Molecular Biology of Resistance: A Brief History of Resistance Mechanisms and the Discovery of Gene Transfer



Molecular Biology of Resistance: A Brief History of Resistance Mechanisms and the Discovery of Gene Transfer


Gopi Patel

Robert A. Bonomo



The widespread availability of penicillin in the 1940s is credited with the survival of many wounded soldiers in World War II. This feat marked the beginning of the antibiotic era. During the next several decades, robust antimicrobial development provided the opportunity to safely and effectively treat a variety of pathogens. Along with every new antimicrobial agent came the inevitable development of resistance. With the increasing number of novel and potent antimicrobials being developed, sporadic antimicrobial resistance resulting from random spontaneous mutations was not initially considered a serious threat. The description of penicillinases (enzymes capable of destroying penicillin) in Escherichia coli (1) followed by Staphylococcus aureus (2) suggested that antimicrobial resistance may be more complex. Active efflux of antimicrobials (3), drug target modification (4), and bacterial expression of drug-modifying enzymes (5) have collectively contributed to the intricate web of antimicrobial resistance. The introduction of carbapenems in the 1980s promised to be a solution for patients and practitioners struggling with the treatment of severe health care-associated drug-resistant gram-negative infections. Unfortunately, reports of resistance to these last resort antibiotics are no longer rare, and despite increasing accounts of resistance and the growing clinical isolation of multidrug-resistant strains, the number of new antibiotics with original mechanisms of action in development is discouraging and many are predicting the end of the era of antibiotics (6).

In the late 1950s, the remarkable finding that susceptible Shigella species were able to acquire a multiple drug resistance phenotype (7) introduced the concept of horizontal transmission of resistance genes and suggested that these resistance determinants could be shuttled between species. Resistance in this case was coupled with the presence of a plasmid, which through the process of conjugation could be transferred between bacteria. Plasmid-mediated transfer of multiple resistance determinants appears to be responsible for much of the multidrug resistance seen in gram-negative bacilli today.

Subsequently, in the 1960s, bacteriophages (or viruses capable of infecting bacteria) were noted to be capable of transferring antimicrobial resistance genes and even small plasmids between bacteria through a process called transduction (8). This was followed in the 1970s by the description of transposons, chromosomal or plasmid-borne transposable elements with the ability to move independent from normal DNA recombination (9). In the 1980s, the basic structure of the integron was reported. Integrons are mobile genetic elements containing antibiotic-resistance genes as well as a site-specific recombination system capable of inserting these genes into plasmid or chromosomal DNA (10,11). In the late 1990s emerged the description of the staphylococcal chromosomal cassette (scc), responsible for housing and transferring the genes associated with methicillin resistance in S. aureus (methicillin-resistant Staphylococcus aureus; MRSA) (12). More recently, an 86 kb “resistance island” capable of accommodating a large number of antimicrobial and heavy metal resistance determinants has been described in an epidemic multidrug-resistant strain of Acinetobacter baumannii (13).

Understanding the growing complexity of antimicrobial resistance may assist in the development of novel agents, rapid and accurate diagnostics, and infection control recommendations.


MECHANISMS OF RESISTANCE


INTRINSIC RESISTANCE

Absence of an intended target, decreased permeability, active drug efflux, or chromosomal expression of a drug-modifying or inactivating enzymes are responsible for intrinsic resistance to some antimicrobials (14). For example, Mycoplasma spp. lack a peptidoglycan cell wall and thus are inherently resistant to penicillins, cephalosporins, and other β-lactam antibiotics. Most enterococci demonstrate low levels of aminoglycoside resistance because of the inability of the drug to traverse the cell wall’s peptidoglycan layer to reach their target, the ribosome. However, in the presence of inhibitors of cell wall synthesis, such as ampicillin or vancomycin, aminoglycosides can reach their site of action and provide a synergistic bactericidal effect (15). The size and structure of vancomycin prevents it from being able to navigate the outer membrane of gram-negative bacilli, thus limiting its activity to gram-positive pathogens. Chromosomally encoded enzymes with the ability to modify or destroy antimicrobials also can be responsible for innate resistance. For example, AmpC β-lactamases are intrinsic to several clinically relevant gram-negative pathogens including Enterobacter spp., Citrobacter freundii, Serratia marcescens, and Pseudomonas aeruginosa (16). In the absence of β-lactam exposure, these enzymes can be expressed at low levels. Spontaneous mutations in regulatory regions can lead to derepression and subsequent
enzyme hyperproduction, thus rendering these bacteria resistant to most commonly employed penicillins, cephalosporins, cephamycins, aztreonam, as well as commercially available β-lactam/β-lactamase inhibitor combinations.


