Molecular Mechanisms of Antibacterial Multidrug Resistance

Leading Edge

Review Molecular Mechanisms of Antibacterial Multidrug Resistance Michael N. Alekshun1,* and Stuart B. Levy2,*

Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA Center for Adaptation Genetics and Drug Resistance, Department of Molecular Biology & Microbiology and Department of Medicine, Tufts University School of Medicine, Boston, MA 02111, USA *Correspondence: [email protected] (M.N.A), [email protected] (S.B.L.) DOI 10.1016/j.cell.2007.03.004 1 2

Treatment of infections is compromised worldwide by the emergence of bacteria that are resistant to multiple antibiotics. Although classically attributed to chromosomal mutations, resistance is most commonly associated with extrachromosomal elements acquired from other bacteria in the environment. These include different types of mobile DNA segments, such as plasmids, transposons, and integrons. However, intrinsic mechanisms not commonly specified by mobile elements—such as efflux pumps that expel multiple kinds of antibiotics—are now recognized as major contributors to multidrug resistance in bacteria. Once established, multidrug-resistant organisms persist and spread worldwide, causing clinical failures in the treatment of infections and public health crises. Efforts aimed at identifying new antibiotics were once a top research and development priority among pharmaceutical companies. The potent broad spectrum drugs that emerged from these endeavors provided extraordinary clinical efficacy. Success, however, has been compromised. We are now faced with a long list of microbes that have found ways to circumvent different structural classes of drugs and are no longer susceptible to most, if not all, therapeutic regimens.

that the drugs we use to treat infectious diseases have long-lasting effects outside of the hospital. Many antimicrobial molecules exist for millennia stably within the environment (Cook et al., 1989), where they select and promote growth of resistant strains. Resistance to single antibiotics became prominent in organisms that encountered the first commercially produced antibiotics. The most notable example is resistance to penicillin among staphylococci, specified by an enzyme (penicillinase) that degraded the antibiotic (Barber, 1947). Over the years, “There is probably no chemotherapeutic drug to which continued selective pressure by different in suitable circumstances the bacteria cannot react drugs has resulted in organisms bearing by in some way acquiring ‘fastness’ [resistance].” additional kinds of resistance mechanisms that led to multidrug resistance (MDR)— novel penicillin-binding proteins (PBPs), —Alexander Fleming, 1946 enzymatic mechanisms of drug modification, mutated drug targets, enhanced efflux The means that microbes use to evade antibiotpump expression, and altered membrane permeability. ics certainly predate and outnumber the therapeutic Some of the most problematic MDR organisms that are interventions themselves. In a recent collection of soilencountered currently include Pseudomonas aeruginosa dwelling Streptomyces (the producers of many clini(another microbe of soil origin), Acinetobacter baumancal therapeutic agents), every organism was multidrug nii, Escherichia coli and Klebsiella pneumoniae bearing resistant. Most were resistant to at least seven different extended-spectrum β-lactamases (ESBL), vancomycinantibiotics, and the phenotype of some included resistresistant enterococci (VRE), methicillin-resistant Staphyance to 15–21 different drugs (D’Costa et al., 2006). lococcus aureus (MRSA), vancomycin-resistant MRSA, Moreover, many isolates were resistant to daptomycin, and extensively drug-resistant (XDR) Mycobacterium quinupristin-dalfopristin, and telithromycin—all drugs tuberculosis (Table 1). Some like methicillin-resistant S. approved by the United States Food and Drug Adminaureus couple MDR with exceptional virulence capabiliistration (FDA) within the last decade—as well as purely ties (Miller et al., 2005). Others, including some strains of synthetic agents such as ciprofloxacin. These data not P. aeruginosa, A. baumannii, and K. pneumoniae, manonly suggest that our surroundings can act as a resage to evade every drug within the physician’s arsenal ervoir for new (and old) resistance mechanisms, but (Levin et al., 1999). Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1037

Figure 1. Acquisition of Antibiotic ­Resistance Bacteria can become antibiotic resistant (Abr) by mutation of the target gene in the chromosome. They can acquire foreign genetic material by incorporating free DNA segments into their chromosome (transformation). Genes are also transferred following infection by bacteriophage (transduction) and through plasmids and conjugative transposons during conjugation. The general term transposable element has been used to designate (1) an insertion sequence, (2) composite (compound), complex, and conjugative transposon, (3) transposing bacteriophage, or (4) integron.

Rather than providing an extensive list of antibiotic resistance mechanisms (which can be found elsewhere [Alekshun and Levy, 2000]), the aim of this Review is to highlight common genetic and molecular mechanisms of resistance as they relate to important pathogens and to drugs used by physicians today. Genetics of Multidrug Resistance Bacterial antibiotic resistance can be attained through intrinsic or acquired mechanisms (Figure 1). Intrinsic mechanisms are those specified by naturally occurring genes found on the host’s chromosome, such as, AmpC β-lactamase of gram-negative bacteria and many MDR

efflux systems (see below). Acquired mechanisms involve mutations in genes targeted by the antibiotic and the transfer of resistance determinants borne on plasmids, bacteriophages, transposons, and other mobile genetic material (Figure 1 and Table 2). In general, this exchange is accomplished through the processes of transduction (via bacteriophages), conjugation (via plasmids and conjugative transposons), and transformation (via incorporation into the chromosome of chromosomal DNA, plasmids, and other DNAs from dying organisms) (Levy and Marshall, 2004). Although gene transfer among organisms within the same genus is common, this process has also been observed between very different genera, including transfer between such evolutionarily distant organisms as gram-positive and gram-negative bacteria (Courvalin, 1994). Plasmids contain genes for resistance and many other traits; they replicate inde-

Table 1. General Characteristics of Multidrug-Resistant Organisms Drugs Considered for ­Treatment of MDRa

Organism

Common ­Infections Key Antibiotic Resistances

P. aeruginosa

Lung, wound

β-lactams, fluoroquinolones, aminoglycosides

Colistin

Acinetobacter spp.

Lung, wound, bone, blood

β-lactams, fluoroquinolones, aminoglycosides

Colistin, tigecycline

E. coli and K. Urinary, biliary, ­ neumoniae bearp gastrointestinal ing extended-spec- tracts, lung, blood trum β-lactamases

β-lactams, fluoroquinolones, aminoglycosides

Colistin (for K. pneumoniae), tigecycline

Vancomycin-resistant enterococci

Vancomycin

Quinupristin-dalfopristin, linezolid, daptomycin

Methicillin-resistant Skin and soft tisS. aureus sue, respiratory tract, blood

β-lactams, fluoroquinolones, macrolides

Quinupristin-dalfopristin, ­daptomycin, linezolid, tigecycline, vancomycin

Multidrug-resistant S. pneumoniae

Ear, lung, blood, cerebrospinal fluid

β-lactams, macrolides, tetracyclines, co-trimoxazole

Fluoroquinolones, ­tigecycline

Extensively drug-resistant M. ­tuberculosis

Lung

Rifampin, isoniazid, and three of the following: ­aminoglycosides, polypeptides, fluoroquinolones, ­thioamides, cycloserine, or para-aminosalicylic acid

3rd line agents, drug ­combinations

Blood, heart, intra-abdominal

Agents either have been approved for use by a regulatory agency (e.g., FDA), have shown usefulness in treating infection, or exhibit promising in vitro activity and await a determination of clinical efficacy. a

1038  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

Table 2. Mobile Genetic Elements Genetic Element

General Characteristics

Resistance Determinant(s) Specified/Examplesa

Plasmid

Variable size (1- >100 kb), conjugative, and mobilizable

R factor: multiple resistances

Insertion sequence

Small (<2.5 kb), contains terminal inverted repeats, and specifies a transposase

IS1, IS3, IS4, etc.

