Enzyme-Mediated Resistance to Antibiotics: Mechanisms, Dissemination, and Prospects for Inhibition

This chapter focuses on varied mechanisms by which aminoglycosidemodifying enzymes (AMEs) confer resistance to aminoglycosides. Clinical resistance to aminoglycosides emerges by one of four mechanisms: (i) loss of cell permeability (decreased uptake); (ii) alterations in the target site (ribosome) that prevent binding; (iii) expulsion by efflux pumps; and (iv) enzymatic inactivation by AMEs. Inhibition of AMEs by drugs such as dibekacin (a semisynthetic aminoglycoside), 5-epi-sisomicin, and 5-epi-gentamicin is discussed in the chapter that ensues. The major thrust of current efforts at AME inhibition has targeted the APH enzymes. Dimeric aminoglycosides and construction of amino-glycosides that undergo a cyclic process of phosphorylation and release of inorganic phosphate are exciting developments. In understanding these mechanisms, the reader should keep in mind the microbiological difference between in vitro synergism and clinical trials showing the benefits of synergistic therapy. Studies of the structure of the ribosome have resulted in an intense interest in aminoglycosides and macrolides as therapeutic agents. The lessons learned from the study of the mechanism of action and resistance of these agents will undoubtedly lead to further development of novel inhibitors of protein synthesis.

“Aminoglycoside (AG) antibiotics” is a large group of antibiotics that inhibit protein biosynthesis in bacteria. This chapter focuses on the general aspects of AG antibiotics and associated resistance factors with particular attention to the literature and developments between 2000 and 2003. AG antibiotics are effective against both gram-negative and gram-positive bacteria. Popular AG antibiotics such as neomycins, paromomycin, kanamycins, gentamicins tobramycin, and amikacin belong to the group of amino-glycosides that contain the 2-deoxystreptamine core moiety. There are several reviews on the activities, structure, resistance, and side effects of AGs. The chapter talks about the molecular mechanism of action, resistance to AGs, epidemiology, and clinical importance. The enzymatic mechanism is the most important one due to its prevalence among various pathogenic bacteria. ANTs catalyze the modification of an AG via transfer of a nucleotide monophosphate (NMP) from a corresponding nucleotide triphosphate (NTP). Among ANTs, ANT(4') and ANT(2')-I are the most mechanistically studied ANTs. ANTs also appear to be part of bifunctional enzymes, which would possess two different types of AG-modifying activities. Members of AACs do not have close sequence homologies to the other two classes of AG-modifying enzymes, ANTs and APHs. Understanding complex ecological, biochemical, and molecular origins of antimicrobial resistance mechanisms in relation to antibiotics, their development, and their dissemination in addition to developing newer inhibitors is important to continue to use effective drugs such as AGs in the clinic.

Three structurally and functionally unrelated classes of inactivation enzymes are known: the aminoglycoside kinases (APH), the adenylyltransferases (ANT), and the acetyltransferases (AAC). The past decade has seen a dramatic increase in our understanding of the structure and function of members of each of these classes, and each is discussed separately in this chapter. It was subsequently demonstrated that APHs maintain vestigial protein kinases activity. In fact, since the structure determination of APH(3’)-IIIa, which represented the first distant relative of the protein kinases, this superfamily has further expanded and includes now lipid kinases and choline kinases. One of the most intriguing differences when comparing Ser/Thr and Tyr kinases with aminoglycoside kinases is the region of the enzymes involved in substrate binding. In fact, the activation loop is completely lacking in APH(3’)-IIIa, and vice versa there are no remnants of an aminoglycoside-binding loop segment in protein kinases. Aminoglycoside adenylyltransferases (ANTs) catalyze the transfer of AMP to aminoglycoside hydroxyl groups. Other members of the GNAT family include other small-molecule acyltransferases such as serotonin acetyltransferase and protein acetyltransferases such as the histone acetyltransferases. The realization following the determination of enzyme three-dimensional structure that aminoglycoside resistance enzymes are members of larger families of proteins that share similar structures and mechanisms has permitted insight into the origins and evolution of resistance. This information will arm us in the continuing efforts to meet the challenge of resistance at the molecular level and apply this work to the management of infectious disease.

