Antibiotics Ian Chopra University of Leeds, Leeds, UK Advanced doi:10.1038/npg.els.0002225 Antibiotics are natural and synthetic compounds with selective bactericidal or bacteriostatic effects that eliminate pathogens or slow their growth such that the host defence mechanisms can clear the infection. Antibiotics are classified mechanistically according to their site of action in the bacterial cell. Introduction The term õ?~antibioticõ?T includes a variety of naturally occurring and synthetic organic molecules that have sufficient selective activity against bacteria to permit systemic, or topical, administration in man or animals as chemotherapeutic agents for the treatment of bacterial infections. Infection usually involves multiplication of bacteria in host tissues and antibiotics prevent this phase by interacting with targets that either kill the pathogens (a bactericidal response) or slow their growth (a bacteriostatic response) to the point where host defence mechanisms can clear the infection. See also: Antibiotic molecules: intracellular; Antimicrobial proteins and peptides; Antimicrobial agents: basis of action; Bacterial antibiotic resistance; Antimicrobials against streptococci, pneumococci and enterococci; Antimicrobials against fastidious Gram-negative rods; Antimicrobials against anaerobic bacteria The origin of anti-infective chemotherapy can probably be traced to about 3000 bc, when it is evident that early Chinese civilizations recognized the therapeutic properties of teas and herbs. However, it was not until 1935 that the antibiotic era truly began when Gerhard Domagk (1895õ?"1964) declared the discovery of Prontosil, the forerunner of the sulfonamides. This was followed in the early 1940s with the introduction of penicillin, originally discovered in 1929 by Alexander Fleming (1881õ?"1955) and developed by Ernst Chain (1906õ?"1979) and colleagues during the late 1930s. During the period from the end of the Second World War to the present, a large number of ad***ional antibiotics have been discovered that match, or even surpass, the efficacy of sulfonamides and penicillin in combating bacterial infections. Undoubtedly the discovery of antibiotics represents one of the most important medical advances of the twentieth century. It has permitted the control of bacterial infections, thereby resulting in substantially reduced morbi***y and mortality from such diseases. Antibiotics can be classified into groups on the basis of their principal sites of action within the bacterial cell (Figure 1). The sections that follow describe the major classes of antibiotic according to the mechanistic classification presented in Figure 1. For each group the target sites and mode of action are described, as well as the basis for selective antiprokaryotic activity. Selective antiprokaryotic activity, or õ?~selective toxicityõ?T, is a vital guiding principle for antibacterial chemotherapy since the infecting bacterium must be inhibited, or killed, without harming the host. As described below, most antibiotics exploit differences in the metabolism or structure of bacterial and mammalian cells to achieve selectivity. Figure 1 Mechanisms of antibiotic action. Schematic illustrating the mode of action of antibiotics on bacteria according to the target site of each group of agents. The principal processes inhibited by antibiotics are nucleotide biosynthesis, DNA replication, RNA transcription, and protein synthesis. In ad***ion, some peptide antibiotics disrupt bacterial cytoplasmic membrane integrity. A limited num ... Inhibitors of Nucleotide Biosynthesis Tetrahydrofolic acid (THFA) is an important cofactor in nucleotide biosynthesis because it is required as a donor of one-carbon units at several stages in purine and pyrimidine synthesis. Therefore, antibiotics that interfere with the biosynthesis of THFA are powerful indirect inhibitors of nucleotide biosynthesis. Synthesis of THFA in bacteria is subject to inhibition by two groups of antibiotics, the sulfonamides (e.g. sulfamethoxazole, Sx) and the 2,4-diaminopyrimidines (e.g. trimethoprim, Tp) (Figure 2). Sulfonamides are structural analogues of p-aminobenzoic acid (PABA), acting as alternative substrates that bind more tightly to dihydropteroate synthase (DHPS) than does PABA itself. This results in the formation of inactive folate-like analogues. Since DHPS is absent from mammalian cells (Figure 2) sulfonamides exhibit selective activity towards bacteria. The 2,4-diaminopyrimidines (e.g. trimethoprim) are competitive inhibitors of bacterial dihydrofolate reductase (DHFR) (Figure 2). Although DHFR is present in mammalian cells, drugs such as trimethoprim are highly selective towards bacterial DHFRs; for example, trimethoprim is 80 000 times more active against E. coli DHFR than against mammalian DHFR. See also: Nucleotides: structure and properties; Nucleotide synthesis de novo Figure 2 The tetrahydrofolic acid biosynthetic pathway in bacteria and mammals and its inhibition by sulfamethoxazole (Sx) and trimethoprim (Tp). A reduced form of folic acid, tetrahydrofolic acid, is an important cofactor in the biosynthesis of nucleotides and is required for growth by both bacterial (a) and mammalian (b) cells. Because mammalian cells are unable to synthesize folic acid (b), this c ... Inhibitors of DNA Synthesis Bacterial DNA-damaging agents The nitroimidazoles are nitroheterocyclic drugs, containing a nitro group at position 2 or 5 of the molecule. Nitroimidazole drugs exert their antimicrobial effect via reduction of the nitro group (Edwards, 1997). The nitro groups are reduced at reduction potentials that are only generated in anaerobes. Consequently, antibiotics of this type (e.g. metronidazole) are active only under anaerobic con***ions and it is against infections caused by anaerobic bacteria that nitroimidazoles have their main