
QUINOLONES
Alterations in drug target. Two topoisomerases involved in
DNA synthesis, DNA gyrase and topoisomerase IV, have been
identified as the major targets of the quinolones [10–12]. In
gram-negative bacteria, DNA gyrase appears to be the favored
target, whereas in gram-positive bacteria, topoisomerase IV appears the preferred target. It is thus not surprising that alterations in the subunits of these enzymes are the single most
prevalent mechanism of resistance encountered among clinical
isolates of both gram-positive and gram-negative bacteria [1,
11, 13–27]. Mutations in gyrA, the gene encoding the A subunit
of DNA gyrase (GyrA), are the most common mechanism involved in quinolone resistance among gram-negative bacteria,
whereas mutations in parC (grlA in staphylococci), the gene
encoding the C subunit of topoisomerase IV (ParC), are most
commonly encountered among quinolone-resistant grampositive bacteria. Mutations affecting the B subunit of the DNA
gyrase (GyrB) and the E subunit of topoisomerase IV (ParE)
S2 • CID 2001:32 (Suppl 1) • Sanders
have also been observed, but these are less commonly encountered and are often found secondary to mutations in the other
2 subunits [1, 11, 12, 27].
Not only are mutations within the genes encoding GyrA and
ParC the most common mechanisms responsible for resistance
to the quinolones among clinical isolates, but there are “hot
spots” for these mutations within similar areas of the genes
encoding them [1, 11]. These hot spots are called quinolone
resistance–determining regions and involve regions between
amino acid 67 and 106 in GyrA, with amino acids 83 and 87
most often involved [1, 11]. Similar quinolone resistance–
determining regions are found in analogous regions of ParC.
Modifications affecting cellular drug concentration. Lowlevel resistance to the quinolones has also been observed in
gram-negative bacteria with altered outer membrane porins
[27, 28]. This suggests that in these organisms, entrance of the
drugs through the outer membrane is retarded and thus inadequate cellular concentrations are achieved. More important
are the recent observations that overexpression of active efflux
pumps in both gram-positive and gram-negative organisms can
lead to quinolone resistance [9, 29–35]. In gram-positive bacteria, multidrug efflux pumps such as NorA in staphylococci
and PmrA in pneumococci are responsible for quinolone resistance [29–32, 36].
In gram-negative bacteria, these pumps are more complex
and are generally composed of 3 subunits: a transporter protein
in the cytoplasmic membrane that removes the drug from the
cytoplasm; an outer membrane channel, usually a porin, that
funnels the drug out of the cell; and a periplasmic linker protein
that connects the other 2 components [9]. This complex arrangement allows for the active efflux of many different drugs,
including those that accumulate primarily in the periplasmic
space. Resistance to the quinolones and other drugs transported
by these efflux pumps arises when those that are constitutively
active become more active or when those that are usually inactive become active [9]. The net effect is that efflux increases,
significantly reducing cellular accumulation of the drugs.
Figure 1. The evolution of resistance to quinolones. Each step in the
evolution represents a spontaneous mutation that diminishes quinolone
susceptibility 4- to 8-fold. Thus the MIC of the quinolone used to select
mutants from the wild type (WT) is 4- to 8-fold diminished for successive
first-step (1M), second-step (2M), and third-step (3M) mutants.
Table 2.
Evolution of quinolone resistance
among Staphylococcus aureus and Escherichia
coli.
MIC, mg/mL
Bacterium
Ciprofloxacin
Ofloxacin
S. aureus
Wild type
0.25
First-step mutant
2.0
Multiple-step mutant
1128
0.12
1.0
32
E. coli
Wild type
0.002
First-step mutant
0.015
Multiple-step mutant
8.0
NOTE.
0.03
0.25
16.0
Data kindly provided by K.S. Thomson.
EVOLUTION OF QUINOLONE RESISTANCE
Virtually all quinolone resistance encountered in clinical isolates
has involved mechanisms that result from mutations in chromosomal genes of the isolates [1, 11, 27, 28, 37, 38]. There
have been rare, often unconfirmed reports of plasmid-mediated
resistance to the quinolones (reviewed in Martinez-Martinez et
al. [39]). Although plasmid-mediated resistance is certainly
possible both theoretically and scientifically, its rarity precludes
its inclusion in any discussion of the evolution of resistance to
quinolones at this time.
The evolution of resistance to quinolones arises in a stepwise
fashion through the accumulation of spontaneous mutations
in chromosomal genes encoding cellular products involved with
the drugs [1, 11, 27, 28, 40–43]. These include the drug targets
themselves, as well as those cellular mechanisms involved in
drug entry into or exit out of the cell. The precise effect of
each mutation, taken as a single event, on quinolone susceptibility varies widely depending on the gene involved, the organism, and the specific quinolone. However, in studies involving well-characterized single-step mutants, each mutation
affecting the quinolones most often diminished susceptibility
4- to 8-fold [11, 26, 27, 31, 33, 43–47]. In wild-type cells (i.e.,
cells with no mutations in their chromosomal genes), the intrinsic susceptibility to a specific quinolone is determined by
the ability of the drug to enter and accumulate in the cell and
the potency of the drug for the primary target enzyme in the
cell—the DNA gyrase or topoisomerase IV.