Aminoglycoside antibiotics are broad-spectrum bactericidal agents that have predictable pharmacokinetics and are used mainly in the treatment of infections caused by gram-negative aerobic bacilli and some gram positives. Aminoglycoside antibiotics are commonly administered by intramuscular injection but also intravenously in cases of severe infections. The aminoglycoside-modifying enzymes are adenylyltransferases (ANTs), phosphotransferases (APHs), also known as kinases, and acetyltransferases (AACs). The three-dimensional structures of two aminoglycoside O-phosphotransferases, APH(3p)-IIIa, from grampositive cocci, and APH(3p)-IIa, from Enterobacteriaceae, have been determined. ANTs catalyze the transfer of an AMP group from the substrate ATP to a hydroxyl group in the aminoglycoside molecule. Acetyltransferases (AACs) belong to the GCN5-related N-acetyltransferase superfamily of proteins, which spans all kingdoms of life and comprises several unrelated kinds of enzymes with a wide variety of functions. The presence of genes coding for aminoglycoside-modifying enzymes within many of these elements permitted them to disseminate at the molecular as well as at the cellular level, contributing to the rise of multiresistant bacteria. Protein kinase inhibitors that disable aminoglycoside-modifying enzymes by targeting the ATP-binding site are showing very promising prospects for their use in combination with aminoglycosides. Antisense oligonucleotide-mediated inhibition of expression of genes coding for aminoglycoside-modifying enzymes shows promise for development as another alternative against resistance.

All the macrolides, lincosamides, streptogramins, ketolides, and oxazolidinones (MLSKO) antibiotics inhibit protein synthesis by binding to the 50S bacterial ribosomal subunit, close to the peptidyltransferase center. The last group of acquired genes, and the focus of this chapter, are those that encode an adenine-N6-methyltransferase (rRNA methylase). Currently, 32 different genes coding for enzymes that modify the 23S rRNA have been described in the chapter. This modification blocks the binding of the MLSK group of antibiotics and allows the bacterial ribosomes to continue to produce protein in the presence of macrolides, lincosamides, and streptogramin B. Mobile elements often carry multiple different genes which confer antibiotic resistance to a variety of different antibiotic classes. Association of the erm genes with mobile elements (plasmids, conjugative transposons, and transposons) provides the potential to move from one species to another, one genus to another, and one ecosystem to another, between bacterial hosts, and from food to man and the environment. If DNA-DNA hybridization is used, confirmation by a PCR assay is advisable. One can speculate that as the newer macrolide derivatives became available in the 1980s this influenced the increase in carriage of macrolide resistance genes. It can be predicted that new innate genes may be found, which when mutated could provide increased levels of resistance to all the MSLKO antibiotics. It is even possible that new mechanisms of MLS resistance will be described in the coming years.

β-lactamases are arguably the most well-studied antibiotic resistance determinants that have been described, with over 450 unique enzymes currently identified. Molecular relationships, as well as functional relationships, exist between the β-lactamases and penicillin binding proteins (PBPs), the essential enzymes involved in the terminal stages of bacterial cell wall synthesis. Two sets of β-lactamases classification schemes have evolved, based on either structural or functional descriptions of the enzymes. Fortuitously, the two schemes fit together according to major divisions but show some diversity among various subgroups for each of the divisions as shown. Microbiological methods have traditionally indicated the potential for β-lactamase production, based on susceptibility profiles for β-lactam-containing agents. In multiple surveys conducted during the late 1970s the most commonly identified plasmid-encoded β-lactamases among the Enterobacteriaceae were the broad-spectrum (group 2b) SHV-1 and TEM-1 or TEM-2 β-lactamase, followed by the oxacillinhydrolyzing OXA-1 enzyme. In the current age of rapid nucleotide sequencing, molecular and epidemiological characterizations of numerous β-lactamase genes in a wide variety of organisms have been reported. The chapter describes important new enzymes characterized in the published literature since the year 2000. The reported incidence of these β-lactamases depends, in part, on the laboratory technique used for detection by the clinical laboratory, and the type of β-lactamase varies by country and by region. Finally, the search for new β-lactamases in environmental isolates will serve to expand this important family of enzymes that are expected to continue as a major chapter in the story on antibiotic resistance.