therapeutic activity and use. The spectrum of metronidazole and other nitroimidazoles also extends to several anaerobic protozoal parasites. The antibacterial (and antiprotozoal) activity of these antibiotics appears to result from the formation of short-lived nitro radical anions following reduction of the nitro group in the antibiotic molecule by nitroreductases present in the bacterial or protozoal cell. The radicals attack DNA, causing strand breaks and helix destabilization. These events prevent DNA synthesis, thereby causing cell death, in both bacteria and protozoal parasites. See also: Bacterial cell division DNA topoisomerase inhibitors Replication of the bacterial chromosome, a circular duplex DNA molecule, requires the separation of the two highly intertwined parental strands. However, separation of strands wound in a helix generates loops, termed positive supercoiled twists, in the single strands. Unless prevented, this situation would stop further unwinding of parental DNA at the replication fork. The enzyme DNA gyrase (a type II DNA topoisomerase) relaxes positively supercoiled DNA by periodically breaking a phosphodiester bond in one of the strands of the double helix, introducing negative supercoils, and finally resealing the nick (Hooper and Wolfson, 1993). See also: Topoisomerases Bacterial DNA gyrase is a tetrameric enzyme composed of four subunits: two gyrase A subunits (each with a molecular weight of 97 kDa) and two gyrase B subunits (each with a molecular weight of 90 kDa). All activities of the enzyme require both A and B subunits, but certain domains mediate different functions. The A subunits of gyrase are involved in the DNA breakage and resealing events associated with supercoiling, while the B subunits are responsible for ATP hydrolysis, reflecting the consumption of energy in the formation of negative supercoils (Hooper and Wolfson, 1993). Bacteria contain a second type II DNA topoisomerase called DNA topoisomerase IV. This is a heterodimer with subunits of molecular weight 70 kDa and 75 kDa. DNA topoisomerase IV catalyses ATP-dependent relaxation of negatively and positively supercoiled DNA. However, unlike gyrase, topoisomerase IV displays no DNA supercoiling activity and appears to have a primary role in chromosomal partitioning (segregation) to daughter cells during cell division (Hooper and Wolfson, 1993). The quinolone antibacterial agents, including nalidixic acid and the newer fluoroquinolones, selectively inhibit DNA gyrase and topoisomerase IV and therefore arrest the essential process of bacterial DNA replication. In Gram-negative bacteria the primary target of quinolones is DNA gyrase, whereas in Gram-positive bacteria the primary target is DNA topisomerase IV (Piddock, 1998). These drugs do not bind directly to DNA gyrase or topoisomerase IV, but form stable drugõ?"DNAõ?"enzyme complexes that prevent the further functioning of the topoisomerases (Shen, 1993). Eukaryotic topoisomerase II displays enzymatic activities that are analogous to those mediated by bacterial DNA gyrase. However, the quinolones developed for use as antibiotics demonstrate inhibition of the eukaryotic enzyme only at concentrations that are two to three orders of magnitude above those required for antibacterial activity (Gootz and Osheroff, 1993). Inhibitors of RNA Synthesis RNA polymerases are enzymes that mediate transcription of structural genes by catalysing the initiation and elongation of RNA molecules on a DNA template. Bacterial RNA polymerases comprise a core enzyme (E) and a specificity subunit, also defined as a sigma factor (ẽf). Core polymerase (E) is a stable noncovalent assembly of four polypeptide chains consisting of two ẻ subunits (each with a molecular weight of 36.5 kDa), one ẻ subunit (molecular weight 151 kDa) and one ẻõ? subunit (molecular weight 155 kDa). Under normal growth con***ions, transcription of most genes is initiated by RNA polymerase holoenzyme Eẽf70, in which the specificity subunit is ẽf70. Under other growth con***ions, e.g. during the stationary phase of growth, alternative sigma factors are used to achieve expression of a different set of genes (Ishihama, 2000). See also: RNA synthesis Since the total number of core enzyme (E) molecules (approximately 2000 per bacterial cell) is less than the number of structural genes (typically 4000 per bacterial cell), ẽf factor interactions with the core enzyme have a vital role in determining which genes are transcribed and hence the pattern of gene expression in the cell. In E. coli there are seven different species of ẽf subunit, each responsible for recognition of a specific set of promoters (Ishihama, 2000). Rifampicin, a member of the rifamycin group of antibiotics, binds selectively to the ẻ subunit of bacterial RNA polymerase, and in so doing interferes with the ability of the holoenzyme (Eẽf) to initiate RNA synthesis. Binding of RNA polymerase to the DNA template is not blocked and inhibition probably results from interference with the formation of the first phosphodiester bond in the RNA chain (Parenti and Lancini, 1997). Rifampicin neither binds to nor inhibits the corresponding mammalian enzyme (Parenti and Lancini, 1997). Rifapentine and rifabutin are two further members of the rifamycin group of antibiotics that have been introduced into clinical practice more recently than rifampicin (Parenti and Lancini, 1997). Their mechanism of antibacterial action is identical to that of rifampicin but their pharmacokinetic properties differ from those of rifampicin, making them more suitable than rifampicin for the chemotherapy of certain types of infection. Other antibiotics, distinct from the rifamycin class, that appear to selectively inhibit bacterial RNA polymerase have also been described. These compounds include streptolydigin, ripostatin, sorangicin and corralopyronin. So far these antibiotics have not been developed for use as chemotherapeutic agents in man (Oõ?TNeill et al., 2000).