Composite (compound) transposon

Flanked by insertion sequences and/or inverted repeats

Tn5: Kan, Bleo, and Str

Complex transposon

Large (>5 kb), flanked by short terminal inverted repeats, and specifies a transposase and recombinase

Tn1 and Tn3: β-lactamase Tn7: Tmp, Str, Spc Tn1546: glycopeptides

Conjugative transposon

Promotes self-transfer

Tn916: Tet and Mino Tn1545: Tet, Mino, Ery, and Kan

Transposable bacteriophage

A bacterial virus that can insert into the ­chromosome

Mu

Other transposable elements

Other than composite, complex, and ­conjugative transposons

Tn4: Amp, Str, Sul, and Hg Tn1691: Gen, Str, Sul, Cm, and Hg

Integron

Facilitates acquisition and dissemination of gene cassettes; specifies an integrase, ­attachments sites, and transcriptional ­elements to drive expression of multiple resistance genes

Class 1: Multiple single determinants and MDR efflux pump (Qac)b Class 2: Tmp, Strp, Str, and Spc (Tn7) Class 3: carbapenems Class 4: Vibrio spp. super-integron

See Figure 1. a Abbreviations: Amp, ampicillin; Bleo, bleomycin; Cm, chloramphenicol; Ery, erythromycin; Fus, fusidic acid; Gen, gentamicin; Hg, mercury; Kan, kanamycin; Mino, minocycline; Spc, spectinomycin; Str, streptomycin; Strp, streptothricin; Sul, sulfonamide; Tet, tetracycline; Tmp, trimethoprim; Van, vancomycin. b See Figure 2.

pendently of the host chromosome and can be distinguished by their origins of replication. Multiple plasmids can exist within a single bacterium, where their genes add to the total genetics of the organism. Transposons are mobile genetic elements that can exist on plasmids or integrate into other transposons or the host’s chromosome. In general, these pieces of DNA contain terminal regions that participate in recombination and specify a protein(s) (e.g., transposase or recombinase) that facilitates incorporation into and from specific genomic regions. Conjugative transposons are unique in having qualities of plasmids and can facilitate the transfer of endogenous plasmids from one organism to another. Integrons contain collections of genes (gene cassettes) that are generally classified according to the sequence of the protein (integrase) that imparts the recombination function (Mazel, 2006) (Figure 2A). They have the ability to integrate stably into regions of other DNAs where they deliver, in a single exchange, multiple new genes, particularly for drug resistance. The super-integron, one which contains hundreds of gene cassettes (representing about ?3% of the host’s genome), is distinct from other integrons; it was first identified in Vibrio cholerae (Mazel et al., 1998). Although Alexander Fleming selected and described mutants resistant to penicillin soon after he had discovered the antibiotic, no one could have predicted the speed with which bacteria would acquire the capacity

for dealing with multiple antibacterial agents. Plasmids, specifying a collection of individual antibiotic resistance determinants (originally termed resistance factors, R factors), were initially described as the vehicle used to rapidly spread resistance traits among bacteria. Strains of Shigella bearing the self-replicating and self-transferable plasmids were easily selected and propagated during a period of considerable sulfonamide use in Japan after World War II (Watanabe, 1963). Antibiotics such as streptomycin, chloramphenicol, and tetracycline were subsequently introduced for the treatment of the sulphonamide-resistant organisms. Shigella and E. coli strains bearing resistance to all four agents, however, were recognized in 1955 (Watanabe, 1963). Now, more than 60 years after antibiotics were first introduced into clinical practice, the prescience of Fleming has come to fruition as the infectious disease community has yet to identify an antibiotic that has managed to circumvent the development of resistance. Resistance through Chromosomal Mutation Fluoroquinolones Virtually all clinically important fluoroquinolone resistance can be attributed to mutations within the drug’s targets, DNA gyrase and topoisomerase IV (Drlica and Malik, 2003). These multisubunit complexes perform critical ATP-dependent functions during DNA replication; each is comprised of two subunits: GyrA and GyrB for DNA Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1039

Figure 2. Transposable Elements that Confer Antibiotic Resistance (A) Organization of a class 1 integron. int1 specifies the integrase (the recombination protein), att1 is the primary recombination site, attC are the 59 base pair elements that are substrates of the integrase, and Pc is a common promoter that drives high-level expression of antibiotic resistance determinants (Abr) such as those for aminoglycosides, b-lactams, chloramphenicol, and trimethoprim inserted downstream of this element. qacE∆ and sul1 specify genes for resistance to quaternary ammonium compounds (partially deleted) and sulfonamides, respectively; these two genes are found specifically in class 1 integrons. (B) The vanA gene cluster (adapted from Courvalin, 2006). Orf1 and Orf2 impart the transposition features. VanR (response regulator) and VanS (histidine kinase) comprise a two-component system that regulates expression of vanHAXYZ. VanA (ligase) and VanH (dehydrogenase) are responsible for the synthesis of the modified depsipeptide (D-Ala-D-Lac, see text) while VanX (D,D-dipeptidase) and VanY (D,D-carboxypeptidase) cleave normal peptidoglycan substrates. VanZ is of unknown function. PR and PH are promoters that control expression of two separate transcriptional units, and the entire element is flanked by imperfect inverted repeats.

gyrase and ParC/GrlA and ParE/GrlB for topoisomerase IV. The GyrA and ParC/GrlA proteins contain the DNAbinding functions and are targeted by the fluoroquinolones (which are purely synthetic antibiotics), whereas GyrB and ParE/GrlB perform the roles of ATP binding and hydrolysis and are inhibited by coumarin antibiotics. In gram-negative bacteria, mutations in DNA gyrase occur first, whereas in gram-positive organisms mutations in topoisomerase IV arise initially in the stepwise movement to clinically relevant fluoroquinolone resistance. Mutations that give rise to fluoroquinolone resistance are predominantly found in the quinolone-resistancedetermining-region (QRDR) of GyrA and ParC/GrlA. Mutations occur more frequently at particular amino acids than at others. Multiple mutational events can be selected in a stepwise manner so as to “train” resistance in a bacterium; successive mutations have additive effects. Isolates that are highly fluoroquinolone resistant bear multiple QRDR mutations and generally other mechanisms of fluoroquinolone resistance, such as drug efflux (see below). Rifamycins Rifamycins are bactericidal drugs that arrest transcription by interacting with RpoB, the β subunit of RNA polymerase (RNAP). Although combination regimens 1040  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

that include rifampin or rifapentine and isoniazid, pyrazinamide, ethambutol, or streptomycin remain primary choices for front-line therapy of M. tuberculosis infection, resistance through point mutations in the rifampin-binding region of rpoB occur at a frequency of 1 × 10 −8 and are widespread. M. tuberculosis resistant to rifampin and isoniazid at a minimum are defined as multidrug resistant and impede successful therapy among patients in all regions of the world (Sharma and Mohan, 2006). Sulfonamide and Trimethoprim Antibiotics The sulfonamides, the first antimicrobials developed for large-scale introduction into clinical practice (in 1935), target dihydropteroate synthase. Their serendipitous discovery (the antibacterial activity was seen initially in vivo when the active compound was released as part of a dye) pales only in comparison with that of Fleming’s chance discovery of penicillin (Levy, 2002). Trimethoprim, introduced in 1968, inhibits dihydrofolate reductase and was the last structurally unique antibiotic approved prior to the release of linezolid in 2000. Mutations in the gene specifying dihydropteroate synthase decrease the enzyme’s affinity for the sulfonamides and have been found in laboratory experiments using E. coli and Streptococcus pneumoniae and in clinical isolates of Campylobacter jejuni and Haemophilus influenzae (Skold,

2001). Mutations in the chromosomal gene specifying dihydrofolate reductase can result in overexpression of an enzyme with a reduced affinity for trimethoprim and thereby offer very high-level trimethoprim resistance in E. coli and H. influenzae (Skold, 2001). Tetracycline, Aminoglycoside, and MLS Antibiotics Agents within the tetracycline, aminoglycoside and macrolide-lincosamide-streptogramin (MLS) classes of antibiotics target the ribosome to inhibit protein translation. As a consequence, resistance via chromosomal mutation is uncommon. Tetracyclines and aminoglycosides interact with 16S rRNA (rrs) and the macrolide-lincosamide-streptogramin (MLS) family binds to 23S rRNA (rrl). In most bacteria, multiple rrs and rrl operons are present and susceptibility mediated by any one of these targets can be dominant, making resistance difficult to attain without a mutation in all or a majority of the other operons. However, in organisms with low rRNA (rrn) copy numbers, chromosomal mutations conferring resistance have appeared. Tetracycline resistance via a point mutation in Propionibacterium acnes (3 rRNA operons) and Helicobacter pylori (2 copies of rrn) (Gerrits et al., 2002; Ross et al., 1998) has been documented. Mutations in rrs conferring resistance to amikacin and kanamycin and alterations in small ribosome protein S12 (rpsL) or rrs affecting streptomycin (all aminoglycoside drugs) susceptibility in clinical M. tuberculosis (1 rrn operon) have been described (Alangaden et al., 1998). The emergence of resistance to erythromycin (a macrolide) during therapy caused by mutant rrl in S. pneumoniae (4 copies of rrn) has been reported (Musher et al., 2002). Moreover, mutations in the large ribosome protein L4 (rplD) have also been shown to alter MLS susceptibility (Tait-Kamradt et al., 2000). The ketolides, recently introduced into clinical practice, were designed to circumvent macrolide resistance; nonetheless decreased susceptibility and resistance to telithromycin (a ketolide) have been found in S. pneumoniae with mutations in rrl, rplD, and large ribosome protein L22 (rplV) (Hisanaga et al., 2005). Mutations in L22 also affect quinupristin-dalfopristin (streptogramins that act individually in a bacteriostatic manner) susceptibility by affecting the synergistic relationship important to the combination’s bactericidal mechanism of action (references in Hershberger et al., 2004). Oxazolidinones Linezolid (another protein synthesis inhibitor) was approved recently as an agent to treat methicillin-resistant S. aureus and vancomycin-resistant enterococci infections. Its availability in both intravenous and oral formulations makes it useful for treating infections in both the hospital and community. Resistance to linezolid in laboratory studies has been linked to point mutations in rrl in S. aureus and E. faecalis. Now, clinical isolates of Staphylococcus epidermidis, S. aureus, Streptococcus oralis, Enterococcus faecium, and E. faecalis resistant to linezolid have been documented and many of the strains bear rrl mutations (Meka and Gold, 2004). As with the