ACQUIRED RESISTANCE

Of greater concern is the development of resistance among traditionally susceptible bacteria through mutation or through the acquisition of new genetic material. Transfer of genetic material often is mediated by integrons, transposons, or plasmids. These mobile DNA elements can encode multiple determinants, thus conferring resistance to more than one type or class of antibiotics. The resultant multidrug resistance severely handicaps practitioners treating patients infected with these pathogens. Examples of acquired resistance mechanisms for several classes of antimicrobial agents are described in the following sections with a focus on common clinically relevant mechanisms as well as mechanisms for multidrug resistance among gram-positive and gram-negative bacteria.


MECHANISMS OF RESISTANCE TO SELECTED ANTIBIOTICS


RESISTANCE TO β-LACTAM AGENTS

β-lactam agents include penicillins, cephalosporins, monobactams, and carbapenems. These agents bind to penicillin-binding proteins (PBPs), which are enzymes (e.g., transpeptidases) involved in the synthesis and maintenance of the peptidoglycan cell wall of gram-positive and gram-negative bacteria. β-lactam resistance can be conferred by decreased access to the PBPs through efflux or outer membrane porin mutations, reduced PBP affinity for β-lactams, and by the expression of β-lactamases.

As previously referenced, staphylococcal β-lactamases are narrow-spectrum penicillinases with relatively poor activity against semisynthetic anti-staphylococcal penicillins like methicillin and oxacillin, cephalosporins, and carbapenems. On the other hand, the majority of β-lactam resistance in clinically relevant gram-positive bacteria (e.g., streptococci, enterococci, and staphylococci) is secondary to expression of low-affinity PBPs (altered drug target).

Methicillin resistance in S. aureus (MRSA) is due to the expression of PBP2a (17). This low-affinity PBP is encoded by the mecA gene. This gene has been mapped to mobile chromosomal elements referred to as staphylococcal chromosomal cassettes (SCCmec) (12,18). Eleven variants of this element have been described (19,20,21,22). The sizes of these chromosomal cassettes vary, and smaller cassettes (e.g., SCCmec IV and SCCmec V) appear to carry fewer concomitant resistance genes (23,24). These cassettes contain the mec gene complex as well as unique cassette recombinase genes (ccr gene complex) responsible for excision and integration (18). Contemporary SCCmec typing is based on variations in these components (19).

Similarly, penicillin resistance among Streptococcus pneumoniae is due to the expression of low-affinity PBPs from mosaic gene products (modifications in genes for PBP2b, 2x, and 1a). These mosaic genes are derived from DNA recombination between chromosomal PBP genes and those from less virulent streptococcal species (25,26). Ampicillin resistance in enterococci can be due to β-lactamase production (27), but high-level ampicillin resistance among Enterococcus faecium is the result of the expression of a low-affinity target, PBP5 (28,29,30). Mosaic genes resulting in altered PBPs are also responsible for penicillin and cephalosporin resistance among Neisseria gonorrhoeae (31,32,33,34).

In gram-negative bacilli, β-lactam resistance primarily results from the expression of β-lactamases (Table 16.1). This heterogenous group of enzymes efficiently hydrolyze the amide bond of the β-lactam ring, thus inactivating the β-lactams. These enzymes can be intrinsic (e.g., AmpC β-lactamases in Enterobacter cloacae and P. aeruginosa) or acquired (e.g., plasmid-mediated), and over 1,000 unique β-lactamases have been described with different structures and spectra of activity (http://www.lahey.org/studies/). These enzymes are classified by two different schemes: the Ambler molecular classification, based on the amino acid structure (35), and the the Bush-Jacoby-Medeiros functional classification scheme, based on the substrate and/or inhibitor profile (36). For the sake of simplicity, we will reference the Ambler classification scheme here.