From these basic principles, the evolution of quinolone resistance can be predicted (figure 1). As wild-type cells replicate,
mistakes are made in duplicating the DNA. These mistakes are
spontaneous mutations that arise at measurable frequencies
usually varying from 1 in 106 (mutation frequency p 1026) to
1 in 109 (mutation frequency p 1029) wild-type cells. Progeny
of wild-type cells with a single mutation affecting quinolone
susceptibility are called first-step mutants and are usually 4- to
8-fold less susceptible than wild-type cells to any quinolone
affected by the mutation (figure 1). As the first-step mutants
replicate, spontaneous mutations can occur among these cells
as well. Progeny of first-step mutants with a second mutation
affecting quinolone susceptibility are called second-step mutants (figure 1). These are 4- to 8-fold less susceptible than the
first-step mutants and 16- to 64-fold less susceptible than the
wild-type cells. Single mutations can be accumulated through
successive generations to produce third-, fourth-, and fifth-step
mutants, and so on (figure 1).
The precise mutation occurring in each step of the pathway
varies greatly, depending on the organism involved and the quinolone used to select each mutant [1, 11, 44, 48–50]. In general,
first-step mutants usually involve alterations in the preferred target of the quinolone for the organism. Thus first-step mutants
of Streptococcus pneumoniae selected with ciprofloxacin tend to
be parC mutants, whereas those selected with sparfloxacin tend
to be gyrA mutants, reflecting the different preferred target of
these 2 quinolones for this species [11, 50]. Furthermore, firststep mutants of gram-negative bacteria tend to have altered gyrA,
whereas first-step mutants of gram-positive bacteria tend to have
altered parC. Precisely where in the stepwise evolution mutations
affecting permeability and efflux occur is highly variable [15, 31,
36]. However, among Staphylococcus aureus, first-step mutants
may be efflux mutants [36].
The evolution of clinically relevant resistance (i.e., MIC above
the susceptible breakpoint) to any specific quinolone is determined by the intrinsic potency of the drug against the wild type,
Figure 2. Cross-resistance among the quinolones. The evolution of
resistance to quinolone A as selected by quinolone A is shown in the
left portion of the figure, with the MIC for each successive mutant (mg/
mL) given below each step. Cross-resistance evolving simultaneously to
quinolone B, which is 4-fold more potent than quinolone A against the
wild type (WT), is shown (right). With each successive mutation selected
by quinolone A, susceptibility to quinolone B diminishes, but differences
in potency between the 2 drugs are maintained. If both quinolones achieve
a concentration of 2 mg/mL at the site of infection, the 8-fold rule would
predict that quinolone B would provide the most effective therapy and
be less likely to select for resistance because achievable concentrations
exceed the MIC for the wild-type and first-step mutants.
Resistance among the Quinolones • CID 2001:32 (Suppl 1) • S3
Figure 3. Dichotomous resistance among the quinolones. The evolution of resistance to quinolone A as selected by quinolone A is shown
(left), with each successive mutation causing diminished susceptibility to
quinolone A. Because the mechanisms responsible for the mutations in
the first-step (1M) and third-step (3M) mutants do not affect susceptibility
to quinolone B, a pattern of dichotomous resistance emerges. Only the
mutation in the second-step (2M) mutant reduces susceptibility to quinolone B. WT, wild type.
the achievable levels of drug at the site of infection, and the
number of successive mutations required before the diminished
susceptibility produced by each mutation exceeds the achievable
levels of drug. The greater the intrinsic potency of the drug and
the higher the achievable levels, the greater the number of mutations required for a clinically relevant level of resistance to be
achieved. For example, with ciprofloxacin, it was not surprising
to see resistance emerging quickly among staphylococci, because
these organisms possessed only marginal susceptibility to the
drug initially (table 2). Thus, a single mutation in grlA was often
sufficient to confer resistance to ciprofloxacin. Such was not the
case for Escherichia coli, which showed a much greater intrinsic
susceptibility to ciprofloxacin (table 2). In this organism, multiple
mutations involving drug target and drug accumulation were
required for resistance to occur.