The transpeptidases are members of the family of penicillin-binding proteins (PBPs), which have become known as the targets of β-lactam antibiotics. PBPs perform a variety of critical functions for the bacterial cell. PBPs are found in all pathogenic bacteria except those of the genus Mycoplasma, which do not have a cell wall. Reduced expression of PBP2 (type B, assignment not certain) is one of the most frequently observed mechanisms of resistance to carbapenems. PBP2 (type B4) and PBP3 (type B5) have been specifically implicated in resistance towards β-lactams. Increased expression of PBP4 (type C2) has been implicated in resistance to both β-lactams and glycopeptides. Hakenbeck et al. implicated the D,D-carboxypeptidase S. pneumoniae PBP3 (type C3) in resistance. PCR-based methods for the detection of antibiotic resistance are becoming increasingly important with the expanding use of molecular techniques for bacteriological diagnosis. Antibody-based tests have also been investigated for detection of methicillin resistance in staphylococci. There are several experimental β-lactams now known to be potent inhibitors of the staphylococcal type B1 PBP that is the primary determinant of β-lactam resistance in these organisms, of which ceftobiprole is the most advanced in clinical development. An understanding of the mechanism of methicillin resistance has led to the discovery of accessory factors that influence the level and nature of methicillin resistance.

The catalytic function of β-lactamases is the primary mechanism of bacterial resistance to β-lactam antibiotics (penicillins, cephalosporins, carbapenems). β-lactamases hydrolyze the β-lactam bond of these antibiotics, a structure modification that abrogates the antibacterial activity. β-lactams include tazobactam, a highly effective sulfone penam inhibitor, penicillanic acid sulfone sulbactam, 6-β-bromopenicillanic acid, and thienamycin. Clavulanate is a potent inhibitor of class A β-lactamases, which incidentally exhibits weak antimicrobial activity as well. A series of molecules-using sulfoxide and sulfone penams have been synthesized as starting points-with sulfhydryl and sulfide moieties at C-6; the goal of this exercise was to arrive at molecules that would simultaneously inhibit classes A and B of β-lactamases. The study also confirmed that the sulfone oxidation state of the penam thiazolidine resulted in greater inhibition. The success of BRL 42715 prompted additional efforts into compounds with a double bond at C-6, leading to the discovery of SYN-1012-with a methyl triazolyl moiety at C6 instead-and another more recent methylidene penem-with a bicyclic and heterocyclic moiety at C-6; both of these compounds show good activity against class A and C β-lactamases. Several routes have been taken towards the development of more effective inhibitors including the syntheses of variants of penam sulfones, penems, alkylidenes, monobactams, transition-state analogs, and the boronates.

This chapter provides an overview of class B β-lactamases (CBBLs), considering both fundamental and clinical aspects. Class B is one of the four classes in the structural classification of β-lactamases, which was created to accommodate the metallo-β-lactamases (MBLs). Expression of the CBBL genes carried on gene cassettes is normally under the control of the integron promoters (Pc and, possibly, P2) located in the 5'-conserved segment of the integron. The constant features of CBBLs include (i) good to excellent carbapenemase activity; (ii) lack of activity on monobactams, which apparently do not interact with these enzymes; and (iii) inhibition by EDTA and other metal ion chelators, and lack of inhibition by the conventional serine-β-lactamase inhibitors. Concerning the structure of the zinc center, which is located at the bottom of a shallow groove between the two β-sheets, in subclass B1 and B3 CBBLs it can accommodate two zinc ions (dinuclear zinc center): a tetrahedrally coordinated zinc ion (Zn1) and a trigonal bipyramidally coordinated zinc ion (Zn2), bridged by a water molecule/hydroxide ion (Wat1). Zn1 is coordinated by three His residues (His 116, His118, and His196) and the bridging Wat1 in all subclass B1 and B3 enzymes, while the strategy of Zn2 coordination is partially different in members of the two subclasses. The clinical relevance of resident CBBLs essentially reflects that of the host species, where these enzymes can variably contribute to intrinsic β-lactam resistance depending on their expression pattern and substrate specificity.