Inhibitors of Bacterial Protein Synthesis Various clinically useful antibacterial agents inhibit protein synthesis by directly inhibiting the functions of ribosomes, the cellular organelles upon which the genetic message is decoded to produce proteins. The specific action of antibacterial protein synthesis inhibitors arises from differences in the structure of bacterial (70S) and mammalian (80S) ribosomes (Table 1) allowing unique binding sites in the bacterial ribosome to be selectively exploited by the antibacterial agents. In some cases differential accumulation by bacterial and mammalian cells also contributes to the selective action of drugs inhibiting protein synthesis; for example, the tetracyclines and fusidic acid are more readily accumulated by bacteria than by mammalian cells. See also: Protein synthesis initiation in prokaryotes Table 1 Bacterial protein synthesis is a complex process involving the formation of active 70S particles capable of catalysing peptide bonds between incoming amino acids in a repetitive process known as the elongation cycle (Figure 3). Most clinically useful inhibitors of bacterial protein synthesis, including lincosamides, macrolides, tetracyclines, chloramphenicol, aminoglycosides, streptogramins and fusidic acid, act by inhibiting or modifying this cycle (Piddock, 1998). Figure 3 Bacterial protein synthesis showing the steps inhibited by antibiotics. In the first stage of bacterial protein synthesis, mRNA, transcribed from a structural gene, binds to the smaller (30S) ribosomal subunit and attracts formylmethionine-tRNA to the initiator codon AUG. The larger (50S) subunit is then added to form a complete (70S) initiation complex. The site occupied by the formylmethio ... Each of these drugs binds to separate sites in the 70S ribosomal particle and affects different stages of the elongation cycle: lincosamides and chloramphenicol inhibit peptidyltransferase activity; the macrolides cause dissociation of peptidyl-tRNA from the ribosome; the tetracyclines inhibit the binding of aminoacyl-tRNA; streptogramins inhibit both the binding of aminoacyl-tRNA and peptidyltransferase activity; and fusidic acid forms a stable complex between the ribosome and the soluble elongation factor G (EF-G), the release and re-association of which are important events that facilitate the elongation cycle. Aminoglycosides have a variety of effects that include prevention of the movement between ribosomes and mRNA (spectinomycin), misreading of proteins being translated (streptomycin), and inhibition of binding of EF-G to the ribosome (neomycins, gentamicins, kanamycins, amikacin and tobramycin). In contrast to most protein synthesis inhibitors, which affect the elongation cycle, linezolid, a member of a new class of antibacterial agents, the oxazolidinones, prevents the formation of the 70S initiation complex (Figure 3) (Burghardt et al., 1998) by binding both to the 50S and 30S ribosomal subunits (Matassova et al., 1999). Membrane-disorganizing Agents Like all cells, bacteria have a cytoplasmic membrane within which the cytoplasm is contained. The structurally related cyclic peptide antibiotics polymyxin B and polymyxin E (colistin), which are primarily active against Gram-negative bacteria, exert their antibacterial activity by disrupting both the outer and inner (cytoplasmic) membranes of the Gram-negative cell. Death of the bacterial cell results from leakage of cytoplasmic contents. The preferential activity of polymyxins against Gram-negative bacteria results in part from their binding to lipopolysaccharide (LPS) in the outer membrane followed by self-promoted uptake and interaction with the cytoplasmic membrane (Han**** and Chapple, 1999). In ad***ion to LPS binding, the Gram-negative spectrum of these cyclic peptides also results from preferential interaction with cytoplasmic membranes containing the phospholipid phosphatidylethanolamine, which is present in Gram-negative bacteria but generally lacking in Gram-positive species. Although polymyxin B and polymyxin E bind less readily to mammalian cell membranes than to the cytoplasmic membrane of Gram-negative bacteria, the interaction with mammalian cells does result in a number of adverse side effects. Consequently, the polymyxins have only a minor role in medicine. See also: Bacterial cell wall; Bacterial cytoplasmic membrane In view of the experience with polymyxins and other relatively nonspecific membrane-disorganizing agents, it was believed until recently that the bacterial cytoplasmic membrane was insufficiently different from the mammalian cell membrane to be a useful selective target for the development of antibacterial agents. However, more recently there has been renewed interest in the discovery of new peptide antibiotics with greater specificity for the bacterial cytoplasmic membrane. This has led to the development of a number of peptides that are currently being evaluated in clinical trials as both topical and systemic agents (Han**** and Chapple, 1999). The antibacterial mode of action of these peptide antibiotics is not fully understood but probably involves a number of distinct stages. Upon insertion into the bacterial cytoplasmic membrane, the peptides probably aggregate to form a micelle-like complex that spans the membrane bilayer. Such micelles are proposed to have water associated with them, which provides channels for the movement of ions across the membrane and also leakage of larger water-soluble molecules from the cell. Antibiotics That Inhibit Cell Wall Biosynthesis The cell walls of most pathogenic bacteria contain peptidoglycan, an essential polymer that confers cell shape and maintains the structural integrity of the cell. A number of antibiotics (Figure 4 and discussed