This predictable, stepwise evolution of resistance to the quinolones has a positive side in that it provides an additional basis
on which to predict which quinolone may provide effective
therapy for a given infection and have a lower chance of selecting resistance. If a quinolone is desired for therapy, the MIC
of the quinolone for the organism causing the infection should
be multiplied by 8. The resultant product is the likely MIC of
the next-step mutant that will arise spontaneously from the
strain. Because the number of cells in an active infection readily
exceeds 109, it is likely that in addition to the parental cells,
there are already a few mutant cells present at the site of the
infection. Thus, the preferred quinolone to treat the infection
would be one that achieves a concentration at least 8-fold above
the MIC of the infecting strain at the site of infection. This
concentration provides adequate coverage not only for the parental cells of the infecting strain but also is likely to cover the
next-step mutant as well.
For example, if the MIC of a particular quinolone is 1 mg/mL
against a bacterial strain, the MIC of the drug for the next-step
mutant is likely to be 4–8 mg/mL. Applying the 8-fold rule, if
S4 • CID 2001:32 (Suppl 1) • Sanders
concentrations of the quinolone at the site of infection do not
exceed 8 mg/mL, another quinolone should be chosen for therapy.
Obviously, the higher the achievable concentration over the MIC
of the infecting strain, the greater the number of successive mutants that will be covered by the therapy, making selection of
resistance highly unlikely. Once this rule of 8 has been invoked,
additional factors that may affect the selection of therapy should,
of course, be taken into consideration, as always.
Drlica [12] has taken a somewhat different approach to this
same general principle. He suggests that, for each quinolone, the
mutant prevention concentration (MPC) should be determined.
The MPC is that concentration that prevents the growth of the
next-step mutant of a bacterial strain. It is determined by plating
∼1010 cells into varying concentrations of the quinolone and
determining the concentration at which no growth occurs [12].
This is certainly a valuable approach for categorizing quinolones
and their relative ability to prevent selection of mutants, and this
strategy should be pursued during the investigation of any quinolone. However, because it involves tests that are not performed
routinely by hospital laboratories, the MPC is unlikely to be
available to a physician when he or she is determining which
quinolone is best for the patient. Because an MIC is often available, the 8-fold rule can be applied readily in the clinical setting.
Were a study to be performed to determine the MPC and the
MIC derived from the 8-fold rule, it is likely that results would
be comparable because both concentrations reflect the amount
of quinolone required to prevent growth of the next-step mutant.
Cross-resistance and dichotomous resistance. Many of the
earlier fluoroquinolones were affected similarly by mutations
that diminished susceptibility to the quinolones. Thus when
one compared the potency of the various quinolones among
organisms resistant to 1 drug, a pattern of cross-resistance was
often seen (figure 2). Cross-resistance implies that as organisms
become less susceptible to one drug in a class, they become
less susceptible to others in the class. Thus, a quinolone that
was 4-fold more potent than another against wild-type cells
would continue to be 4-fold more potent against each successive mutant, but each mutant would be less susceptible to both
drugs. If the achievable concentrations of both drugs were 2
mg/mL, as shown in figure 2, the more potent drug would be
preferable for therapy because it would cover both wild-type
and first-step mutants. However, because of the cross-resistance
pattern of the mutants, neither drug would provide adequate
coverage for any successive mutants. Cross-resistance between
ciprofloxacin and ofloxacin is illustrated in the data shown for
wild types and first-step mutants in table 2.
With the development of newer fluoroquinolones, especially
those with an 8-methoxy group such as moxifloxacin, a different pattern of resistance among quinolones became apparent.
It is now clear that not all quinolones are equally affected by
all mutations involved in resistance [1, 17, 19, 20, 26, 36, 44,
Table 3.
Examples of cross-resistance and dichotomous resistance selected by quinolones in vitro.