The class C enzymes are considered to be the most homogeneous and least effective among the different groups of β-lactamases. The design of novel inhibitors is discussed in this chapter. Besides the “classical” AmpC class C β-lactamases, two related enzymes can be mentioned. The first is an ES class C β-lactamase (Enterobacter cloacae GC1) that deserves special attention from a structural point of view, to understand both its activity and inhibition potential. The second is a cold-adapted enzyme from a psychrophilic microorganism (Psychrobacter immobilis A5) that could lead to a better understanding of bacterial adaptation to temperature. Structures of complexes with class C β-lactamases were used to approach the mechanism of these enzymes. Several consensus binding sites were identified from the crystal structures, and predictions by the computational programs showed some correlation with the experimentally observed binding sites. To investigate the structural bases of these energies, X-ray crystal structures of N289A/13 and N289A/14 were determined to 1.49 and 1.39 Å, respectively. Crystallographic structures of a large number of class C β-lactamases have been reported and studied. As was presented in “Structural Features of the Active Sites of Class C β-Lactamases and Mechanistic Considerations” above, those structures produced a clear picture of different intermediates along the catalytic route of class C β- lactamases. The section entitled “Complexes with Class C β-Lactamases and Drug Design” selected a number of examples where structure-based drug design approaches were successful in identifying and optimizing original inhibitors.

β-lactams are the most widely used antimicrobial agents. This success reflects their broad spectrum, low toxicity, limited side effects, bactericidal action, good pharmacokinetic properties, and the diversity of useful analogs that can be synthesized, including penams, cephems, oxapenems, carbapenems, and monobactams. Most of the class D β-lactamases found in important pathogens are encoded by acquired genes, and the dissemination of their host plasmids defines their distribution. In this article, the authors have adopted the DBL numbering system as a general scheme and with the actual position in the sequence of a specific enzyme indicated afterwards in brackets. Both OXA-10 and OXA-13 have a single disulfide bridge between Cys 45 (44 OXA-10) and Cys 54 (51 OXA-10), linking β-sheets 2 and 3, and this is conserved in all other group I enzymes except the carbapenemases OXA-48, -54, and -55. Class D β-lactamases are similar to class A and C types in their three-dimensional structure, with the three same conserved elements, positioned similarly. OXA-10 β-lactamase, the most prevalent member of group I, hydrolyzes a wide range of β-lactams, including aztreonam, cefotaxime, and ceftriaxone but not ceftazidime. Once cloned into Escherichia coli, OXA-28 was again associated with resistance to ceftazidime, and reduced susceptibility or resistance to cefotaxime. This enzyme hydrolyzed cloxacillin, benzylpenicillin, cefotaxime, and ceftazidime, but not piperacillin; clavulanic acid had an IC50 of 10 μM. OXA-28 had a Trp164Gly substitution, reinforcing the view that the Trp164 side chain ordinarily limits the spectrum of OXA-10-related enzymes.