Resistance to ampicillin among E. coli and Klebsiella pneumoniae is mediated by narrow-spectrum penicillinases (e.g., TEM-1 in E. coli and SHV-1 in K. pneumoniae) (37,38). The clinical introduction of third-generation cephalosporins in the 1980s, which were stable against these penicillinases, was soon followed by the description of β-lactamases capable of inactivating these broad-spectrum antibiotics. Remarkably, a single point mutation in the genes encoding the narrow-spectrum penicillinases was responsible for expanding the substrate spectra (39). These group of enzymes were designated as extended-spectrum β-lactamases, or ESBLs (40). The expression of ESBLs confers resistance to penicillins, first-, second-, and third-generation cephalosporins, and aztreonam. In the absence of other resistance mechanisms, cephamycins and carbapenems retain activity. Interestingly, ESBLs are readily inhibited by β-lactamase inhibitors (e.g., clavulanic acid). This property has been an important characteristic used for detection of ESBL production in clinical microbiology laboratories. Clinically, however, ESBLproducing K. pneumoniae and E. coli tend to be resistant to available β-lactam/β-lactamase inhibitor combinations (41,42).

Recently, there has been a dramatic shift in the prevalence and types of ESBLs identified both in the community and in many health care settings. The CTX-M family of β-lactamases, particularly CTX-M-15, have become the dominant ESBL type in Europe, North and South America, and Asia (43,44,45,46). Unlike SHV-type (e.g., SHV-2) and TEM-type ESBLS, which evolved primarily from mutations in plasmid-mediated penicillinases, CTX-M type ESBLs appear to have evolved from a chromosomal cephalosporinase of Kluyvera spp. (47). CTX-M-15 has been associated with a specific clone of E. coli, ST131 (44,48,49,50,51), that has contributed to the successful global dissemination of this particular resistance determinant. CTX-M type enzymes were considered rare in the United States a few years ago, but contemporary surveys demonstrate that its prevalence is increasing at an impressive rate (52,53). Many plasmids carrying ESBLs also harbor genes associated with fluoroquinolone and/or aminoglycoside resistance (44,54).

Carbapenems are considered the drug of choice to treat patients at risk for or with documented severe infections caused by ESBL-producing organisms, even in the setting of reported in vitro susceptibility, due to clinical failures observed with alternative therapy (55).









TABLE 16.1 Examples of Commonly Encountered β-lactamases among Epidemiologically Significant Gram-Negative Bacilli















































Ambler Class


Bush-Jacoby-Medeiros Class


Common Name


β-lactam Resistance Profile


Representative Examples


A


2b


Penicillinase


Penicillins, first generation cephalosporins (narrow spectrum)


SHV-1, TEM-1


A


2be


ESBL


Penicillins, cephalosporins, aztreonam, not inhibited by clinically available β-lactamase inhibitorsa


SHV-2, CTX-M


A


2f


Carbapenemase


Penicillins, cephalosporins, aztreonam, β-lactam/β-lactamase inhibitor combinations, carbapenems


KPC


B


3


Metallo-β-lactamases


All β-lactams including carbapenems EXCEPT aztreonam, not inhibited by β-lactamase inhibitors but inhibited by EDTAb


IMP, VIM, NDM-1


C


1


Cephalosporinase


Penicillins, cephalosporins, not inhibited not by β-lactamase inhibitors


AmpC β-lactamases


D


2df


Carbapenemase


Penicillins, cephalosporins, aztreonam, β-lactam/β-lactamase inhibitor combinations, carbapenems


OXA-23, OXA-48, OXA-58


a clavulanic acid

b ethylenediaminetetraacetic acid


Phenotypic carbapenem resistance in Enterobacteriaceae was initially considered sporadic, but during the last decade has become alarmingly more common. Carbapenem resistance results from one or more of the following mechanisms: derepression and hyperproduction of AmpC β-lactamases or ESBLs with concomitant alteration in outer membrane porins, augmented drug efflux, alterations in PBPs, and/or carbapenemase production (56).

Currently, most carbapenem resistance among Enterobacteriaceae in the United States and Israel is attributed to plasmid-mediated expression of a KPC-type (K. pneumoniae carbapenemase) carbapenemase. These Class A serine carbapenemases hydrolyze carbapenems as well as penicillins, cephalosporins, and aztreonam and are not overcome in vitro by clinically available β-lactamase inhibitors (57). KPCs have been identified in several species of Enterobacteriaceae as well as Pseudomonas spp. and A. baumannii (56), with KPC-2 and KPC-3 being most commonly isolated subtypes in Enterobacteriaceae. The blaKPC gene has been mapped to a Tn3 based transposon, Tn4401 (58), explaining the efficient transfer of these β-lactamases between strains and species. Five isoforms of Tn4401 have been described (59,60). Reports of many of the plasmids harboring blaKPC include genes conferring resistance to fluoroquinolones and aminoglycosides are unsettling and not uncommon (61,62).