a
Reference, strain
[31] Streptococcus pneumoniae
ATCC 49619
1
[26] VPH AQ3815
Mutant
selected
by
Location of mutation
(presumptive mechanism)b
—
Nfx
—
(efflux)
—
MIC, mg/mL
Cpfx
0.5
2–4 (C)
—
0.2
1
Cpfx
gyrA (Ser83Ile)
0.78
2
Cpfx
gyrA 1 unknownc
6.25
3
Cpfx
gyrA 1 unknown 1 parC (Ser85Phe)
4
Cpfx
gyrA 1 unknown 1 parC (Ser85Phe) 1 unknown
[48] Staphylococcus aureus MS
5935
—
—
Nfx
4
16–32
0.2
0.78 (C)
25 (C)
0.2
0.10 (D)
0.78 (C)
3
Tofx
grlA 1 gyrA 1 grlA (Glu84Lys)
100 (C)
200 (D)
4
Spfx
grlA 1 gyrA 1 grlA 1 gyrA (glu88Val)
100 (D)
200 (D)
12.5 (C)
2
Cpfx
Unknown 1 parC (79)
3
Cpfx
Unknown 1 parC (79) 1 gyrA (83)
1
Spfx
gyrA (83)
2
Spfx
gyrA (83) 1 parC (79) or (83)
1
Clin
gyrA (83) or (87) or gyrB (474)
2
Clin
gyrA (83) or (87) 1 parC (79)
0.78 (D)
6.25 (C)
0.05
grlA 1 gyrA (Ser84Lui)
Unknown
0.39 (D)
3.13 (C)
0.78
Ofx
Cpfx
0.39
200 (C)
2
1
0.39
50 (C)
1.56 (C)
12.5
100 (C)
Clin
0.12
100 (C)
0.2
Tofx
0.25–0.5 (D) 0.12–0.25 (D)
1200 (C)
grlA (Ser80Phe)
—
Ofx
50
Nfx
—
0.5
Moxi
200
1
[49] S. pneumoniae 7785
Spfx
12.5 (C)
50 (C)
0.025
0.2 (C)
25 (C)
12.5
50 (D)
100 (C)
25
400 (C)
25 (D)
200
3.13 (C)
1
0.25
0.12
3
0.25 (D)
0.12 (D)
8
64
0.25 (D)
32 (C)
1 (D)
2
32 (C)
16
1–2 (D)
1–2 (C)
32–64 (C)
16–32 (C)
0.25 (D)
1 (C)
0.25 (D)
1 (C)
0.25
1
3
Clin
gyrA (83) 1 (87) 1 parC (79)
64 (D)
64 (D)
4
Clin
gyrA (83) 1 (87) 1 parC (79) 1 parE (454)
128 (D)
64 (D)
32–64
4
Clin
gyrA (83) 1 (87) 1 parC (79) 1 (83)
128 (D)
64 (D)
64
6
NOTE. MIC must differ at least 4-fold from parental strain for cross-resistance. Clin, clinafloxacin; cpfx, ciprofloxacin; moxi, moxifloxacin; nfx, norfloxacin; ofx, ofloxacin; spfx, sparfloxacin; tofx, tosufloxacin;
VPH, Vibrio parahaemolyticus.
a
The numbers 1, 2, 3, and 4 refer to first-step mutant, second-step mutant, third-step mutant, and fourth-step mutant.
The precise location of mutation is indicated by gene, if known. If phenotypic tests were used to ascertain the mutation, the result listed as a presumptive mechanism. Mutations of parental mutant used to
select successive mutant are indicated by gene only. The type of resistance relative to the drug used to select the mutant is indicated in parentheses by C if cross-resistant and by D if dichotomous resistant.
c
Unknown. Genetic lesion was not found among the genes investigated in the study.
b
Table 4.
Examples of cross-resistance and dichotomous resistance to quinolones among clinical isolates.
MIC, mg/mL
Reference
d
b
c
Phenotype
Genotype
Cpfx
Trfx
Spfx
a
Lvfx
[16]
Cpfx S–trfx S Wild type
0.25
0.02–0.03
0.06
0.12–0.19
[16]
Cpfx R–trfx S gyrA (84) 1 grlA (80)
8–64
0.75
8
8–12
[16]
Cpfx R–trfx R gyrA (84) 1 grlA (80) 1
grlA (84)
8–16
8–16
[17]e
Cpfx S–trfx S Wild type
0.12–0.25
0.01–0.03
0.06–0.12
0.12–0.25
3–48
0.25–2
1–8
2–8
8–32
8
8–16
128
Ofx
Gefx
Clin
32
[17]
Cpfx R–trfx S gyrA (84) 1 grlA (80) or (84)
[17]
Cpfx R–trfx R gyrA (84) 1 grlA (80) 1
grlA (84)
[19]
Susceptible
Wild type
0.25–0.5
1
2
0.25 <0.12
[19]
Ofx R–trfx S
parC (79)
0.5–1
1–2
4
8
1–2
[19]
Ofx R–trfx S
gyrA (81) 1 parE (435)
0.5
2
8
[19]
Ofx R–trfx R
gyrA (81) 1 parC (80) 1
parC (83)
8
16
16
116
8
0.5
[19]
8
16
16
116
8
0.5
64–128
<0.12
Ofx R–trfx R
gyrA (81) 1 parC (79)
[20]
Susceptible
Wild type
[20]
Cpfx R–trfx S gyrA (114) 1 parC (91)
[20]
Cpfx R–trfx S parC (79) 1 parE (460)
2
0.5
0.5
4
[20]
Cpfx R–trfx S gyrA (114) 1 parC (79) 1 (91) 1
parE (493)
4
0.5
1
4
f
[20]
Cpfx R–trfx S parC (83) 1 parE (460)
[20]
Cpfx R–trfx R gyrA (81) 1 parE (460) 1 (435)
NOTE.