The discovery of novel β-lactamases and penicillin-binding proteins (PBPs) often requires kinetic characterization. As such, the rate at which a β-lactamase hydrolyzes a β-lactam is influenced by several factors. The first is concentration of β-lactam, which is designated [S] and is expressed in units of molarity. The second is temperature. As the temperature rises, molecular motion, and hence collisions between β-lactamase and β-lactam, and the rates of interconversion of intermediates increase. The third factor is the presence of inhibitors. β-lactamase inhibitors are clinically used to hinder the activity of the β-lactamase. The last is pH: the charge of active-site groups and the conformation of a protein are influenced by pH, and enzyme activity is crucially dependent on both these factors. The equations of enzyme kinetics are conceptual tools that allow us to interpret quantitative measurements of enzyme activity. Nitrocefin is the most practical reference compound, since the accumulation of ER* can be monitored at 480 to 490 nm and no interference is expected with most other β-lactams which do not yield acyl enzymes absorbing in this wavelength range. Several class D β-lactamases also exhibit substrate-induced inactivation (or biphasic kinetics) with a significant number of substrates. Indeed, in some but not all cases, the substrate-induced inactivation disappeared in the presence of a saturating concentration of bicarbonate which was assumed to completely maintain the Lys in the carboxylated form.

One might argue that the existence of the modern pharmaceutical industry relies on the discovery of compounds first identified by their ability to inhibit the growth of bacteria on agar (e.g., sulfonamides and penicillin). Some, including the authors, argue that the new technologies are not yet mature enough to deliver antibacterials, especially given the safety hurdles for antibacterials (currently marketed products are among the safest ever produced) and the desire for broad-spectrum activity expressed by the marketers and many physicians. An extension of the concept of essential gene targeting is the idea of attacking an essential metabolic pathway. The potential advantages of such a strategy for avoidance of the emergence of resistance are obvious. A number of approaches to this have been elucidated and are discussed in this chapter. The chapter examines the issue of virulence functions as targets. The concept of protein-protein interactions as drug targets has been a controversial one in the pharmaceutical industry. The distinctive nature of virulence inhibition as a method to eradicate bacterial infection will likely require special tools in the drug development process. The topic of enzyme inhibitors as antibacterial therapeutics must be taken in the context of an overall decrease in the effort to discover such therapies in the pharmaceutical industry. Given the difficulties outlined in the chapter, the challenge for researchers to meet the demands of physicians and their patients for therapies for infections increasingly caused by resistant pathogens becomes all that much greater.

This chapter discusses structure and function studies of β-lactamase inhibitory protein (BLIP) with a focus on the interactions between TEM-1 β-lactamase and BLIP. The first proteinaceous inhibitor of β-lactamases, BLIP, was isolated from Streptomyces clavuligerus by a researcher's group in 1990. A BLIP nonproducer mutant and a BLIP/clavulanic acid nonproducer double mutant were constructed with the aim of elucidating BLIP's physiological function. Two hypotheses have been proposed for the function of BLIP. First, BLIP may be produced in response to the production of β-lactamases by other organisms in the surrounding environment in order to inhibit these β-lactamases and prevent the hydrolysis of antibiotics produced by S. clavuligerus. Alternatively, BLIP may play a role in cell wall growth or morphogenesis. Protein-protein interactions play a significant role in most cellular processes. The importance of such interactions in biology has made protein-protein recognition an area of considerable interest. The two domains join with each other to form an 8-strand antiparallel β-sheet. The study of a series of homologous interfaces provides an opportunity to study specificity as well as overall affinity determinants.

The major facilitator superfamily (MFS) comprises drug-specific pumps like the tet- and otrB-encoded tetracycline transporters found in many gram-positive bacteria and gram-negative bacteria. The majority of macrolide-resistant pneumococci in the United States, Canada, South America, Hong Kong, Singapore, Thailand, and Malaysia were cases of mef-mediated resistance. Fluoroquinolone resistance in gram-negative bacteria due to target site alteration can be substantially supported by efflux. Multiple-antibiotic-resistant (mar) mutants of Escherichia coli have decreased outer membrane permeability due to reduced porin expression and at the same time show increased efflux. An interesting aspect of the VceC structure is that the resolved positions of two octyl-β-glucoside molecules were as expected for lipopolysaccharides (LPS) of the bacterial outer membrane. The majority of the multidrug and toxic compound extrusion family (MATE) transporters identified until now use the electro-chemical potential of Na+ across the membrane to drive multidrug export. Multidrug ABC transporters contribute to multiple antibiotic resistance in bacteria and cause multiple-cancer-drug resistance in humans. Apart from contributing to resistance against anticancer chemotherapy, drug efflux is highly relevant for the successful treatment of bacterial infections by tetracycline, macrolide, and fluoroquinolone antibiotics. NorA in S. aureus, PmrA in S. pneumoniae, and (clinically less relevant) Bmr and Blt in B. subtilis are clear examples of the high relevance of drug efflux for fluoroquinolone resistance in gram-positive bacteria. The MFS members CmlA and FloR specifically export chloramphenicol and the structurally related veterinary drug florfenicol. They confer inducible resistance in many gram-negative bacteria including P. aeruginosa, H. influenzae, and the Enterobacteriaceae.