fluoroquinolones, the level of resistance in S. aureus increases with mutations in multiple rrl alleles (Wilson et al., 2003) leading to clinically relevant resistance. Lipopeptides Daptomycin (a drug acting on bacterial membranes) was approved by the FDA in 2003 and while it has been successfully used for treating infections caused by methicillin-resistant S. aureus and other gram-positive bacteria, treatment failures have been noted (Hayden et al., 2005). Laboratory experiments have established that mutations in multiple chromosomal loci (i.e., mprF, yycG, rpoB, and rpoC) affect daptomycin susceptibility (Friedman et al., 2006). With regard to mprF, which specifies a lysylphosphatidylglycerol synthetase, similar mutations have been identified in nonsusceptible clinical isolates as well (Friedman et al., 2006). Genomic Duplications A mechanism for drug resistance common in eukaryotic cells (e.g., certain mammalian cancers and parasites) is gene amplification, which leads to overexpression of multidrug transporters and drug targets (Albertson, 2006). Recently, a large tandem duplication of the E. coli acrAB locus in a mutant isolated in the presence of tetracycline was found to overexpress the AcrAB drug efflux pump, producing an MDR phenotype (Nicoloff et al., 2006). The mutants, however, were unstable and reverted to the wild-type phenotype in the absence of drug (Nicoloff et al., 2007). Genomic amplification has also been shown to affect susceptibility to methicillin in S. aureus (Matthews and Stewart, 1988). It is expected that examples of gene duplication as a mechanism of resistance, not mutation, will increase in recognition among bacterial isolates. The likely phenotype will be an unstable form of resistance. Acquired Resistance: Enzymatic Drug Modification Enzymes that modify antibacterial drugs are divided into two general classes: those such as β-lactamases that degrade antibiotics and others (including the macrolide and aminoglycoside-modifying proteins) that perform chemical transformations. β-Lactam Antibiotics There are hundreds of β-lactamases that have been discovered and characterized, and a more complete overview of the individual enzymes within this protein family is available elsewhere (Jacoby and Munoz-Price, 2005). Most of these resistances are specified by genes located on plasmids and transposons; others are chromosomal and provide intrinsic resistance. β-lactamases are classified using schemes based on function (the system of Bush-Jacoby-Medeiros) or structure (Ambler classification) but in general are broadly separated into enzymes with a serine active site and those that require a metal ion cofactor (references in Jacoby and Munoz-Price, 2005). The classification scheme of Ambler divides the β-lactamases into four groups: class A, C, and D enzymes are proteins with a Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1041

serine residue at their active sites and the class B proteins are zinc-dependent metalloenzymes. Some class A proteins function as extended-spectrum β-lactamases and as carbapenemases. The class B metallo-enzymes that hydrolyze carbapenems are susceptible to inhibition by EDTA, whereas they are not susceptible to inhibition by clavulanate (a β-lactamase inhibitor). AmpC, an inducible and usually chromosomal enzyme found in many species of the Enterobacteriaceae and P. aeruginosa, is the prototype class C enzyme. Recent reports have described plasmid-borne ampC genes that can be transferred among E. coli, Klebsiella spp., and Salmonella spp. (references in (Jacoby and Munoz-Price, 2005). Class D enzymes have been found in only a few species such as P. aeruginosa, Acinetobacter spp., and Aeromonas spp. Aminoglycosides A large number of aminoglycoside-modifying enzymes specified by genes on transferable elements are widespread in clinically relevant organisms (Davies and Wright, 1997). Here, resistance is accomplished with proteins that N-acetylate (acetyltransferases), phosphorylate (phosphotransferases), and adenylate (nucleotidyltransferases) the aminoglycosides. The acetyltransferases are capable of modifying tobramycin, gentamicin, netilmicin, and amikacin; the nucleotidyltransferase proteins alter the activity of tobramycin; and the phosphotransferases affect amikacin susceptibility. Many of the aminoglycoside-modifying enzymes are found on integrons and other mobile genetic elements where they associate with additional resistance determinants. For example, three acetyltransferase genes were found on a class 1 integron from P. aeruginosa that also specified resistance to carbapenems and sulfonamides (Poirel et al., 2001). MLS Antibiotics There are a number of inactivating enzymes that act on the MLS antibiotics. The genes encode esterases, hydrolases, glycosylases, phosphotransferases, nucleotidyltransferases, and acetyltransferases and are found less frequently than efflux and ribosome-modifying genes in clinical isolates (Weisblum, 1998). Esterases act on 14(e.g., erythromycin) and 15- (e.g., azithromycin) membered macrolides; the hydrolases affect streptogramin B drugs; the acetyltransferases inactivate streptogramin A antibiotics; and the nucleotidyltransferases confer resistance to the lincosamides (e.g., clindamycin). Phosphotransferases modify 14-, 15-, and 16-membered macrolides with varying specificities and telithromycin (Matsuoka and Sasaki, 2004). Chloramphenicol Acetyltransferases that inactivate chloramphenicol (a generally bacteriostatic inhibitor of protein synthesis) are the most common resistance mechanisms for this antibiotic and are separated into type A and B proteins (Schwarz et al., 2004). Both types of enzymes function as homotrimers but are unrelated based on amino acid sequence analyses. The type B enzymes are also 1042  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

termed xenobiotic acetyltransferases and appear to share an evolutionary lineage that includes some streptogramin-inactivating enzymes found in the enterococci and staphylococci. Constitutive expression as well as translational attenuation underlie the regulation of type A and B proteins. Tetracyclines Recently, a flavin-dependent monooxygenase, designated tet(X), originally identified in Bacteroides fragilis, that acts on older tetracyclines (such as tetracycline, oxytetracycline, and chlortetracycline) as well as newer compounds (such as doxycycline, minocycline, and tigecycline) has been characterized (Moore et al., 2005). The enzyme catalyzes regioselective hydroxylation to inactivate its substrates, but the modification products produced by TetX are not stable at physiological pH. The tet(X) determinant has not yet been found in common, clinically relevant tetracycline-resistant isolates, although a related tet(X)-like gene in P. aeruginosa has been reported (A. Nakamura et al., 2002, paper presented at Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, CA). Altered, Substituted, and Protected Drug Targets β-Lactam Antibiotics The first penicillin-resistant S. aureus identified in the mid-1940s expressed a β-lactamase (termed PCI) that afforded resistance. A penicillin derivative, methicillin, that was resistant to this β-lactamase was subsequently introduced in 1959 to treat the penicillin-resistant isolates, but methicillin-resistant S. aureus were identified shortly thereafter (Barber, 1961). Here, the β-lactam resistance was linked to the acquisition of a gene for an altered PBP. Resistance in the staphylococci (Fuda et al., 2005) and streptococci (Jacobs, 1999) frequently occurs following the acquisition of genes for PBPs that are not sensitive to β-lactam inhibition. The altered PBP of methicillin-resistant S. aureus, PBP2a, is specified by mecA and is transported on a mobile genetic element called the “staphylococcal cassette chromosome” (SCCmec) (Fuda et al., 2005). In addition to mecA, SCCmec contains the mecR1-mecI regulatory loci and encodes the enzymes that are involved in site-specific recombination. Under normal circumstances, S. aureus uses multiple PBPs during cell wall biosynthesis. One, PBP2, is a bifunctional enzyme with transpeptidase and transglycosylase activities. When methicillin-resistant S. aureus is exposed to methicillin, PBP2 functions as the transglycosylase whereas PBP2a contributes the transpeptidase activity in order to confer resistance to nearly all β-lactam antibiotics. Removal of the transglycosylase function of PBP2 imparts β-lactam susceptibility and demonstrates the importance of both attributes of the enzyme (reviewed in Fuda et al., 2005). The new strains of methicillin-resistant S. aureus that have emerged recently outside of hospitals in the community share some features with hospital-associated