0.5
0.12
0.25
1
2
0.12
0.5
2
4
14
0.5
0.5
2
4
16 to 116 2
0.25
0.5–1
4
18
Clin, clinafloxacin; cpfx, ciprofloxacin; gefx, grepafloxacin; lvfx, levofloxacin; ofx, ofloxacin; spfx, sparfloxacin; trfx, trovafloxacin.
a
Range is shown when multiple isolates were involved.
b
The phenotype is based on susceptibility pattern; S, susceptible; R, resistant.
c
The gene and the location of the mutation are shown. If no mutations were found among genes examined, the wild-type genotype was presumed if the
phenotype also reflected wild-type susceptibility.
d
Staphylococcus aureus.
e
Coagulase-negative staphylococci.
f
Streptococcus pneumoniae.
48–52]. For example, differences in primary targets are responsible for resistance affecting one quinolone but not another. As mentioned above, in S. pneumoniae, the primary target for ciprofloxacin is topoisomerase IV, whereas the primary
target for sparfloxacin is DNA gyrase. Thus, single-step mutants
selected by sparfloxacin that are altered in gyrA will be less
susceptible to sparfloxacin but not to ciprofloxacin. Conversely,
those mutants selected with ciprofloxacin that are altered in
parC will be less susceptible to ciprofloxacin but not to sparfloxacin. Differences in quinolone hydrophobicity also influence the effect of specific efflux pumps on quinolone resistance
[9, 31, 48, 52–54]. In staphylococci, NorA has a greater effect
on susceptibility to norfloxacin and ciprofloxacin than it has
on susceptibility to sparfloxacin, gatifloxacin, clinafloxacin, grepafloxacin, and moxifloxacin.
Because not all mechanisms of resistance affect all quinolones
equally, a pattern of dichotomous resistance has emerged, especially between older and newer quinolones in studies with
gram-positive bacteria (figure 3). With dichotomous resistance,
the pattern of evolving resistance to 2 quinolones resembles
S6 • CID 2001:32 (Suppl 1) • Sanders
that of a branching tree, with the path of one drug leaving the
other at each mutation step that affects only 1 of the 2. In the
example shown, the evolution of resistance to quinolone A as
selected by quinolone A is shown, with each successive mutation reducing susceptibility 4- to 8-fold. However, simultaneous evolution of resistance to quinolone B does not follow
the same path because mutations occurring in the first- and
third-step mutants do not affect quinolone B.
Examples of dichotomous resistance and cross-resistance
among quinolones are shown in table 3 for mutants isolated
sequentially in vitro from wild-type strains. The different pathways to resistance depending on the drug used in selection of
the mutants are illustrated in the data shown in table 3 from
Pan and Fisher [49]. Examples of dichotomous resistance and
cross-resistance for clinical isolates found to be quinolone resistant are shown in table 4.
Selection of resistance. It may be argued that because of
the dichotomous resistance between older and newer fluoroquinolones, the latter should be held in reserve for use with
infections caused by gram-positive strains resistant to the older
agents. However, it should be noted that the newer fluoroquinolones, in addition to being less affected by mechanisms causing
resistance to older agents, are less likely to select resistant mutants
(i.e., the mutation frequencies for first-step mutants selected by
the newer quinolones are lower than those for mutants selected
by the older drugs) [12, 44, 47, 48]. Thus the likelihood of a
spontaneous mutation occurring that affects the newer quinolones is much smaller. This has been hypothesized to be due to
the greater lethal effect of the newer quinolones [12]. Although
the precise mechanism or mechanisms responsible for the lethal
effect of the quinolones is unknown, induction of the SOS response appears to be involved [11, 12]. The more rapid the lethal
effect, the less likely cells containing mutations arising from an
activated SOS system are to arise. The smaller likelihood of selection of resistance by newer quinolones may also be related to
the fact that some appear to target both DNA gyrase and topoisomerase IV equally [49]. In S. pneumoniae, clinafloxacin
appears to target both DNA gyrase and topoisomerase IV. Thus,
for significant resistance to clinafloxacin to appear in this organism, mutations in genes encoding both targets need to occur.
Along these same lines, if resistance to newer quinolones
among gram-positive organisms requires 2 mutations, and firststep mutants are less likely to be selected by them, older agents
should definitely be avoided. Use of older agents would be more
likely to select first-step mutants. Many of these mutants, although appearing fully susceptible to newer agents, have a silent
mutation that will be expressed when a second mutation occurs.
If in this scenario a newer quinolone is introduced, the nextstep mutant to be selected is a second-step mutant. Studies
have shown that mutation frequencies for second-step mutants
selected by newer quinolones are similar to those selected by
older agents [12, 48]. Thus resistance to the newer quinolones
will arise sooner in this scenario because the strains have been
primed for resistance by use of older agents than if the newer
agents had been used from the start.