below) interfere with the synthesis of bacterial peptidoglycan and are lethal to the cell through this activity. In ad***ion to peptidoglycan, the cell walls of mycobacteria contain two unique polymers, mycolic acid and arabinogalactan, whose synthesis is disrupted by a number of specific antimycobacterial agents that have applications for the treatment of tuberculosis and other mycobacterial infections. The absence of peptidoglycan, mycolic acid and arabinogalactan from mammalian cells provides an opportunity for specific inhibition of the biosynthetic pathways for these macromolecules by antibiotics. See also: Bacterial cell wall Figure 4 Bacterial peptidoglycan synthesis and its inhibition by antibiotics. The first dedicated step in bacterial peptidoglycan synthesis involves the synthesis of N-acetylmuramic acid (NAMA) from N-acetylglucosamine (NAG) by the ad***ion of a lactic acid substituent, derived from phosphoenolpyruvate, to NAG. This reaction is blocked by fosfomycin, which inhibits the pyruvyltransferase catalysing t ... Inhibitors of peptidoglycan synthesis Several stages in the formation of the bacterial cell wall provide targets for therapeutically useful antibacterial agents (Figure 4). The first step is the formation of N-acetylmuramic acid (NAMA) by the condensation of phosphoenolpyruvate with N-acetylglucosamine (NAG), a reaction that is inhibited by the phosphonic acid antibiotic fosfomycin. NAMA is next substituted by five amino acids, the last two of which, d-alanyl-d-alanine (d-ala-d-ala) are added as a dipeptide unit. d-Alanine is derived from l-alanine by alanine racemase and the dipeptide is formed by a ligase; both reactions are competitively inhibited by cycloserine. Any amino acids needed for interpeptide bridges (e.g. pentaglycine in Staphylococcus aureus) are now added and the cell wall unit is completed by the ad***ion of NAG to the NAMA-peptide. See also: Peptidoglycan The biosynthetic steps described above occur in the cytoplasm and the cell wall unit (NAG?"NAMA?"pentapeptide) is now transported across the cytoplasmic membrane to the peptidoglycan growth site. A lipid (55-carbon isoprenyl phosphate) carrier molecule in the membrane accomplishes this. In this process the lipid acquires an ad***ional phosphate group that is then removed to regenerate the carrier function. This dephosphorylation reaction is inhibited by the cyclic peptide antibiotic bacitracin. The process of transglycosylation now transfers the nascent cell wall unit to the growing end of the peptidoglycan chain. Transglycosylation, whereby glycan units are polymerized within the peptidoglycan, is inhibited by glycopeptide antibiotics such as vancomycin and teicoplanin. These large, saccharide-containing antibiotics have no useful activity against Gram-negative bacteria because they fail to penetrate the outer membrane of these organisms. However, in Gram-positive bacteria these antibiotics bind specifically to the terminal d-alanyl-d-alanine group on the pentapeptide side-chain of both the lipid-linked NAG?"NAMA?"pentapeptide and to the pentapeptide in the growing point of the peptidoglycan. The interactions are mediated by the formation of key hydrogen bonds between the dipeptide and the antibiotics (for example see Figure 5 for vancomycin interaction) (Nicas and Cooper, 1997). Strictly speaking, the activity of the transglycosylase enzyme is not inhibited, but the complex of glycopeptides with the pentapeptide units blocks the incorporation of the disaccharide (NAG?"NAMA) pentapeptide into the growing peptidoglycan chain. The association of glycopeptides with the NAG?"NAMA?"pentapeptide also prevents subsequent crosslinking (transpeptidation, see below) between adjacent peptidoglycan chains. Figure 5 Interaction of vancomycin with the peptidoglycan precursor NAG?"NAMA?"pentapeptide. Vancomycin (a) interacts with nascent peptidoglycan (b) through key hydrogen-bonding interactions (dashed lines) between functional groups on the antibiotic and sites in the D-alanyl-D-alanine dipeptide unit of NAMA?"pentapeptide. Binding of vancomycin to nascent peptidoglycan prevents translocation, whereby NA ... In the final stages of peptidoglycan biosynthesis, adjacent peptidoglycan chains are crosslinked to give the wall its mechanical strength and this transpeptidation reaction is the site of action of penicillins and other Ỵ-lactam antibiotics. Bacterial transpeptidases bind to the nascent peptidoglycan and the active-site serine residue cleaves the terminal d-alanine residue of the substrate to give an intermediate acylated d-alaninê?"enzyme species (Figure 6b) (Dax, 1997). An acceptor glycan strand then reacts at the acyl carbonyl centre of this complex and transpeptidation is achieved, releasing free enzyme for further rounds of transpeptidation (Figure 6a). Ỵ-Lactam antibiotics bind to transpeptidases and adopt a conformation that closely mimics the d-alanyl-d-alanine terminus of the peptidê?"glycan complex primed for transpeptidation. Consequently, Ỵ-lactam antibiotics are able to acylate active-site serine residues within transpeptidases to form inactive enzyme complexes that are essentially unable to regenerate a functional transpeptidase through hydrolysis of the enzymê?"