KPC-production confers variable levels of carbapenem resistance with reported minimum inhibitory concentrations (MICs) ranging from susceptible to ≥16 µg/mL. Analysis of 14 KPC harboring K. pneumoniae isolates with MIC ≥16 µg/mL demonstrated that high level of resistance may be secondary to increased gene copy number (i.e., dose response) or the loss of a functional outer membrane porin, OmpK35 and/or OmpK36. The highest level of resistance was seen with isolates lacking both porins and with augmented KPC enzyme production (59).

Carbapenem resistance among gram-negative bacilli also can be mediated by the Class B β-lactamases, the metallo-β-lactamases (MBLs) (63). These enzymes use metal, commonly zinc, as a cofactor for β-lactam hydrolysis. With the exception of aztreonam, MBLs can hydrolyze all β-lactam antibiotics and are not inhibited by commercially available β-lactamase inhibitors. Co-expression of MBLs with ESBLs, AmpC β-lactamases, and/or other carbapenemases can result in concomitant resistance to aztreonam.

Intrinsic carbapenem resistance in Stenotrophomonas maltophilia is due to chromosomal MBLs. Initially these MBLs were described in Pseudomonas spp. but reports in Enterobacteriaceae are now quite common. Frequently detected MBLs include the IMP-type (active against imipenem) and VIM-type (Verona integron-encoded MBL), with the most prevalent being VIM-2. As of 2009, international attention has turned toward the increasing recovery of NDM-1, that is, New Delhi MBL. Due to the conveniences of travel and medical tourism, NDM-1 is not only endemic to the Indian subcontinent, this relatively novel MBL has disseminated worldwide (64,65,66,67,68).

The rapid spread of NDM-1 exemplifies the fluidity of gene transfer between bacterial species. Although blaNDM-1 was initially and repeatedly mapped to plasmids isolated from carbapenem-resistant E. coli and K. pneumoniae, reports of both plasmid and chromosomal expression of blaNDM-1 has been noted in other species of Enterobacteriaceae as well as Acinetobacter spp. and P. aeruginosa (56,66). It is currently held that blaNDM-1 is a chimeric gene that may have evolved from A. baumannii (69).

A small percentage of the Class D β-lactamases, commonly referred to as oxacillinases, demonstrate low-level carbapenemase activity. These carbapenemases contribute to the carbapenem resistance in Acinetobacter spp. and can be chromosomal or plasmid-mediated. Examples of acquired Class D carbapenemases include OXA-23, OXA 24/40, OXA-58, and
OXA-48. OXA-23 (originally ARI-1) was the first of these types of carbapenem-hydrolyzing oxacillinases described (70). Both OXA-23 and OXA-58 are responsible for the carbapenem resistance reported in many isolates of A. baumannii recovered from skin and soft tissue infections in military and civilian personnel returning from the military operations in the Middle East and Afghanistan (71). OXA-48 appears to have the highest affinity for carbapenems of this class of carbapenemases and has been described with increasing frequency in Turkey, the Middle East, and Europe (72). High-level carbapenem resistance is usually conferred by the concurrent presence of other resistance determinants including alterations in porins (e.g., CarO in A. baumannii) (73), modifications in PBPs, increased transcription mediated by insertion sequences, increased gene copy number, and amplified drug efflux (74).

Carbapenem resistance in P. aeruginosa is frequently due to a variety of mechanisms including expression of carbapenem-hydrolyzing β-lactamases. Other contributors to the carbapenem-resistant phenotype in this species include efflux and alterations in outer membrane proteins. Unlike carbapenemases that, in general, do not discriminate between carbapenems, upregulated expression of certain multidrug efflux pumps (e.g., MexAB-OprM) found in P. aeruginosa appear to exclude imipenem but confer resistance to meropenem (3,75). The loss or alteration of the outer membrane porin OprD, however, is specific for imipenem resistance exclusive of other β-lactams (76).


RESISTANCE TO ERYTHROMYCIN AND OTHER MACROLIDES

Macrolide resistance has been reported in numerous bacteria, but is of great clinical significance when treating community-associated gram-positive pathogens like S. pneumoniae and S. aureus. In pneumococci, two phenotypes of macrolide resistance appear to be common worldwide, although prevalence varies geographically (77). One is isolated low-level macrolide resistance and the other is high-level resistance to macrolides, lincosamindes (e.g., clindamycin), and streptogramin B (MLSB) (78).