This chapter provides an overview of dissemination mechanisms of genes coding for resistance to antibiotics. The emergence of antibiotic resistance determinants would not be as devastating if it were not for the inherent ability of bacteria to exchange genes at the cellular and molecular level. A plasmid genome database containing all sequenced plasmids was recently established. While plasmids participate in bacterial antibiotic resistance mainly by disseminating genes coding for drug resistance at the cellular level, other elements promote gene exchange at the molecular level. Integrative and conjugative elements (ICEs) encode diverse excision, recombination, and conjugation systems, in addition to specific functions, including resistance to antibiotics. The combination of dissemination at the cellular level through conjugation, natural transformation, transduction, with dissemination at the molecular level, and mutagenesis permits genes coding for antibiotic resistance to reach virtually all bacterial cells, resulting in a virtual elimination of barriers between types of bacteria. However, while the acquisition of resistance genes provides an advantage to the bacterial cells when in the presence of antibiotics, it has been shown that their presence comes with an associated fitness cost when the cells are growing in the absence of antibiotic selective pressure.

Transfer of conjugative transposons requires cell-to-cell contact and occurs at relatively low frequency. However, the host range of the various elements is, in some cases, quite diverse. In this chapter, the author reviews the current state of knowledge of the epidemiology, resistance profiles, and transposition mechanisms of the major types of characterized conjugative transposons found in human pathogenic bacteria. A recently published survey directly implicates conjugative transposons in the steady and significant rise in erythromycin and tetracycline resistance in Bacteroides species over the past three decades. Unlike Tn916 family transposons, Bacteroides conjugative transposons can mobilize coresident nonconjugative transposons. Bacteroides conjugative transposons can be transferred in vitro into distantly related species, such as Escherichia coli. The Tn916 family elements are the most thoroughly studied and characterized of the conjugative transposons. The single "family" of Bacteroides conjugative transposons described to date is the CTnDOT family. This family is characterized by the presence of right and left ends that differ from each other but are conserved within the family. The rteA, rteB, and rteC genes are also involved in excision, presumably through a regulatory role. rteA and rteB are required for expression of rteC. Expression of rteC increases transcription of exc, thereby increasing excision of CTnDOT. Recent advances in understanding the structural details of λ and related integrase interactions with DNA targets and cooperative proteins have the potential to stimulate intelligent, structure-based design of such chemical inhibitors.

In areas where trachoma is highly endemic, that is, where about 70% of the population is infected, many individuals typically carry Streptococcus pneumoniae. Of these, 2% were originally macrolide resistant. After 2 weeks of trachoma treatment, the carrier rate for macrolide-resistant pneumococci rose to 50%, with an increase in the carrier rate of the more highly resistant strains. This study dramatically illustrates the potential for rapid in vivo selection of resistance. In this study it was also found that one chronically ill child was carrying a strain with very high level resistance. This chapter defines gene dissemination mechanisms and shows how and why conjugative mechanisms are the most proficient in multiple drug resistance (MDR) transfer. The plasmid is an interesting variation on the theme of MDR dissemination. The chapter reviews aspects of bacterial conjugation in terms of basic science and clinical relevance. A unifying model can be developed because the general mechanism for classical bacterial conjugation appears to be conserved in conjugative transposons. It is reasonable to suggest that bacterial conjugation is the greatest mover of genes in the microbial world and, in the clinical world, that these genes are often antibiotic resistance genes. The cycle of antibiotic resistance and pathogen genome sequencing are showcasing the prominent role of bacterial conjugation in gene dissemination.