This possibility is illustrated by results obtained in studies
performed by Fukuda et al. [48], shown in table 3. First-step
mutants of S. aureus MS 5935 selected by older quinolones
showed dichotomous resistance to sparfloxacin and were altered in grlA. Second-step mutants altered in both grlA and
gyrA were 256-fold less susceptible to sparfloxacin than the
wild-type or first-step mutant. This greatly exceeds the 8-fold
decrease in susceptibility seen with single mutations and suggests that the grlA mutation was silent until the second mutation affecting gyrA occurred. When the second mutation occurred, both contributed significantly to the greatly reduced
susceptibility to sparfloxacin. These results, coupled with the
observations that sparfloxacin-resistant clinical isolates of
staphylococci usually possess mutations in both grlA and gyrA
[16, 17], strongly suggest that a mutation in both targets is
necessary for resistance to occur.
CONCLUSIONS
The mechanisms responsible for resistance to the quinolones
are many and varied. As more quinolones are developed, the
mechanisms responsible for resistance promise to become more
numerous and complex. However, as the evolution of resistance
becomes better understood and the mechanisms responsible
for cross-resistance and dichotomous resistance become elucidated, it should be possible to predict which quinolones are
more likely to provide effective therapy and which are less likely
to select for resistance to this important drug class.
Acknowledgments
I acknowledge the support and encouragement received from
W. Eugene Sanders, Jr., during my career. Grateful thanks are
extended to members of the Center for Research in AntiInfectives and Biotechnology for their involvement in the many
quinolone projects that led to this article. They include Kenneth
S. Thomson, Anton F. Ehrhardt, Philip D. Lister, Nancy D.
Hanson, Ellen S. Moland, Jennifer A. Black, and Patti J. Karl.
References
1. Sanders CC, Sanders WE Jr. Resistance to antibacterial agents. In: Jungkind DL, Mortensen JE, Fraimow HF, Calandra GD, eds. Antimicrobial
resistance: a crisis in health care. New York: Plenum Press, 1995; 15–23.
2. Tomasz A. Antibiotic resistance in Streptococcus pneumoniae. Clin Infect
Dis 1997; 24(Suppl 1):S85–8.
3. Grebe T, Hakenbeck R. Penicillin-binding proteins 2b and 2x of Streptococcus pneumoniae are primary resistance determinants for different
classes of b-lactam antibiotics. Antimicrob Agents Chemother 1996; 40:
829–34.
4. Murray BE. The life and times of the Enterococcus. Clin Microbiol
Rev 1990; 3:46–65.
5. Murray IA, Shaw WV. O-Acetyltransferases for chloramphenicol and
other natural products. Antimicrob Agents Chemother 1997; 41:1–6.
6. Davies JE. Aminoglycoside-aminocyclitol antibiotics and their modifying enzymes. In: Lorian V, ed. Antibiotics in laboratory medicine.
2d ed. Baltimore: Williams & Wilkins, 1986.
7. Sanders CC. b-Lactamases of gram-negative bacteria: new challenges
for new drugs. Clin Infect Dis 1992; 14:1089–99.
8. Bush K, Jacoby GA, Medeiros AA. A functional classification scheme
for b-lactamases and its correlation with molecular structure. Antimicrob Agents Chemother 1995; 39:1211–33.
9. Nikaido H. Antibiotic resistance caused by gram-negative multidrug
efflux pumps. Clin Infect Dis 1998; 27(Suppl 1):S32–41.
10. Hooper DC. Bacterial topoisomerases, anti-topoisomerases, and antitopoisomerase resistance. Clin Infect Dis 1998; 27(Suppl 1):S54–63.
11. Drlica K, Zhao X. DNA gyrase, topoisomerase IV, and the 4-quinolones.
Microbiol Mol Biol Rev 1997; 61:377–92.
12. Drlica K. Refining the fluoroquinolones. ASM News 1999; 65:410–5.
13. Knapp JS, Fox KK, Trees DL, Whittington WL. Fluoroquinolone resistance in Neisseria gonorrhoeae. Emerg Infect Dis 1997; 3:33–9 [erratum: Emerg Infect Dis 1997; 3:584].
14. Piddock LJ. Quinolone resistance and Campylobacter spp. J Antimicrob
Chemother 1995; 36:891–8.
15. Jalal S, Wretlind B. Mechanisms of quinolone resistance in clinical
strains of Pseudomonas aeruginosa. Microb Drug Resist 1998; 4:257–61.
16. Fitzgibbon JE, John JF, Delucia JL, Dubin DT. Topoisomerase mutations
Resistance among the Quinolones • CID 2001:32 (Suppl 1) • S7
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
in trovafloxacin-resistant Staphylococcus aureus. Antimicrob Agents
Chemother 1998; 42:2122–4.
Dubin DT, Fitzgibbon JE, Nahvi MD, John JF. Topoisomerase sequences
of coagulase-negative staphylococcal isolates resistant to ciprofloxacin or
trovafloxacin. Antimicrob Agents Chemother 1999; 43:1631–7.