strains (Naimi et al., 2003). These include the nearly complete β-lactam resistance phenotype and the ability to cause a range of serious life-threatening infections. Although many community-associated strains are multidrug susceptible (except to β-lactam antibiotics), the genome of an epidemic clone distributed throughout the US (USA300) contains chromosomal mutations that confer fluoroquinolone resistance and plasmids that mediate resistance to tetracycline, erythromycin, clindamycin, streptogramin B agents, and mupirocin (Diep et al., 2006). Glycopeptides Glycopeptides (including vancomycin and teicoplanin) interact with peptidoglycan precursors to exert bactericidal activity. Resistance to the glycopeptides in grampositive cocci is another illustration of an altered drug target (Courvalin, 2006). In the enterococci, acquired glycopeptide resistance is attributable to VanA, B, D, E, and G phenotypes whereas VanC affords intrinsic resistance. VanA and VanD provide resistance to both vancomycin and teicoplanin, whereas the others provide resistance to vancomycin alone. The resistance phenotype is accomplished using multiple proteins specified in gene clusters and each result in the production of a modified peptidoglycan. Of the many drug-resistance determinants currently known, the determinant for glycopeptide resistance is probably the most complex (Figure 2B). The activity of many enzymes specified by the gene cluster are involved in conferring glycopeptide resistance (Figure 2B). Either a racemase or dehydrogenase results in the production of serine (VanC, E, or G) or lactate from pyruvate (VanA, B, or D), which a ligase uses to form a C-terminal D-Ala-D-Ser or D-Ala-D-Lac in the altered peptidoglycan. A two-component regulatory system controls expression of the biosynthetic machinery. Although the glycopeptides have a lower affinity for the D-Ala-D-Ser or D-Ala-D-Lac substrates, they can still bind and inhibit peptidoglycan biosynthesis if an intact D-Ala target remains. Two additional enzymes (or a single bi-functional protein for VanC) complete the phenotype by removing the antibiotic’s normal target: a dipeptidase cleaves the C-terminal D-Ala-D-Ala and a carboxypeptidase provides a redundant function of removing the terminal D-Ala in the absence of, or under circumstances when, the dipeptidase is less active. Recently, the enterococcal vanA gene cluster has made its way into methicillin-resistant S. aureus, imparting full-blown vancomycin resistance on this prominent pathogen (Weigel et al., 2003), although this molecular event had been demonstrated in laboratory experiments years before (Noble et al., 1992). Vancomycin-resistant S. aureus have now been identified in patients from three different US locales, Michigan, Pennsylvania, and New York. In all instances, the resistance determinant resides on a plasmid-specified transposon, Tn1546. Vancomycin-resistant E. faecalis and methicillin-resistant S. aureus were also obtained from the Michigan patient

that bore vancomycin-resistant S. aureus and each contained plasmids that were identical, except for the presence of Tn1546 in the isolate of vancomycin-resistant E. faecalis. It is assumed that the plasmid from E. faecalis was the vehicle for vanA entry into S. aureus and that the vanA gene cluster was subsequently transferred by means of transposition into the S. aureus plasmid. Tetracyclines and MLS Antibiotics The most prevalent forms of resistance to the tetracyclines in the clinic are drug efflux (see below) and ribosome protection. Determinants of ribosome protection (Connell et al., 2003) have sequence similarity to bacterial elongation factors (EF-G an EF-Tu). They also possess GTPase activity and function to facilitate release of tetracycline from the ribosome in an energy-dependent manner (Connell et al., 2003). Bacteria that exhibit resistance to minocycline generally contain a ribosome protection determinant. In Megasphaera elsdenii (a rumen bacterium), a mosaic gene containing two unique ribosome protection determinants (TetO and TetW) has been described (Stanton and Humphrey, 2003). Moreover, the tet(P) determinant of Clostridium perfringens specifies both ribosome protection and efflux mechanisms in an overlapping genetic unit (Sloan et al., 1994). Binding of drugs within the MLSK (macrolide-lincosamide-streptogramin-ketolide) family to 23S rRNA is affected by erm (erythromycin resistance methylase or erythromycin ribosome methylation) specified gene products (Zhanel et al., 2001). These mechanisms represent the predominant macrolide resistance mechanisms in Europe and South Africa. Currently, there are 34 different classes of Erm proteins (http://faculty.­washington. edu/marilynr/), and each functions by methylating a single adenine residue of the E. coli 23S rRNA (at position A2058). Methylation results in the MLSB phenotype, which encompasses resistance to 14-, 15-, and 16-membered macrolides, lincosamides, and streptogramin B drugs. Some of the resistance determinants are induced following exposure to MLS drugs, whereas others are constitutively synthesized. In the staphylococci, agents like erythromycin and azithromycin induce erm expression, but 16-membered macrolides do not (Zhanel et al., 2001). Constitutive erm expression in some clinical and laboratory strains confers telithromycin resistance (Hisanaga et al., 2005). Sulfonamide and Trimethoprim Antibiotics The activities of the sulfonamides and trimethoprim are also affected by acquired genes specifying enzymes that are insensitive to drug inhibition (Skold, 2001). sul1 and sul2 are the main determinants of clinical resistance to sulfonamide, whereas sul3 was found to be prevalent in farm animals (Perreten and Boerlin, 2003). In contrast, more than 20 trimethoprim resistance determinants (numbered chronologically from dfr1) are documented. The genes specifying the sulfonamide-insensitive dihydropteroate synthases are present on class 1 integrons (sul1) or plasmids (sul2), whereas dfr variants (with dfr1 Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1043

Figure 3. Antibiotic Efflux Systems For simplicity, the illustration represents a gram-negative bacterium. Some substrates of AcrAB (e.g., aminoglycosides) are presumably recognized by a portion of AcrB that faces the cytoplasm while others (e.g., tetracycline) migrate thorough the inner membrane to an AcrB-binding pocket(s) (Murakami and Yamaguchi, 2003). The E. coli AcrAB-TolC and Tet efflux systems are secondary transporters that use proton motive force (H+) while NorM (Vibrio spp.) is a secondary transporter that exploits a gradient of sodium ions (Na+) to drive efflux (Li and Nikaido, 2004). MacAB is a primary transporter that uses the energy derived from ATP hydrolysis to drive efflux (Li and Nikaido, 2004). Single-component systems (e.g., NorM, Tet) transport substrates across the inner membrane where they then diffuse through porins (e.g., OmpF) or the outer membrane into the extracellular medium (Li and Nikaido, 2004). Multicomponent systems like AcrAB-TolC and MacAB-TolC direct the transport of substrates to the extracellular medium. Abbreviations: LPS, lipopolysaccharide; Ab, antibiotic; Tc, tetracycline; Mac, macrolide.

being the most common in gram-negative bacteria) move from organism to organism on class 1 and 2 integrons. dfr1 is present on the Tn7 transposon, thereby facilitating its integration into the E. coli chromosome. Fluoroquinolones The plasmid-specified qnr determinants provide an unusual mechanism of decreased fluoroquinolone susceptibility. Variants of the qnr element, first identified in K. pneumoniae (Martinez-Martinez et al., 1998), have been found in E. coli, Enterobacter cloacae, Providencia stuartii, Citrobacter freundii, C. koseri, Shigella flexneri 2b, and non-Typhi Salmonella enterica. Qnr belongs to the pentapeptide repeat family of proteins and mediates resistance presumably by protecting DNA gyrase and topoisomerase IV from the inhibitory action of the fluoroquinolones (Tran and Jacoby, 2002). Another fluoroquinolone resistance mechanism, MfpA, has been found in M. tuberculosis, which contains a single type II topoisomerase (Hegde et al., 2005). It, too, is a member of the pentapeptide repeat family of proteins but also acts as an inhibitor of DNA gyrase from M. tuberculosis by virtue of its ability to mimic the structure of B-form DNA. The interaction of MfpA with DNA gyrase likely interferes with the inhibitory action of agents like ciprofloxacin. By themselves, Qnr and MfpA confer only low levels of fluoroquinolone resistance but can augment resistance when coupled with other mechanisms (see below). Moreover, selection for fluoroquinolone resistance may be favored in organisms bearing Qnr or MfpA if they survive the presence of low fluoroquinolone concentrations. Efflux Systems, Porins, and Altered Membranes Efflux as a mechanism of antibiotic resistance was originally described for tetracyclines (McMurry et al., 1980) and is now well-appreciated and commonly found. Most drug efflux proteins belong to five distinct protein families: the resistance-nodulation-cell division (RND), major facilitator (MF), staphylococcal/ small multidrug resistance (SMR), ATP-binding cas1044  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