Ỵ-lactam complex (Figure 6b) (Dax, 1997). Most bacteria contain more than one transpeptidase (also known as penicillin-binding proteins) and some Ỵ-lactam antibiotics display different affinities for these essential bacterial targets. Figure 6 Mechanism of the transpeptidase reaction mediating cross-linkage of bacterial peptidoglycan (upper) and its inhibition by Ỵ-lactam antibiotics (lower). The final stage of peptidoglycan synthesis involves crosslinkage between adjacent peptide side-chains of the NAG?"NAMA?"pentapeptide units. This requires the activity of a transpeptidase, containing an active-site serine, which recognizes the t ... Inhibitors of mycolic acid and arabinogalactan synthesis in mycobacteria The antibiotics isoniazid, ethionamide and ethambutol inhibit the synthesis of specific wall-associated polymers in mycobacteria (Kremer et al., 2000; Zhang and Telenti, 2000). Mycolic acids, whose synthesis is inhibited by the structurally related drugs isoniazid and ethionamide, occur primarily as esters bound to arabinogalactan, a polymer that itself is covalently linked to mycobacterial peptidoglycan. Although extensive mode-of-action studies have been conducted on both isoniazid and ethionamide, their precise modes of action have yet to be elucidated. However, it is assumed that these drugs inhibit one or more key steps in the complex biosynthetic pathway leading to the synthesis of mycolic acids (Kremer et al., 2000). Ethambutol acts on the biosynthesis of arabinogalactan, the major polysaccharide of the mycobacterial cell wall. Although the precise target is unknown, ethambutol is assumed to inhibit one or more of the arabinosyltransferases involved in the polymerization of cell wall arabinogalactan (Zhang and Telenti, 2000). Future Directions The ability to treat bacterial infections with chemotherapeutic agents, introduced with the discovery of penicillin and Prontosil in the 1930s and 1940s, represents one of the most important medical achievements of the twentieth century. Indeed, the rapid advances made in the discovery of new antibiotics and other antibacterial agents during the so-called ?~golden?T period between 1940 and the mid-1960s led to widespread optimism that bacterial infections could be completely conquered. This period of optimism is captured by the famous remark made in 1969 by the US Surgeon General who testified to Congress that ?~the time has come to close the book on infectious diseasê?T. However, even from the very earliest period of the antibiotic era the potential for the emergence of drug-resistant bacteria has been recognized. Unfortunately, the selection of organisms resistant to antibiotics has continued to the present day and the new millennium has arrived with the dramatic emergence of resistance to antibiotics in all significant bacterial pathogens. Furthermore, bacteria resistant to virtually all chemotherapeutically useful antibiotics have been identified among clinical isolates of some bacterial species. See also: Bacterial antibiotic resistance; Antibiotic resistance plasmids in bacteria; Antimicrobial resistance: epidemiology; Antimicrobial resistance: control The discovery of new antibiotics is now a matter of urgency and it is likely that future research will be primarily concerned with the identification of novel agents active against previously unexploited bacterial targets (Chopra et al., 1997). With the availability of several bacterial genome sequences, we are now in a good position to discover new antibiotics directed against novel targets (Moir et al., 1999). Originally published: June 2001
Antibiotic Resistance In the 1940s, the clinical use of antibiotics first curbed the widespread threat of deadly bacterial infection. These drugs effectively inhibited bacterial growth that had gone unchecked for decades. Antibiotics did not, however, eradicate the threat of bacterial infection. In fact, the widespread use of antibiotics gave a selective advantage to bacteria that had antibiotic resistance. Many strains of bacteria have developed antibiotic resistance, or insensitivity to antibiotic drugs, in response to antibiotic selection pressures. Now, bacteria employ a myriad of resistance mechanisms to circumvent the best efforts of antibiotic researchers and clinicians. Only with the development of new and potent antibiotics and the appropriate use of existing antibiotics will researchers regain control over this resilient lifeform, bacteria. 1. A Historical Perspective The development of antibiotics as therapeutic agents began in the late 1930s to combat the most common cause of death, infectious disease. In the preantibiotic era, any infection could prove mortal. Subsequently, over 150 different antibiotics have been synthesized or discovered, and these drugs are used to treat bacterial infections, such as pneumonia, malaria, and tuberculosis (1, 2). Antibiotics are a collection of natural products and synthetic compounds that kill bacteria. Naturally occurring antibiotics are isolated from molds, yeasts, and bacteria. These organisms use antibiotics as defense mechanisms to kill other bacteria. Alternatively, synthetic antibiotics are developed by understanding the architecture and function of bacteria (see Fig. 1). Some bacteria have cell walls, and many effective antibiotics, such as penicillin, bacitracin, and cephalosporin, inhibit the synthesis of this cell wall. The bacterial machinery for protein biosynthesis differs from that of many host organisms and, therefore, is a good target for antibiotics, such as tetracycline and chloramphenicol. Ad***ionally, antibiotics, such as rifampin and quinolones, specifically inhibit DNA replication in bacteria (1). Figure 1. Mechanisms of action of some common antibiotics. A schematic of some basic functions of bacterial cells to which antibiotics are targeted. The X indicates an inhibition of that function by the antibiotics noted (see text for details). Antibiotics were considered the â?owonder drugsâ? of their time, and in retrospect, this highly favored opinion resulted in their overuse. Antibiotics were commonly prescribed by physicians to cure and to appease their patients. Some patients would have genuine bacterial infections, for which antibiotic treatment is appropriate, whereas others would request antibiotics for viral infections that are not susceptible to these drugs. In ad***ion, individuals who have suppressed immune systems, such as AIDS patients or organ transplants