Schmitz F-J, Kohrer K, Scheuring S, et al. The stability of grlA, grlB,
gyrA, gyrB and norA mutations and MIC values of five fluoroquinolones
in three different clonal populations of methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect 1999:287–90.
Jorgensen JH, Weigel LM, Ferraro MJ, Swenson JM, Tenover FC. Activities of newer fluoroquinolones against Streptococcus pneumoniae
clinical isolates including those with mutations in the gyrA, parC, and
parE loci. Antimicrob Agents Chemother 1999; 43:329–34.
Pestova E, Beyer R, Cianciotto NP, Noskin GA, Peterson LR. Contribution of topoisomerase IV and DNA gyrase mutations in Streptococcus
pneumoniae to resistance to novel fluoroquinolones. Antimicrob Agents
Chemother 1999; 43:2000–4.
Takenouchi T, Sakagawa E, Sugawara M. Detection of gyrA mutations
among 335 Pseudomonas aeruginosa strains isolated in Japan and their
susceptibilities to fluoroquinolones. Antimicrob Agents Chemother
1999; 43:406–9.
Mouneimne H, Robert J, Jarlier V, Cambau E. Type II topoisomerase
mutations in ciprofloxacin-resistant strains of Pseudomonas aeruginosa.
Antimicrob Agents Chemother 1999; 43:62–6.
Bachoual R, Tankovic J, Soussy CJ. Analysis of the mutations involved
in fluoroquinolone resistance of in vivo and in vitro mutants of Escherichia coli. Microb Drug Resist 1998; 4:271–6.
Brisse S, Milatovic D, Fluit AC, et al. Comparative in vitro activities
of ciprofloxacin, clinafloxacin, gatifloxacin, levofloxacin, moxifloxacin,
and trovafloxacin against Klebsiella pneumoniae, Klebsiella oxytoca, Enterobacter cloacae, and Enterobacter aerogenes clinical isolates with alterations in GyrA and ParC proteins. Antimicrob Agents Chemother
1999; 43:2051–5.
El Amin NA, Jalal S, Wretlind B. Alterations in GyrA and ParC associated with fluoroquinolone resistance in Enterococcus faecium. Antimicrob Agents Chemother 1999; 43:947–9.
Okuda J, Hayakawa E, Nishibuchi M, Nishino T. Sequence analysis of
the gyrA and parC homologues of a wild-type strain of Vibrio parahaemolyticus and its fluoroquinolone-resistant mutants. Antimicrob
Agents Chemother 1999; 43:1156–62.
Acar JF, Goldstein FW. Trends in bacterial resistance to fluoroquinolones. Clin Infect Dis 1997; 24(Suppl 1):S67-S73.
Sanders CC. Microbiology of fluoroquinolones. In: Sanders WE Jr,
Sanders CC, eds. Fluoroquinolones in the treatment of infectious diseases. Glenview, IL: Physicians and Scientists Publishing, 1990:1–28.
Muñoz-Bellido JL, Alonso Manzanares MA, Martinez Andrés JA, et
al. Efflux pump-mediated quinolone resistance in Staphylococcus aureus
strains wild type for gyrA, gyrB, grlA, and norA. Antimicrob Agents
Chemother 1999; 43:354–6.
Aeschlimann JR, Dresser LD, Kaatz GW, Rybak MJ. Effects of NorA
inhibitors on in vitro antibacterial activities and postantibiotic effects
of levofloxacin, ciprofloxacin, and norfloxacin in genetically related
strains of Staphylococcus aureus. Antimicrob Agents Chemother
1999; 43:335–40.
Brenwald NP, Gill MJ, Wise R. Prevalence of a putative efflux mechanism among fluoroquinolone-resistant clinical isolates of Streptococcus
pneumoniae. Antimicrob Agents Chemother 1998; 42:2032–5.
Gill MJ, Brenwald NP, Wise R. Identification of an efflux pump gene,
pmrA, associated with fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob Agents Chemother 1999; 43:187–9.
Lomovskaya O, Lee A, Hoshino K, et al. Use of a genetic approach to
evaluate the consequences of inhibition of efflux pumps in Pseudomonas aeruginosa. Antimicrob Agents Chemother 1999; 43:1340–6.
Mine T, Morita Y, Kataoka A, Mizushima T, Tsuchiya T. Expression in
Escherichia coli of a new multidrug efflux pump, MexXY, from Pseudomonas aeruginosa. Antimicrob Agents Chemother 1999; 43:415–7.
S8 • CID 2001:32 (Suppl 1) • Sanders
35. Ziha-Zarifi I, Llanes C, Kohler T, Pechere JC, Plesiat P. In vivo emergence of multidrug-resistant mutants of Pseudomonas aeruginosa overexpressing the active efflux system MexA-MexB-OprM. Antimicrob
Agents Chemother 1999; 43:287–91.