sette (ABC), and multidrug and toxic compound extrusion (MATE) families (Li and Nikaido, 2004). Efflux by proteins in the RND, SMR, MF, and MATE families is driven by proton (and sodium) motive force and is therefore called secondary transport (Figure 3). ATP hydrolysis drives efflux in the primary (ABC) transporters (Figure 3). Efflux proteins also come in two general types. Some, like the tetracycline (Tet) (Levy, 1992) and macrolide (Mef) (Zhanel et al., 2001) transporters, are single component efflux systems that have narrow substrate profiles, acting on a few agents or multiple agents within the same drug class. Others, such as the RND family members, require two additional proteins to confer resistance but are capable of binding multiple structurally unrelated compounds and confer broad resistance phenotypes (Li and Nikaido, 2004). The organizations of the RND-based efflux systems, which are found in a number of gram-negative bacteria, allow them to transport drugs from the cytoplasm and across the inner and outer membranes of the cell envelope (Figure 3). Tetracyclines There are now more than 20 different tetracycline efflux proteins that have been allocated to six different groups (Sapunaric et al., 2005). The proteins contain either 12(e.g., TetA-E in gram-negative bacteria) or 14- (e.g., TetK and TetL in gram-positive organisms) putative transmembrane-spanning segments. Expression of proteins within group 1 is controlled by a transcription repressor such as TetR. The antibiotic inactivates the repressor and allows expression of the tetracycline efflux systems. Alternatively, synthesis of TetK and TetL is inducible by tetracyclines through mechanisms that involve translation attenuation and reinitiation, respectively. Also, the mechanism of tetracycline efflux is not conserved among all members; some, such as Tet(B), engage in electroneutral reactions (that is, one tetracycline-Mg chelate exits the cell as a proton enters whereas others, such as TetL, accomplish efflux in an electrogenic manner [references in Sapunaric et al., 2005]).

MLS Antibiotics MsrA (macrolide-streptogramin resistance) is an ABC efflux protein that produces resistance to 14- and 15membered macrolides and streptogramin B antibiotics in the streptococci and staphylococci, but susceptibility to clindamycin is not affected by this system (Li and Nikaido, 2004). In the staphylococci, relatives of MsrA (VgaA and VgaB) are specified on plasmids. VgaA confers resistance to streptogramin A antibiotics and lincosamides, and VgaB affects pristinamycin (a mixture of streptogramin A and B antibiotics) susceptibility (Chesneau et al., 2005). Alternatively, the Mef efflux transporters, which are the predominant macrolide resistance proteins in the United States, effectively handle 14- and 15-membered macrolides in the streptococci, but strains that express the proteins are susceptible to 16-membered macrolides, lincosamides, and streptogramin B drugs (Li and Nikaido, 2004). E. faecalis are naturally resistant to quinupristin-dalfopristin and this characteristic is attributed to the expression of lsa, which specifies a streptogramin efflux protein belonging to the ABC family, as mutational inactivation of Lsa confers susceptibility to quinupristin-dalfopristin in this organism (references in Hershberger et al., 2004). This scenario has been observed with other (MDR) efflux proteins where removal of an efflux system renders an organism susceptible to antibiotics, even in the presence of chromosomal mutations that reduce binding of the drug to its target and would otherwise confer clinical resistance (Lomovskaya et al., 1999; Oethinger et al., 2000). Phenicols Chloramphenicol and florfenicol are broad-spectrum agents that have utility in human clinical practice (chloramphenicol) and animal husbandry (florfenicol). Because of toxicity (aplastic anemia) seen with the former, the current use of chloramphenicol has been limited to the treatment of some severe life-threatening infections such as bacterial meningitis in patients with allergies to the penicillins. Efflux proteins specific for the phenicols have been described in a number of clinically important bacteria and are classified into eight different groups (E-1 through E-8) (Schwarz et al., 2004). In general, these proteins provide levels of resistance greater than that of the multidrug efflux proteins described below, and members of groups E-3 and E-4 confer resistance to both phenicols. For some chloramphenicol-specific efflux proteins like CmlA, expression of the resistance determinant is mediated through an inducible mechanism of translational attenuation similar to that used for particular chloramphenicol acetyltransferase resistance genes (see above). Multidrug Resistance Efflux Systems Historically, the gram-negative cell envelope was thought to affect antibiotic susceptibility by greatly restricting drug penetration (Li and Nikaido, 2004). Contemporary studies, however, have shown that most antibacterial agents effectively penetrate gram-negative organisms (Li and Nikaido, 2004) but fail to reach intracellular tar-

gets because of efflux (Levy, 1992). In gram-negative bacteria, including E. coli and nonfermenting organisms such as P. aeruginosa, Acinetobacter spp., Stenotrophomonas maltophilia, and Burkholderia cepacia, “intrinsic” resistance is attributed to the expression of the RND efflux system(s) (Figure 3). This mechanism is an effective means for dealing with different antibiotic classes using a single resistance determinant. The natural function of the E. coli AcrAB efflux system is thought to have evolved to protect the cell from the inhibitory activity of toxic substances such as bile salts (Li and Nikaido, 2004). Other related systems function similarly in Neisseria gonorrhoeae (Hagman et al., 1995) and export molecules involved in quorum sensing in P. aeruginosa (Pesci et al., 1999). Tigecycline, which gained FDA approval in 2005, has poor activity against P. aeruginosa, Proteus mirabilis, Morganella morganii, and Klebsiella pneumonia; this is attributed to RND efflux systems (references in Stein and Craig, 2006). Elimination of an AcrA ortholog in M. morganii, MexXY-OprM in P. aeruginosa, and an AcrB ortholog in P. mirabilis increased susceptibility to tigecycline 16- to 133-fold (references in Stein and Craig, 2006), whereas deletion in E. coli of AcrAB and AcrEF had a more modest (4-fold) effect (Hirata et al., 2004). Bacillus subtilis Bmr (Bacillus multidrug resistance) (Neyfakh et al., 1991) and S. aureus Qac (quaternary ammonium compound) (Tennent et al., 1989) are two MDR efflux proteins (members of the MF superfamily) that were first characterized in gram-positive cells. Like many members of the RND family in gram-negative bacteria, Bmr is constitutively expressed and therefore engenders intrinsic resistance to chloramphenicol and fluoroquinolones. Another MDR efflux pump from B. subtilis, Blt, includes spermidine among its list of substrates. It is now thought that the natural function of Blt is to facilitate the removal of polyamines from the cell (Woolridge et al., 1997). The staphylococcal Qac systems provide resistance to antiseptics and disinfectants (e.g., quaternary ammonium compounds, chlorhexidine, and diamidines). Unlike most other MDR efflux proteins, these are specified on plasmids, a feature that facilitates their dissemination. Permeability As alluded to previously, the outer membrane of the gram-negative cell envelope is a barrier to both hydrophobic and hydrophilic compounds. In order to circumvent this permeability barrier, these organisms have evolved porin proteins (e.g., OmpF in E. coli and OprD in P. aeruginosa) that function as “nonspecific” entry and exit points for antibiotics and other small-molecule organic chemicals. Imipenem (and to a lesser extent meropenem) and basic amino acids pass through OprD; mutations that decrease expression of the porin contribute to clinical imipenem resistance. Moreover, the expression of OprD and the MexEF-OprM efflux system is coregulated (Ochs et al., 1999), and in this manner resistance to the carbapenems and other substrates of Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1045

Table 3. MDR Efflux Systems of Clinically Important Bacteria Bacterial Organism

Efflux System(s)

Representative Antibiotic Resistancea

P. aeruginosa

MExAB-OprM

BLA and FQ

MexCD-OprJ

4th gen ceph

MexEF-OprN

FQ, Cm, Tmp, and Tri

MexHI-OprD

EtBr, Nor, and Acr

MexJK-OprM

Cip, Tet, Ery, and Tri

MexVW-OprM

FQ, Cm, Tet, Ery, EtBr, and Acr

MexXY-OprM

AG and Tig

A. baumannii

AdeABC

AG, FQ, TET, Ctx, Cm, Ery, and Tmp

S. maltophilia

SmeABC

AG, BLA, and FQ

SmeDEF

MC, TET, FQ, CAR, Cm, and Ery

B. cepacia

CeoAB-OpcM

Cm, Cip, and Tmp

B. pseudomallei

AmrAB-AprA

MAC and AG

E. coli

AcrAB-TolC

FQ, BLA, TET, Cm, Acr, Tri

K. pneumoniae

AcrAB-TolC

FQ, BLA, TET, and Cm

S. aureus

MepA

Tig, Mino, Tet, Cip, Nor EtBr, and TPP

E. faecalis

EmeA

Nor, EtBr, Clind, Ery, and Nov

Lsa

Clind and QD

PmrA

FQ, Acr, and EtBr

S. pneumoniae

Abbreviations: 4th gen ceph, fourth-generation cephalosporins; Acr, acriflavin; AG, aminoglycosides; BLA, β-lactams; CAR, carbapenems; Cip, ciprofloxacin; Clind, clindamycin; Cm, chloramphenicol; Ctx, cefotaxime; Ery, erythromycin; EtBr, ethidium bromide; FQ, fluoroquinolones; MC, macrolides; Mino, minocycline; Nor, norfloxacin; Nov, novobiocin; QD, quinupristin-dalfopristin; TET, tetracycline family; Tet, tetracycline; Tig, tigecycline; Tmp, trimethoprim; TPP, tetraphenylphosphonium bromide; Tri, triclosan. a