This chapter focuses on the proteic toxin-antitoxin (TA) systems. TA systems function as vitally important regulatory systems in bacteria and represent ideal targets for the development of novel antibiotic therapeutic agents. A broad mechanistic understanding of TA systems at physiological, biochemical, biophysical, and structural levels provides the scientific framework needed both for rational drug design and for elegant selection schemes using large pools of compounds. The proteins of chromosome-encoded TA systems (relBE, yefM-yoeB, and dinJ-yafQ) from gram-negative bacteria, namely, CcdA-CcdB, Phd-Doc, ParD-ParE, YefM-YoeB, and one system from a plasmid from a G+ bacterium, have been studied in vitro with respect to their properties in solution and binding to DNA. The chapter summarizes the knowledge accumulated on these proteins. Pathogenic bacteria are subjected to an enormous selective pressure because of the indiscriminate overuse and misuse of broad-spectrum antibiotics. The recognition of the importance of protein-protein interactions within the cell has led to their investigation as targets for novel inhibitors. Here, the approaches that can be used for screening of inhibitors of protein-protein interactions are highlighted by recent research on the TA systems. The chapter focuses on two resonance energy transfer techniques, namely, fluorescent resonance energy transfer (FRET) and, especially, bioluminescence resonance energy transfer (BRET), since they have demonstrated to be highly useful for studying interactions between two proteins that have been shown to form complexes.

Numerous studies have appeared that describe the importance of resistance integrons (RIs) and superintegrons (SIs) in antibiotic resistance, microbial physiology, and environmental adaptation in phylogenetically diverse gram-negative bacteria. This chapter describes the genetic organization of an integron; summarizes the different classes of RIs and highlights their importance in antibiotic resistance; describes the organization of an SI; and highlights the structure of the key component of the integron, the integrase, and its binding to an attC site. There are two types of recombinases: tyrosine recombinases (integrases) and serine recombinases (resolvases or invertases). The attC site is an imperfect inverted repeat located at the 3' end of the gene. Cassettes are always integrated in the same orientation and are cotranscribed from one or two common promoters (P1 or P2) located in the 5' conserved segment (CS). To date, five distinct integron classes have been found associated with cassettes that contain antibiotic resistance genes. Three main classes of integrons (classes 1, 2, and 3) have been described in gram-negative bacteria. The similarity between the three integrases (40 to 58% genetic identity) suggests that their evolutionary divergence extended beyond the introduction of antibiotics into clinical medicine.

This chapter focuses on the impact of antibiotic resistance caused by chromosomal mutations on bacterial fitness. An approach to estimate the biological costs associated with resistance is to use epidemiological data to prospectively follow the rate at which a patient infected with a resistant or susceptible bacterial strain transmits it to other people. Such experiments would allow one to measure the basic reproductive number, which is the most relevant parameter to use when predicting the relative rate of spread of the resistant and susceptible bacteria. Chromosomal mutations alter the intracellular level of the transcriptional regulator molecule ppGpp, which might cause additional pleiotrophic fitness effects. Certain mutational alterations in ribosomal protein S12 (encoded by the rpsL gene) causing streptomycin resistance reduce translational efficiency. In isoniazid-resistant Mycobacterium tuberculosis, katG mutants with decreased fitness can be compensated by overproduction of another enzyme that may substitute for the defective catalase. But the most common compensation mechanism is restoration of the function itself, either by intragenic or extragenic mutations. A final implication emerging from studies of fitness costs and their genetic compensation concerns the development of new antibiotics. At present, the key parameter from a resistance development point of view that is considered by drug developers is the rate of appearance of the initial resistance mutation (or plasmid). Even though these rates do influence the rate of resistance development, their importance might be overestimated.