36. Sulavik MC, Barg NL. Examination of methicillin-resistant and methicillin-susceptible Staphylococcus aureus mutants with low-level fluoroquinolone resistance. Antimicrob Agents Chemother 1998; 42:3317–9.
37. Martinez JL, Alonso A, Gomez-Gomez JM, Baquero F. Quinolone resistance by mutations in chromosomal gyrase genes. Just the tip of the
iceberg? J Antimicrob Chemother 1998; 42:683–8.
38. Courvalin P. Plasmid-mediated 4-quinolone resistance: a real or apparent absence? Antimicrob Agents Chemother 1990; 34:681–4.
39. Martinez-Martinez L, Pascual A, Jacoby GA. Quinolone resistance from
a transferable plasmid. Lancet 1998; 351:797–9.
40. Trucksis M, Hooper DC, Wolfson JS. Emerging resistance to fluoroquinolones in staphylococci: an alert [editorial]. Ann Intern Med
1991; 114:424–6.
41. Ferrero L, Cameron B, Crouzet J. Analysis of gyrA and grlA mutations
in stepwise-selected ciprofloxacin-resistant mutants of Staphylococcus
aureus. Antimicrob Agents Chemother 1995; 39:1554–8.
42. Piddock LJ. Mechanisms of resistance to fluoroquinolones: state-ofthe-art 1992–1994. Drugs 1995; 49(Suppl 2):29–35.
43. Sanders CC, Sanders WE Jr, Thomson KS. Fluoroquinolone resistance
in staphylococci: new challenges. Eur J Clin Microbiol Infect Dis
1995; 14(Suppl 1):6–11.
44. Gootz TD, Zaniewski RP, Haskell SL, Kaczmarek FS, Maurice AE. Activities of trovafloxacin compared with those of other fluoroquinolones
against purified topoisomerases and gyrA and grlA mutants of Staphylococcus aureus. Antimicrob Agents Chemother 1999; 43:1845–55.
45. Varon E, Janoir C, Kitzis M-D, Gutmann L. ParC and GyrA may be
interchangeable initial targets of some fluoroquinolones in Streptococcus pneumoniae. Antimicrob Agents Chemother 1999; 43:302–6.
46. Tankovic J, Bachoual R, Ouabdesselam S, Boudjadja A, Soussy CJ. In
vitro activity of moxifloxacin against fluoroquinolone-resistant strains
of aerobic gram-negative bacilli and Enterococcus faecalis. J Antimicrob
Chemother 1999; 43(Suppl B):19–23.
47. Schedletzky H, Wiedemann B, Heisig P. The effect of moxifloxacin on
its target topoisomerases from Escherichia coli and Staphylococcus aureus. J Antimicrob Chemother 1999; 43(Suppl B):31–7.
48. Fukuda H, Hori S, Hiramatsu K. Antibacterial activity of gatifloxacin
(AM-1155, CG5501, BMS-206584), a newly developed fluoroquinolone, against sequentially acquired quinolone-resistant mutants and the
norA transformant of Staphylococcus aureus. Antimicrob Agents
Chemother 1998; 42:1917–22.
49. Pan X-S, Fisher LM. DNA gyrase and topoisomerase IV are dual targets
of clinafloxacin action in Streptococcus pneumoniae. Antimicrob Agents
Chemother 1998; 42:2810–6.
50. Fukuda H, Hiramatsu K. Primary targets of fluoroquinolones in Streptococcus pneumoniae. Antimicrob Agents Chemother 1999; 43:410–2.
51. Schmitz FJ, Hofmann B, Hansen B, et al. Relationship between ciprofloxacin, ofloxacin, levofloxacin, sparfloxacin and moxifloxacin (BAY
12-8039) MICs and mutations in grlA, grlB, gyrA and gyrB in 116
unrelated clinical isolates of Staphylococcus aureus. J Antimicrob
Chemother 1998; 41:481–4.
52. Muñoz-Bellido JL, Alonso Manzanares MA, Yague Guirao G, et al. In
vitro activities of 13 fluoroquinolones against Staphylococcus aureus isolates with characterized mutations in gyrA, gyrB, grlA, and norA and
against wild-type isolates. Antimicrob Agents Chemother 1999; 43:966–8.
53. Kaatz GW, Leo SM, Ruble CA. Efflux-mediated fluoroquinolone resistance in Staphylococcus aureus. Antimicrob Agents Chemother
1993; 37:1086–95.
54. Schmitz FJ, Fluit AC, Luckefahr M, et al. The effect of reserpine, an
inhibitor of multidrug efflux pumps, on the in vitro activities of ciprofloxacin, sparfloxacin and moxifloxacin against clinical isolates of
Staphylococcus aureus. J Antimicrob Chemother 1998; 42:807–10.