MexEF-OprM (Table 3) can develop in mutants where the expression of OprD and the efflux pump has been altered. In 1996, Hiramatsu and colleagues identified clinical S. aureus isolates in Japan that exhibited an intermediate level of resistance to vancomycin (the vancomycin intermediate-resistant S. aureus [VISA] strains) (Hiramatsu et al., 1997), and shortly thereafter additional organisms with a glycopeptide-insensitive phenotype appeared in the United States (the GISA isolates). A prominent feature of the VISA isolates is the presence of a thickened cell wall. It is presumed that this property traps vancomycin and prevents the antibiotic from reaching its target (Cui et al., 2006a). More recently, the same group performed a preliminary characterization of the VISA isolates and found reduced susceptibility to daptomycin (Cui et al., 2006b). Although additional molecular and biochemical characterizations are certainly needed, the findings do carry a cautionary note. Mutations that offer resistance to polymyxin B in P. aeruginosa presumably involve changes in the bacterial cell envelope that do not involve porins (references in Falagas and Kasiakou, 2005). In Salmonella enterica serovar Typhimurium, PmrAB (a two-component regulatory system) regulates resistance to polymyxin by modi-

1046  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

fying lipopolysaccharide and lipid A (references in Falagas and Kasiakou, 2005). The RosAB efflux system of Yersinia enterocolitica also affects susceptibility to polymyxin B (references in Falagas and Kasiakou, 2005). Regulatory Genes in Multidrug Resistance Chromosomal loci specifying global regulatory proteins that control MDR have been characterized in bacteria including E. coli, K. pneumoniae, and other Enterobacteriaceae, Neisseria gonorrhoeae, Bacillus subtilis, and S. aureus (Grkovic et al., 2002). In most instances, the regulatory proteins act as activators or repressors of transcription and contain either a single DNA-binding domain or two separate regions responsible for DNA binding and interacting with small-molecule substrates (Schumacher and Brennan, 2002). In the case of E. coli MarA, the protein regulates the expression of a large number of other chromosome genes (the Mar regulon), including those specifying an MDR efflux pump and other proteins (e.g., porins) that mediate antibiotic susceptibility (Barbosa and Levy, 2000). In a clinical K. pneumoniae isolate exhibiting reduced susceptibility to tigecycline and an MDR phenotype, transposon mutagenesis was used to find that inactivation of ramA, which specifies a MarA ortholog, produced a 16-fold increase in suscep-

tibility to tigecycline and abrogated the MDR phenotype (Ruzin et al., 2005). Although other clinical isolates that overexpress MarA or a MarA ortholog and exhibit a MDR phenotype have been described previously (Schneiders et al., 2003 and references therein), the studies that involved K. pneumoniae offer further proof that chromosomal transcription factors that regulate MDR efflux systems can produce clinical drug resistance. Single Determinants of Multidrug Resistance Except for the case of MDR efflux pumps, a single resistance mechanism commonly affords protection against one antibiotic or, at the most, drugs within the same general class, e.g., PBP2 of MRSA. Still, resistance determinants that specify erythromycin (Erm) methyltransferases in a variety of pathogens are single proteins that give rise to macrolide, lincosamide, and type B streptogramin resistance: structurally unique agents that share a common target and mechanism of action (Roberts, 2004). A mutant aminoglycoside acetyltransferase (specified by aac(6’)-Ib-cr) that gave rise to aminoglycoside (amikacin, kanamycin, and tobramycin) and fluoroquinolone (ciprofloxacin and norfloxacin) resistance was identified recently (Robicsek et al., 2006). Although the level of resistance conferred was low, aac(6’)-Ib-cr was located on a plasmid bearing another unusual mechanism (the qnr determinant [see above]) of fluoroquinolone resistance, and the two determinants together functioned in an additive manner to yield clinical levels of fluoroquinolone resistance. Reduced susceptibility to the macrolides, chloramphenicol, and linezolid in clinical S. pneumoniae isolates has been attributed to mutations in large ribosomal protein (L4) (Wolter et al., 2005). Although mutations in the genes that specify 23S rRNA, L4, or L22 commonly lead to macrolide resistance, susceptibility to chloramphenicol frequently occurs following the acquisition of a modifying enzyme (see above). For linezolid, previous resistance in the enterococci and S. aureus was attributed solely to mutations in the locus that encodes 23S rRNA (see above). What the Past Can Teach Us about the Future More than half a century has passed since the first antibiotics were introduced commercially. It did not take long for microbes to develop “drug fastness” to the original “magic bullets,” and widespread use of many antibacterial drugs provides ideal conditions for the spread of MDR organisms. Many early researchers were expertly skilled in identifying the means used to evade the inhibitory effects of these drugs—from investigators such as Esther and Joshua Lederberg, who characterized the random nature of mutational events conferring resistance to streptomycin, to others such as Tsutomu Watanabe, who surmised that mutation alone would not be sufficient for explaining the MDR phenotype. Resistance transfer factors (RTFs), later called resistance (R)

factors and then plasmids, would provide the basis for multiple drug resistance caused by “infective heredity” (reviewed in Watanabe, 1963). For bacteria, there is more than one way to evade a drug class, and organisms unresponsive to all antimicrobial agents, even ones not encountered previously, are now familiar. Microbes have means for collecting single resistance traits. Theoretically, one needs only a single (broad-spectrum) agent that effectively deals with resistances affecting just one class of drugs. The MDR efflux pumps that render multiple classes ineffective, however, have greatly confounded efforts in finding new agents. Multidrug resistance is a worldwide problem that does not obey international borders and can indiscriminately affect members of all socioeconomic classes. A slew of epidemiological studies have documented the clonal, worldwide spread of infectious disease entities such as MDR S. pneumoniae from specific regions of the world. Also, the use of antibiotics in animal husbandry has been shown to foster human colonization with drugresistant microbes (Levy et al., 1976). There are compelling data that link use of agents like virginiamycin (a streptogramin) and avoparcin (a glycopeptide) in animal husbandry to clinical drug-resistant strains. Bacteria adopt intricate strategies to avoid the lethal effects of antibiotics. Awareness of these mechanisms of resistance can help in the design of new drugs. As we face this critical problem, we need to be aware of the fluidity of the microbial genome and the relative ease with which resistance can emerge by mutation or gene acquisition. Recognizing the potential for the emergence of resistance should garner newfound respect for the discovery of new agents so urgently needed to cure infectious diseases. Acknowledgments S.B.L. is supported by the US Public Health Service (National Institutes of Health). The authors thank Bonnie Marshall for her help in the preparation of the manuscript and figures. References Alangaden, G.J., Kreiswirth, B.N., Aouad, A., Khetarpal, M., Igno, F.R., Moghazeh, S.L., Manavathu, E.K., and Lerner, S.A. (1998). Mechanism of resistance to amikacin and kanamycin in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 42, 1295–1297. Albertson, D.G. (2006). Gene amplification in cancer. Trends Genet. 22, 447–455. Alekshun, M.N., and Levy, S.B. (2000). Bacterial drug resistance: response to survival threats. In Bacterial Stress Responses, G. Storz, and R. Hengge-Aronis, eds. (Washington, D.C.: ASM Press), pp. 323-366. Barber, M. (1947). Staphylococci infection due to penicillin-resistant strains. BMJ 2, 863–865. Barber, M. (1961). Methicillin-resistant staphylococci. J. Clin. Pathol. 14, 385–393. Barbosa, T.M., and Levy, S.B. (2000). Differential expression of over

Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1047

60 chromosomal genes in Escherichia coli by constitutive expression of MarA. J. Bacteriol. 182, 3467–3474. Chesneau, O., Ligeret, H., Hosan-Aghaie, N., Morvan, A., and Dassa, E. (2005). Molecular analysis of resistance to streptogramin A compounds conferred by the Vga proteins of staphylococci. Antimicrob. Agents Chemother. 49, 973–980. Connell, S.R., Tracz, D.M., Nierhaus, K.H., and Taylor, D.E. (2003). Ribosomal protection proteins and their mechanism of tetracycline resistance. Antimicrob. Agents Chemother. 47, 3675–3681. Cook, M., Molto, E., and Anderson, C. (1989). Fluorochrome labeling in Roman period skeletons from Dakhleh Oasis, Egypt. Am. J. Phys. Anthropol. 80, 137–143. Courvalin, P. (1994). Transfer of antibiotic resistance genes between gram-positive and gram-negative bacteria. Antimicrob. Agents Chemother. 38, 1447–1451. Courvalin, P. (2006). Vancomycin resistance in gram-positive cocci. Clin. Infect. Dis. 42 (Suppl 1), S25–S34. Cui, L., Iwamoto, A., Lian, J.Q., Neoh, H.M., Maruyama, T., Horikawa, Y., and Hiramatsu, K. (2006a). Novel mechanism of antibiotic resistance originating in vancomycin-intermediate Staphylococcus aureus. Antimicrob. Agents Chemother. 50, 428–438. Cui, L., Tominaga, E., Neoh, H.M., and Hiramatsu, K. (2006b). Correlation between reduced daptomycin susceptibility and vancomycin resistance in vancomycin-intermediate Staphylococcus aureus. Antimicrob. Agents Chemother. 50, 1079–1082. Davies, J., and Wright, G.D. (1997). Bacterial resistance to aminoglycoside antibiotics. Trends Microbiol. 5, 234–240. D’Costa, V.M., McGrann, K.M., Hughes, D.W., and Wright, G.D. (2006). Sampling the antibiotic resistome. Science 311, 374–377. Diep, B.A., Gill, S.R., Chang, R.F., Phan, T.H., Chen, J.H., Davidson, M.G., Lin, F., Lin, J., Carleton, H.A., Mongodin, E.F., et al. (2006). Complete genome sequence of USA300, an epidemic clone of community-acquired methicillin-resistant Staphylococcus aureus. Lancet 367, 731–739.

Hegde, S.S., Vetting, M.W., Roderick, S.L., Mitchenall, L.A., Maxwell, A., Takiff, H.E., and Blanchard, J.S. (2005). A fluoroquinolone resistance protein from Mycobacterium tuberculosis that mimics DNA. Science 308, 1480–1483. Hershberger, E., Donabedian, S., Konstantinou, K., and Zervos, M.J. (2004). Quinupristin-dalfopristin resistance in gram-positive bacteria: mechanism of resistance and epidemiology. Clin. Infect. Dis. 38, 92–98. Hiramatsu, K., Aritaka, N., Hanaki, H., Kawasaki, S., Hosoda, Y., Hori, S., Fukuchi, Y., and Kobayashi, I. (1997). Dissemination in Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin. Lancet 350, 1670–1673. Hirata, T., Saito, A., Nishino, K., Tamura, N., and Yamaguchi, A. (2004). Effects of efflux transporter genes on susceptibility of Escherichia coli to tigecycline (GAR-936). Antimicrob. Agents Chemother. 48, 2179–2184. Hisanaga, T., Hoban, D.J., and Zhanel, G.G. (2005). Mechanisms of resistance to telithromycin in Streptococcus pneumoniae. J. Antimicrob. Chemother. 56, 447–450. Jacobs, M.R. (1999). Drug-resistant Streptococcus pneumoniae: rational antibiotic choices. Am. J. Med. 106, 19S–25S. Jacoby, G.A., and Munoz-Price, L.S. (2005). The new beta-lactamases. N. Engl. J. Med. 352, 380–391. Levin, A.S., Barone, A.A., Penco, J., Santos, M.V., Marinho, I.S., Arruda, E.A., Manrique, E.I., and Costa, S.F. (1999). Intravenous colistin as therapy for nosocomial infections caused by multidrugresistant Pseudomonas aeruginosa and Acinetobacter baumannii. Clin. Infect. Dis. 28, 1008–1011. Levy, S.B. (1992). Active efflux mechanisms for antimicrobial resistance. Antimicrob. Agents Chemother. 36, 695–703. Levy, S.B. (2002). The Antibiotic Paradox: How the Misuse of Antibiotics Destroys Their Curative Powers (Cambridge, MA: Perseus Publishing).

Drlica, K., and Malik, M. (2003). Fluoroquinolones: action and resistance. Curr. Top. Med. Chem. 3, 249–282.

Levy, S.B., and Marshall, B. (2004). Antibacterial resistance worldwide: causes, challenges and responses. Nat. Med. 10, S122– S129.

Falagas, M.E., and Kasiakou, S.K. (2005). Colistin: the revival of polymyxins for the management of multidrug-resistant gram-negative bacterial infections. Clin. Infect. Dis. 40, 1333–1341.

Levy, S.B., Fitzgerald, G.B., and Macone, A.B. (1976). Spread of antibiotic-resistance plasmids from chicken to chicken and from chicken to man. Nature 260, 40–42.

Friedman, L., Alder, J.D., and Silverman, J.A. (2006). Genetic changes that correlate with reduced susceptibility to daptomycin in Staphylococcus aureus. Antimicrob. Agents Chemother. 50, 2137–2145.

Li, X.Z., and Nikaido, H. (2004). Efflux-mediated drug resistance in bacteria. Drugs 64, 159–204.

Fuda, C.C., Fisher, J.F., and Mobashery, S. (2005). Beta-lactam resistance in Staphylococcus aureus: the adaptive resistance of a plastic genome. Cell. Mol. Life Sci. 62, 2617–2633. Gerrits, M.M., de Zoete, M.R., Arents, N.L., Kuipers, E.J., and Kusters, J.G. (2002). 16S rRNA mutation-mediated tetracycline resistance in Helicobacter pylori. Antimicrob. Agents Chemother. 46, 2996–3000. Grkovic, S., Brown, M.H., and Skurray, R.A. (2002). Regulation of bacterial drug export systems. Microbiol. Mol. Biol. Rev. 66, 671– 701. Hagman, K.E., Pan, W., Spratt, B.G., Balthazar, J.T., Judd, R.C., and Shafer, W.M. (1995). Resistance of Neisseria gonorrhoeae to antimicrobial hydrophobic agents is modulated by the mtrRCDE efflux system. Microbiol. 141, 611–622. Hayden, M.K., Rezai, K., Hayes, R.A., Lolans, K., Quinn, J.P., and Weinstein, R.A. (2005). Development of daptomycin resistance in vivo in methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol. 43, 5285–5287.

1048  Cell 128, March 23, 2007 ©2007 Elsevier Inc.

Lomovskaya, O., Lee, A., Hoshino, K., Ishida, H., Mistry, A., Warren, M.S., Boyer, E., Chamberland, S., and Lee, V.J. (1999). Use of a genetic approach to evaluate the consequences of inhibition of efflux pumps in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 43, 1340–1346. Martinez-Martinez, L., Pascual, A., and Jacoby, G.A. (1998). Quinolone resistance from a transferable plasmid. Lancet 351, 797–799. Matsuoka, M., and Sasaki, T. (2004). Inactivation of macrolides by producers and pathogens. Curr. Drug Targets Infect. Disord. 4, 217–240. Matthews, P.R., and Stewart, P.R. (1988). Amplification of a section of chromosomal DNA in methicillin-resistant Staphylococcus aureus following growth in high concentrations of methicillin. J. Gen. Microbiol. 134, 1455–1464. Mazel, D. (2006). Integrons: agents of bacterial evolution. Nat. Rev. Microbiol. 4, 608–620. Mazel, D., Dychinco, B., Webb, V.A., and Davies, J. (1998). A distinctive class of integron in the Vibrio cholerae genome. Science 280, 605–608.

McMurry, L., Petrucci, R.E., Jr., and Levy, S.B. (1980). Active efflux of tetracycline encoded by four genetically different tetracycline resistance determinants in Escherichia coli. Proc. Natl. Acad. Sci. USA 77, 3974–3977. Meka, V.G., and Gold, H.S. (2004). Antimicrobial resistance to linezolid. Clin. Infect. Dis. 39, 1010–1015. Miller, L.G., Perdreau-Remington, F., Rieg, G., Mehdi, S., Perlroth, J., Bayer, A.S., Tang, A.W., Phung, T.O., and Spellberg, B. (2005). Necrotizing fasciitis caused by community-associated methicillinresistant Staphylococcus aureus in Los Angeles. N. Engl. J. Med. 352, 1445–1453. Moore, I.F., Hughes, D.W., and Wright, G.D. (2005). Tigecycline is modified by the flavin-dependent monooxygenase TetX. Biochemistry 44, 11829–11835. Murakami, S., and Yamaguchi, A. (2003). Multidrug-exporting secondary transporters. Curr. Opin. Struct. Biol. 13, 443–452. Musher, D.M., Dowell, M.E., Shortridge, V.D., Flamm, R.K., Jorgensen, J.H., Le Magueres, P., and Krause, K.L. (2002). Emergence of macrolide resistance during treatment of pneumococcal pneumonia. N. Engl. J. Med. 346, 630–631. Naimi, T.S., LeDell, K.H., Como-Sabetti, K., Borchardt, S.M., Boxrud, D.J., Etienne, J., Johnson, S.K., Vandenesch, F., Fridkin, S., O’Boyle, C., et al. (2003). Comparison of community- and health care-associated methicillin-resistant Staphylococcus aureus infection. JAMA 290, 2976–2984. Neyfakh, A.A., Bidnenko, V., and Chen, L.B. (1991). Efflux-mediated multidrug resistance in Bacillus subtilis: similarities and dissimilarities with mammalian systems. Proc. Natl. Acad. Sci. USA 88, 4781–4785. Nicoloff, H., Perreten, V., McMurry, L.M., and Levy, S.B. (2006). Role for tandem duplication and lon protease in AcrAB-TolC- dependent multiple antibiotic resistance (Mar) in an Escherichia coli mutant without mutations in marRAB or acrRAB. J. Bacteriol. 188, 4413–4423. Nicoloff, H., Perreten, V., and Levy, S.B. (2007). Increased genome instability in Escherichia coli lon mutants: relation to emergence of multiple antibiotic resistant (Mar) mutants caused by insertion sequence elements and large tandem genomic amplifications. Antimicrob. Agents Chemother. Published online January 12, 2007. 10.1128/AAC.01128-06.