patients, would harbor bacteria that acquire resistance more easily. Antibiotics were also used prophylactically in agriculture and aquaculture industries to keep livestock healthy (3). By the late 1960s, infectious disease appeared to be under control by the use of a variety of antibiotics. However, antibiotics simply depressed the propagation of bacteria. They did not eradicate it. Nonetheless, research in human health changed its focus from infectious diseases to chronic diseases, and new antibiotics were no longer being developed (2, 4). Many microbiologists warned the human health community that bacteria were potent, infectious pathogens that should not be underestimated. Bacteria have survived for more than three billion years despite numerous environmental changes on earth, and industrial wastes, insecticides, and herbicides. Obviously, the mechanisms bacteria use *****rvive and adapt to these adversities were very effective. 2. The Origins of Antibiotic Resistance Even before the first clinical application of antibiotics, antibiotic resistance, the ability of bacteria to evade the deleterious effects of antibiotics, was postulated. In 1940, Abraham and Chain identified a bacterial enzyme that inactivates one of the first antibiotics, penicillin. Then, any bacterium that produces this enzyme would be resistant to penicillin (5). Moreover, microorganisms that use an antibiotic as defense mechanisms would require immunity to that antibiotic. This inherent resistance to a particular antibiotic was defined as a naturally-occurring trait called intrinsic resistance. A few strains of bacteria with intrinsic resistance to a particular antibiotic would not constitute a clinical threat because many diverse antibiotics are available. However, the ability of bacteria to propagate this antibiotic resistance to other strains of bacteria had been underestimated. The exchange of genetic information between bacteria of the same strain is a common, yet typically slow process. Mechanisms of exchange include (1) conjugationâ?"a single DNA strand from one bacterium enters another bacterium and is replicated as a part of that genome, (2) transductionâ?"foreign DNA is introduced into bacteria by transducing bacteriophages, and (3) transformationâ?"autonomously replicating circular DNA plasmids are obtained by bacteria (1). Originally, it was believed that these genetic exchange mechanisms were restricted to bacteria of the same strain. However, a new method of gene exchange, using integrons, was recently identified (6, 7). Integrons are independent, mobile elements that encode genes for protein functions, and encode ad***ional DNA to guarantee the integron''s expression and integration into the bacterial genome. Integrons effectively generate widespread antibiotic resistance by donating antibiotic resistance genes to any strain of bacteria. Ironically, it is widely believed that integrons evolved only recently in response to antibiotic selection pressure. In other words, the use of antibiotics advanced the widespread occurrence of antibiotic resistance. In ad***ion to the acquisition of genes by the exchange of genetic information, bacteria also have a high mutation rate that allows them to respond to the selective pressure of antibiotics by using their own genome. For example, if bacteria were subjected to tetracycline, a random mutation in the 30S ribosome to weaken tetracycline binding would be advantageous and, therefore, would be perpetuated by the survival of the tetracycline-resistant bacteria (1). It has also been postulated that housekeeping genes, like acyltransferases, may have mutated to gain the ability to modify and inactivate aminoglycoside antibiotics (8-10). The widespread phenomenon of antibiotic resistance has developed from the promiscuity of bacteria and their genomes. Initially, intrinsic antibiotic resistances were isolated incidents, but the threat of antibiotics has been readily circumvented using the acquisition of antibiotic resistance genes and the high mutational frequency of individual bacteria. 3. Mechanisms of Antibiotic Resistance Using both newly acquired genes and their own mutated genes, bacteria utilize three basic mechanisms *****pport antibiotic resistance. Enzymes that degrade antibiotics inside the cell are key players in antibiotic resistance. Bacteria can also alter their permeability barriers to keep antibiotic concentrations below toxic levels inside the cell. Furthermore, the cellular targets of antibiotics can be modified to evade the effects of the antibiotics. These mechanisms of antibiotic resistance are distinct, but all are obtained through the acquisition or mutation of genes. 3.1. Enzymatic Inactivation of Antibiotics Degradative enzymes are a common mechanism by which bacteria become resistant to antibiotics. Such enzymes chemically modify antibiotics so that they no longer function. The genes for these degradative enzymes are obtained by acquisition of exogenous genes or mutation of endogenous genes. The expression of these genes also governs the level of antibiotic resistance in bacteria. b-Lactamases are common examples of degradative enzymes that generate antibiotic resistance. b-Lactamases inactivate b-lactam antibiotics, a structurally similar group of penicillin-like antibiotics, all of which have a b-lactam ring structure. b-Lactamases are typically grouped into two major classes, penicillinases and cephalosporinases, based on their substrate affinity (11, 12). In ad***ion to b-lactamases, other enzymes also degrade different antibiotics, such as the aminoglycosides, gentamycin, tobramycin, and amikacin. All b-lactam antibiotics function similarly. Their b-lactam ring structure inhibits the final step of bacterial cell wall synthesis. Bacterial cell walls are constructed of alternating N-acetylglucosamine and N-acetyl-muramic acid