Low-level macrolide resistance is mediated through expression of an efflux pump (encoded by mefA) that efficiently removes macrolides from the cytoplasm preventing interaction between drug and the bacterial ribosome (79,80). High-level macrolide resistance, the MLSB phenotype, is secondary to the expression of erm genes (erythromycin rRNA methylase), specifically ermB, which results in dimethylation of adenine 2058 in the 23S rRNA of the 50S ribosomal subunit, thus modifying the drug target for erythromycin, clindamycin, and the streptogramin quinupristin. Expression of the erm genes can be constitutive or induced.

The expression of incompletely homologous erm genes (ermA and ermC) appears to be responsible for a similar phenotype in S. aureus, irrespective of methicillin susceptibility (81,82). In vitro, S. aureus isolates with constitutive expression of the erm gene demonstrate resistance to both erythromycin and clindamycin. Isolates with inducible resistance, however, can test susceptibility to clindamycin despite testing resistance to erythromycin. The inducible nature of these genes and the associated poor clinical outcomes led to the development of the phenotypic D-test (83,84). Macrolide efflux pumps (e.g., msrA) are responsible for isolated macrolide resistance in Staphylococcus spp. (85).


AMINOGLYCOSIDE RESISTANCE

Aminoglycoside resistance is common in both gram-positive and gram-negative organisms. Resistance can be due to target modification, alterations in outer membrane proteins, resulting in decreased intracellular drug concentrations (86,87), efflux (88), and enzymatic drug modification (5). The latter is the most common and includes phosphorylation, acetylation, or adenylylation of the antimicrobial agent by plasmid- or transposon-encoded enzymes. This group of enzymes is exceedingly diverse and a detailed comprehensive review was recently published (5).

AAC(6′)-Ib is a commonly produced aminoglycoside N-acetyltransferase responsible for amikacin resistance among Acinetobacter spp., Vibrio spp., Pseudomonas spp., and Enterobacteriaceae. Like other aminoglycoside-modifying enzymes, the genes encoding these enzymes often are mapped to integrons, transposons, or plasmids. Variants of these enzymes have reduced susceptibility to other aminoglycosides like gentamicin and some transmissible elements carry additional genes associated with fluoroquinolone resistance (89) and β-lactamase production including carbapenemase production (90).

AAC(6′)-Ib-cr is a notable variant of AAC(6′)-Ib in that a two base pair change has permitted acetylation of ciprofloxacin and norfloxacin in addition to amikacin. This appears to be the first description of a single function drug-modifying enzyme that can inactivate unrelated antibiotics (91).

Another mechanism of high-level aminoglycoside resistance mechanism involves the expression of 16S rRNA methylases (e.g., rmt genes and armA). These enzymes methylate the ribosomal RNA (rRNA) associated with the 30S ribosomal subunit preventing aminoglycoside binding (92,93). The genes encoding these enzymes also have been mapped to transposons and plasmids harboring other resistance determinants (92,94).


TETRACYCLINE RESISTANCE

Tetracycline inhibits protein synthesis by preventing the association between aminoacyl-tRNA and the ribosome. Tetracycline resistance is widespread in both gram-positive and gram-negative bacteria and is most often mediated by drug efflux or ribosomal protection (95). Most acquired tetracycline resistance genes that mediate efflux reside on transposons, plasmids, or integrons allowing for efficient horizontal gene transfer (96).

Efflux effectively decreases intracellular drug concentrations, thus preventing inhibition of protein synthesis. Over 30 tet efflux genes have been characterized, all of which encode energy-dependent membrane-associated proteins (95,96). TetK and tetL are found primarily in gram-positive species including S. aureus and confer resistance to tetracycline but not minocycline (97). TetB appears to have the widest host range among gram-negative bacilli and confers resistance to tetracycline and minocycline but not the glycylcycline, tigecycline (95).

Ribosomal protection proteins confer resistance to tetracycline, doxycycline, and minocycline. TetM is widely dispersed among Staphylococcus spp. as well as Enterococcus faecalis, Neisseria. gonorrhoeae, Mycoplasma pneumoniae, and Bacteroides fragilis (95). Due to association with transmissible genetic elements, description of cotransfer with other resistance elements including the aforementioned erm genes has been described (96).

Jun 16, 2016 | Posted by in INFECTIOUS DISEASE | Comments Off on Molecular Biology of Resistance: A Brief History of Resistance Mechanisms and the Discovery of Gene Transfer

Full access? Get Clinical Tree

Get Clinical Tree app for offline access