hydrolyzing beta-lactamase gene and of two novel aminoglycoside resistance gene cassettes. Antimicrob. Agents Chemother. 45, 546–552. Roberts, M.C. (2004). Resistance to macrolide, lincosamide, streptogramin, ketolide, and oxazolidinone antibiotics. Mol. Biotechnol. 28, 47–62. Robicsek, A., Strahilevitz, J., Jacoby, G.A., Macielag, M., Abbanat, D., Park, C.H., Bush, K., and Hooper, D.C. (2006). Fluoroquinolonemodifying enzyme: a new adaptation of a common aminoglycoside acetyltransferase. Nat. Med. 12, 83–88. Ross, J.I., Eady, E.A., Cove, J.H., and Cunliffe, W.J. (1998). 16S rRNA mutation associated with tetracycline resistance is a gram-positive bacterium. Antimicrob. Agents Chemother. 52, 1702–1705. Ruzin, A., Visalli, M.A., Keeney, D., and Bradford, P.A. (2005). Influence of transcriptional activator RamA on expression of multidrug efflux pump AcrAB and tigecycline susceptibility in Klebsiella pneumoniae. Antimicrob. Agents Chemother. 49, 1017–1022. Sapunaric, F.M., Aldema-Ramos, M., and McMurry, L.M. (2005). Tetracycline resistance: efflux, mutation, and other mechanisms. In Frontiers in Antimicrobial Resistance: A Tribute to Stuart B. Levy, D.G. White, M.N. Alekshun, and P.F. McDermott, eds. (Washington, DC: ASM Press), pp. 3-18. Schneiders, T., Amyes, S.G., and Levy, S.B. (2003). Role of AcrR and ramA in fluoroquinolone resistance in clinical Klebsiella pneumoniae isolates from Singapore. Antimicrob. Agents Chemother. 47, 2831–2837. Schumacher, M.A., and Brennan, R.G. (2002). Structural mechanisms of multidrug recognition and regulation by bacterial multidrug transcription factors. Mol. Microbiol. 45, 885–893. Schwarz, S., Kehrenberg, C., Doublet, B., and Cloeckaert, A. (2004). Molecular basis of bacterial resistance to chloramphenicol and florfenicol. FEMS Microbiol. Rev. 28, 519–542. Sharma, S.K., and Mohan, A. (2006). Multidrug-resistant tuberculosis: A menace that threatens to destabilize tuberculosis control. Chest 130, 261–272. Skold, O. (2001). Resistance to trimethoprim and sulfonamides. Vet. Res. 32, 261–273.

Noble, W.C., Virani, Z., and Cree, R.G. (1992). Co-transfer of vancomycin and other resistance genes from Enterococcus faecalis NCTC 12201 to Staphylococcus aureus. FEMS Microbiol. Lett. 72, 195–198.

Sloan, J., McMurry, L.M., Lyras, D., Levy, S.B., and Rood, J.I. (1994). The Clostridium perfringens Tet P determinant comprises two overlapping genes: tetA(P), which mediates active tetracycline efflux, and tetB(P), which is related to the ribosomal protection family of tetracycline-resistance determinants. Mol. Microbiol. 11, 403–415.

Ochs, M.M., McCusker, M.P., Bains, M., and Hancock, R.E.W. (1999). Negative regulation of the Pseudomonas aeruginosa outer membrane porin OprD selective for imipenem and basic amino acids. Antimicrob. Agents Chemother. 43, 1085–1090.

Stanton, T.B., and Humphrey, S.B. (2003). Isolation of tetracyclineresistant Megasphaera elsdenii strains with novel mosaic gene combinations of tet(O) and tet(W) from swine. Appl. Environ. Microbiol. 69, 3874–3882.

Oethinger, M., Kern, W.V., Jellen-Ritter, A.S., McMurry, L.M., and Levy, S.B. (2000). Ineffectiveness of topoisomerase mutations in mediating clinically significant fluoroquinolone resistance in Escherichia coli in the absence of the AcrAB efflux pump. Antimicrob. Agents Chemother. 44, 10–13.

Stein, G.E., and Craig, W.A. (2006). Tigecycline: A critical analysis. Clin. Infect. Dis. 43, 518–524.

Perreten, V., and Boerlin, P. (2003). A new sulfonamide resistance gene (sul3) in Escherichia coli is widespread in the pig population of Switzerland. Antimicrob. Agents Chemother. 47, 1169–1172.

Tait-Kamradt, A., Davies, T., Appelbaum, P.C., Depardieu, F., Courvalin, P., Petitpas, J., Wondrack, L., Walker, A., Jacobs, M.R., and Sutcliffe, J. (2000). Two new mechanisms of macrolide resistance in clinical strains of Streptococcus pneumoniae from Eastern Europe and North America. Antimicrob. Agents Chemother. 44, 3395– 3401.

Pesci, E.C., Milbank, J.B.J., Pearson, J.P., McKnight, S., Kende, A.S., Greenberg, E.P., and Iglewski, B.H. (1999). Quinolone signaling in the cell-to-cell communication system of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 96, 11229–11234.

Tennent, J.M., Lyon, B.R., Midgley, M., Jones, I.G., Purewal, A.S., and Skurray, R.A. (1989). Physical and biochemical characterization of the qacA gene encoding antiseptic and disinfectant resistance in Staphylococcus aureus. J. Gen. Microbiol. 135, 1–10.

Poirel, L., Lambert, T., Turkoglu, S., Ronco, E., Gaillard, J., and Nordmann, P. (2001). Characterization of Class 1 integrons from Pseudomonas aeruginosa that contain the bla(VIM-2) carbapenem-

Tran, J.H., and Jacoby, G.A. (2002). Mechanism of plasmid-mediated quinolone resistance. Proc. Natl. Acad. Sci. USA 99, 5638– 5642.

Cell 128, March 23, 2007 ©2007 Elsevier Inc.  1049

Watanabe, T. (1963). Infective heredity of multiple drug resistance in bacteria. Bacteriol. Rev. 27, 87–115. Weigel, L.M., Clewell, D.B., Gill, S.R., Clark, N.C., McDougal, L.K., Flannagan, S.E., Kolonay, J.F., Shetty, J., Killgore, G.E., and Tenover, F.C. (2003). Genetic analysis of a high-level vancomycin-resistant isolate of Staphylococcus aureus. Science 302, 1569–1571.

Wolter, N., Smith, A.M., Farrell, D.J., Schaffner, W., Moore, M., Whitney, C.G., Jorgensen, J.H., and Klugman, K.P. (2005). Novel mechanism of resistance to oxazolidinones, macrolides, and chloramphenicol in ribosomal protein L4 of the pneumococcus. Antimicrob. Agents Chemother. 49, 3554–3557.

Weisblum, B. (1998). Macrolide resistance. Drug Resist. Updat. 1, 29–41.

Woolridge, D.P., Vazquez-Laslop, N., Markham, P.N., Chevalier, M.S., Gerner, E.W., and Neyfakh, A.A. (1997). Efflux of the natural polyamine spermidine facilitated by the Bacillus subtilis multidrug transporter Blt. J. Biol. Chem. 272, 8864–8866.

Wilson, P., Andrews, J.A., Charlesworth, R., Walesby, R., Singer, M., Farrell, D.J., and Robbins, M. (2003). Linezolid resistance in clinical isolates of Staphylococcus aureus. J. Antimicrob. Chemother. 51, 186–188.

Zhanel, G.G., Dueck, M., Hoban, D.J., Vercaigne, L.M., Embil, J.M., Gin, A.S., and Karlowsky, J.A. (2001). Review of macrolides and ketolides: focus on respiratory tract infections. Drugs 61, 443–498.

1050  Cell 128, March 23, 2007 ©2007 Elsevier Inc.