residues that form long peptidoglycan chains, and the final step in cell wall synthesis involves the enzymatic crosslinking of these peptidoglycan chains by a transpeptidase. Because the b-lactam bond resembles a portion of the peptidoglycan chains, this transpeptidase can mistake a b-lactam antibiotic for its natural substrate and hydrolyze the b-lactam bond. This hydrolysis covalently links the b-lactam drug to the transpeptidase and renders it nonfunctional (13) (see also Penicillin-Binding Proteins). Although b-lactamases are effective at degrading some antibiotics, their mere presence is not sufficient to cause clinically relevant antibiotic resistance. In fact, these enzymes are found ubiquitously in almost all bacteria, and in some blue-green algae and mammalian tissues. b-lactamases must be present in sufficient quantities to degrade the b-lactam antibiotics effectively before they inhibit cell wall synthesis. The cellular concentration of a b-lactamase depends on its gene expression, and b-lactam antibiotics are inducers of b-lactamase expression. Furthermore, particular b-lactamases have variable affinities for b-lactam antibiotics. Therefore, the degree of antibiotic resistance due to b-lactamases is based on a combination of the ability of the b-lactam antibiotic to induce the expression of b-lactamase, and its ability to be a substrate for b-lactamase (14). As researchers began to understand this mechanism of antibiotic resistance, more effective b-lactam antibiotics were developed. The early cephalosporins, like penicillin and amoxicillin, are extremely sensitive to b-lactamases, because these b-lactam antibiotics are potent inducers of b-lactamase expression and good substrates for the b-lactamase. In contrast, the more recently developed antibiotic, imipenem, is a strong inducer of b-lactamase expression but maintains its antibiotic activity because it is a poor substrate for most b-lactamases (15). In ad***ion to these new antibiotics, combination therapies are also being implemented to combat antibiotic resistance. Such therapies are comprised of a b-lactam antibiotic together with b-lactamase inhibitors, like clavulanic acid, sulbactam, and tazobactum (16).
3.2. Altered Permeability Barriers: Pore Proteins and Efflux Systems. The bacterial cell membrane is the major permeable barrier separating the outside of the cell from the inside. The flui***y of the membrane is generally balanced to include most nutrients, while excluding many toxins. Adjusting this flui***y impedes the function of the membrane. Therefore, bacteria cannot protect themselves by changing the flui***y of their membrane. Instead, bacteria have ad***ional structures that surround the cytoplasmic membrane or form pores through it. The alteration of these structures to exclude antibiotics is another mechanism of antibiotic resistance. Gram-positive and Gram-Negative Bacteria have distinct structures that surround their cytoplasmic membranes. Most Gram-positive bacteria have thick cell walls that are mechanically quite strong, although very porous. Although the cell wall helps Gram-positive bacteria retain their shape, it does not exclude most antibiotics and, therefore, is not a good barrier. Thus, Gram-positive bacteria are relatively susceptible to the influx of antibiotics. Alternatively, a more effective barrier, a second lipid bilayer or membrane, surrounds Gram-negative bacteria. This outer membrane is partially composed of a lipid, lipopolysaccharide (LPS), that is not commonly found in cytoplasmic membranes. The distinguishing feature of LPS is its decreased flui***y, which makes the LPS bilayer an efficient barrier that prevents the permeation of most hydrophobic antibiotics into Gram-negative bacteria (17). Enveloped by effective barriers, bacteria use pore-forming proteins, called porins, to obtain nutrients from outside the cell. Porins are transmembrane proteins that function as nonspecific, aqueous channels, and allows nutrients to diffuse across the membrane. Porins generally exclude antibiotics because they are narrow and restrictive. Most antibiotics are large, uncharged molecules that cannot easily traverse the narrow porin channels that are lined with charged amino acid residues. However, some antibiotics enter the bacteria through porins, and the deletion or alteration of these porins to exclude particular antibiotics is linked to antibiotic resistance. Because bacteria cannot develop barriers that are impermeable to all molecules, some toxins do diffuse into bacteria along with nutrients. Therefore, bacterial cell membranes also contain transport proteins that cross the membranes and use energy to remove toxins. They are called active efflux systems, and some are directly identified as another significant cause of antibiotic resistance. Many active efflux systems resemble other transport proteins that catalyze the efflux of common, small molecules, like glucose or cations, and it is likely that mutation has modified them to transport antibiotics. Based on their overall structure, mechanism, and sequence homologies, these transport proteins are classified into four families: (1) the major facilitator family; (2) the resistance nodulation division family; (3) the staphylococcal multidrug resistance family; and (4) the ATP-binding cassette (ABC) transporters. Of these four families, only the ABC transporters use the chemical energy generated from the hydrolysis of ATP to drive molecules across the membrane. Members of the three other families use an electrochemical proton gradient, or proton-motive force, as the source of energy (18, 19). Some active efflux systems exclude a variety of unrelated toxins from the cell. These multidrug resistance (MDR) efflux systems in bacteria are comparable to those found in mammalian cells (see Drug Resistance). An example of a bacterial MDR efflux system is the Bmr transporter that transports drugs which have diverse chemical structures and physical properties and include cationic dyes, rhodamine-6G, ethidium bromide, and the antibiotics netropsin, puromycin, and fluoroquinone (20). Other MDR efflux systems characterized in bacteria include NorA in Staphylococcus aureus, MexB in Pseudomonas aeruginosa, and EmrB in Escherichia coli. If MDR efflux systems occur extensively in bacteria as the source of many antibiotic resistances, they pose a far more formidable challenge than more specific mechanisms of resistance. 3.3. Modification of the Antibiotic''s Target Antibiotics inhibit bacterial growth by inactivating different key proteins that are essential for bacterial survival (see Fig. 1). However, the antibiotic sensitivity of these target proteins can be altered. Typically, antibiotic targets are altered by reducing their affinity for the antibiotic. Bacteria accomplish this change in affinity several ways. Bacteria acquire exogenous DNA for a mutated target protein that no longer interacts with the antibiotic, yet retains the original target protein''s function. Alternatively, bacteria''s endogenous genes can be mutated to achieve the same end. In contrast, DNA for novel modifying enzymes can be acquired to alter the antibiotic target posttranslationally, reducing its affinity for the antibiotic. Altering an antibiotic''s target protein directly at the DNA level is a common mechanism of target modification. An example of this modification is the mutation of genes for penicillin-binding proteins (PBPs). PBPs are transpeptidases, previously discussed, that catalyze the final step in bacterial cell wall synthesis. These PBPs have high affinity for penicillin and its derivatives, and the binding of penicillin permanently inactivates PBPs. Originating from both endogenous and exogenous DNA sources, mutated PBPs can have a lower affinity for penicillin. Therefore, PBPs are resistant to the antibiotic, yet still provide a crucial function in bacterial cell wall synthesis (21, 22). Another example of target modification via mutated DNA is a single amino acid mutation in the quinolone resistance-determining region of the DNA gyrase gene, gyrA, that can provide up to a 20-fold increase in quinolone resistance (23). In ad***ion to using mutated antibiotic targets, bacteria can acquire new genes that produce proteins that, in turn, alter antibiotic targets. A well-studied example is the resistance of Staphylococci to erythromycin. It is known that Staphylococci have acquired genes to produce a protein that methylates a residue on the 23S ribosome. The 23S ribosome is the target of erythromycin, but methylated 23S ribosome has a low affinity for erythromycin. This exogenous gene is expressed and prevents the binding of erythromycin to the ribosomes, making the bacteria erythromycin-resistant (24). Bibliography Bibliography 1. H. C. Neu (1992) Science 257, 1064õ?"1073. 2. J. Travis (1994) Science 264, 360õ?"362. 3. M. Castiglia and R. A. J. Smego (1997) J. Am. Pharm. Assoc. NS37, 383õ?"387. 4. G. H. Cassell (1997) FEMS Immunol. Med. Microbiol. 18, 271õ?"274. 5. E. P. Abraham and E. Chain (1940) Nature 146, 837. 6. H. W. Stokes and R. M. Hall (1989) Mol. Microbiol. 3, 1669õ?"1683. 7. C. M. Collis, G. Grammaticopoulos, J. Briton, H. W. Stokes, and R. M. Hall (1993) Mol. Microbiol. 9, 41õ?"52. 8. T. Udou, Y. Mizuguchi, and R. J. J. Wallace (1989) FEMS Microbiol. Lett. 48, 227õ?"230. 9. K. J. Shaw et al. (1992) Antimicrob. Agents Chemother. 36, 1447õ?"1455. 10. P. N. Rather, E. Orosz, K. J. Shaw, R. Hare, and G. Miller (1993) J. Bacteriol. 175, 6492õ?"6498. 11. M. H. Richmond and R. B. Sykes (1973) Adv. Microb. Physiol. 9, 31õ?"88. 12. K. Bush (1989) Antimicrob. Agents Chemother. 33, 259õ?"276. 13. A. Tomasz (1979) Annu. Rev. Microbiol. 33, 113õ?"137. 14. D. M. Livermore (1993) J. Antimicrob. Chemother. 31 (suppl. A), 9õ?"21. 15. J. Y. Jacobs, D. M. Livermore, and K. W. M. Davy (1984) J. Antimicrob. Chemother. 14, 221õ?"229. 16. K. Coleman et al. (1994) J. Antimicrob. Chemother. 33, 1091õ?"1116. 17. P. R. Cullis and M. J. Hope (1985) In Biochemistry of Lipids and Membranes (D. E. Vance and J. E. Vance, eds.), Benjamin and Cummings, New York, Chap. "2". 18. S. B. Levy (1992) Antimicrob. Agents Chemother. 36, 695õ?"703. 19. K. Lewis, D. C. Hooper, and M. Ouellette (1997) ASM News 63, 605õ?"610. 20. A. A. Neyfakh, V. E. Bidnenko, and L. B. Chen (1991) Proc. Natl. Acad. Sci. USA 88, 4781õ?"4785. 21. B. G. Spratt and K. D. Cromie (1988) Rev. Infect. Dis. 10, 699õ?"711. 22. J. M. Ghuysen (1991) Annu. Rev. Microbiol. 45, 37õ?"67. 23. G. A. Jacoby and A. A. Medeiros (1991) Antimicrob. Agents Chemother. 35, 1697õ?"1704. 24. R. Leclercq and P. Courvalin (1991) Antimicrob. Agents Chemother. 35, 1267õ?"1272. Suggestions for Further Reading 25. A short news article: J. Davies (1996) Bacteria on the rampage, Nature 383, 219õ?"220. 26. Two well written reviews in an excellent anthology: 1. A. Bauernfeind and N. H. Georgopapadalou (1995) In Drug Transport in Antimicrobial and Anticancer Chemotherapy (N. H. Georgopapadalou, ed.), Dekker, New York, Vol. 17, pp. 1õ?"19. 2. R. E. W. Han**** (1995) In Drug Transport in Antimicrobial and Anticancer Chemotherapy (N. H. Georgopapadalou, ed.), Dekker, New York, Vol. 17, pp. 289õ?"306. 27. Three exhaustive articles in a single issue of Science devoted to antibiotic resistance: 1. J. Davies (1994) Inactivation of antibiotics and the dissemination of resistance Genes, Science 264, 375õ?"382. 2. H. Nikaido (1994) Prevention of drug access to bacterial targets: Permeability barriers and active efflux systems, Science 264, 382õ?"388. 3. B. G. Spratt (1994) Resistance to antibiotics mediated by target alterations, Science 264, 388õ?"393. Encyclopedia of Molecular Biology Copyright â1999 by John Wiley & Sons, Inc.
