Download Sir Richard Sykes - Oxford Academic

J Antimicrob Chemother 2010; 65: 1842 – 1852
doi:10.1093/jac/dkq217 Advance Access publication 23 June 2010
The 2009 Garrod Lecture: The evolution of antimicrobial
resistance: a Darwinian perspective
Sir Richard Sykes *
Department of Biosurgery and Surgical Technology, Imperial College, Hammersmith Campus, Du Cane Road, London W12 0NN, UK
*Tel: +44-20-8383-2273; E-mail: [email protected]
Microbes have evolved over 3.5 billion years and are arguably the most adaptable organisms on earth.
Restricted genetically by their inability to reproduce sexually, bacteria have acquired several additional mechanisms by which to exchange genetic material horizontally. Such mechanisms have allowed bacteria to inhabit
some of the most inhospitable environments on earth. It is thus hardly surprising that when faced with a
barrage of inimical chemicals (antibiotics) they have responded with an equal and opposite force. This
article compares and contrasts the evolution of antimicrobial resistance to b-lactam antibiotics over the last
70 years in two bacterial species, namely Staphylococcus aureus, a highly evolved human pathogen, and
Pseudomonas aeruginosa, an opportunistic nosocomial pathogen.
Keywords: microbes, b-lactam antibiotics, Staphylococcus aureus, Pseudomonas aeruginosa
Introduction
‘Infection 2009’ is a very significant event in this year’s calendar
and I feel very privileged having been invited to present the
Garrod Lecture at this prestigious meeting. This is also a very
special year as we are celebrating two famous anniversaries,
the 200th anniversary of Darwin’s birth and the 150th anniversary of On the Origin of Species, a publication that enunciated
the greatest general principles in biology. These principles
relate to the concept of evolution through a process of natural
selection: adapt or perish. Darwin had eventually come to this
view following his voyage on HMS Beagle, and almost 30 years
of painstaking research, leaving no stone unturned.
Darwin worked in a hostile environment, during a period when
the general belief was that the world had been created in 6 days,
and that this had been accomplished 6000 years in the past. It
was a period when nothing was known about genetics or the
microbial basis for infectious diseases. The concept of spontaneous generation was still regarded as the route through
which new life forms appeared. However, as with the concept
of evolution, things were changing rapidly. In 1866 Gregor
Mendel working in a monastery garden had published his work
on experiments with plant hybrids and the basic principles of
heredity. However, they remained largely undiscovered in the
monastery library at Brunn for many years and were never available to Darwin. At around the same time Pasteur was attacking
the concept of spontaneous generation with his experiments on
fermentation. This work, as we all know, led to the understanding of the role of microorganisms as agents of infectious disease.
The next 30 years or so was a period of intense activity in the
identification and characterization of the agents of disease in
man, animals and plants. It soon became clear that these microorganisms were remarkable creatures as they could be found in
all ecological niches examined. A perfect example of survival:
adapt or perish.
We can only imagine what impact this work on microorganisms would have had on Darwin had it been carried out in the
early nineteenth century as it would have provided a perfect
example for Darwin’s theory. Darwin was driven to his conclusions through fossil evidence and the changes that he
observed through both natural and artificial selection. He had
to befriend the pigeon fanciers and the dog breeders who were
using artificial selection techniques to create new breeds.
However, his work was dedicated to creatures that reproduce
sexually where the overriding factor for genetic variation is
during the process of meiosis (Figure 1a) and the mixing of
genes between partners. This is a slow process due to long gestation periods. However, bacteria multiply asexually by binary
fission (Figure 1b) with generation times of minutes rather
than days, months or years.
Microbes, which have evolved over 3.5 billion years, are arguably the most adaptable organisms on earth, inhabiting nearly
every corner of the globe. Some invade the cavities of a wide
variety of insects and invertebrates while others colonize the
skin, blood, eyes and internal organs of animals and man.
They thrive in the most inhospitable places such as hydrothermal vents on the ocean floor and at the Poles. As already
described, they divide by binary fission and thus produce
clones of each other. Any genetic variability under these circumstances requires random point mutations within the genome
that probably occur at a rate of 1×1025 – 1×1028. Although
this is a very low frequency, it is important to remember the
# The Author 2010. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]
1842
JAC
Garrod Lecture
(a)
Daughter
nuclei II
Daughter
nuclei
Interphase
Meiosis I
Meiosis II
Homologous
chromosomes
Prokaryotic
chromosome
Plasma
membrane
Cell wall
(b)
Duplication
of chromosome
Continued growth
of the cell
Division into
two cells
Figure 1. Comparison of cell division in eukaryotic cells (meiosis; a) and bacteria (binary fission; b).
very high concentrations of organisms in the gut, for example,
and so point mutations do play a role in genetic variability.
However, if this were the only mechanism open to bacteria to
adapt their genetic structure then it is inconceivable they
could have been so successful. Because they are unable to
reproduce sexually, bacterial species have acquired several
additional mechanisms by which to exchange genetic material
horizontally, giving them tremendous genetic flexibility. The
horizontal transfer of genetic material between the same and
different species can be accomplished through three main
mechanisms (Table 1). Each of these methods of genetic
exchange can introduce sequences of DNA that share little or
no homology with the DNA of the recipient cell. The inherited
DNA can remain outside the chromosome (extrachromosomal)
or it can be integrated.
All three mechanisms are capable of introducing antibiotic
resistance genes associated with a number of different genetic
entities, the three major ones being: (i) plasmids,1 which
are circular double-stranded DNA molecules separate from
chromosomal DNA and capable of replicating independently
of chromosomal DNA (host-to-host transfer requires direct
mechanical transfer by conjugation or uptake by transduction);
(ii) transposons (or jumping genes), which are mobile sequences
of DNA that can move to different positions within the genome
of a single cell (transposition)2 (in the process, such activity
can produce mutations; transposons in bacteria often carry
1843
Garrod Lecture
Table 1. Mechanisms of genetic transfer between bacteria
Mechanism
Transformation
Transduction
Conjugation
Mode of DNA uptake
uptake of naked DNA
transfer of DNA by bacteriophage
transfer of DNA by cell-to-cell contact
and plasmid transfer
genes for antibiotic resistance and can jump from chromosome
DNA to plasmid DNA and back, a good mechanism
for developing multiresistant strains and also for causing
mutations); and (iii) integrons,3 which are gene-capture
systems found in plasmids, chromosomes and transposons,
through which pieces of DNA called gene cassettes can be incorporated, expressed and disseminated. The integron contains an
integrase to allow its insertion into the genome, a promoter
that drives expression and a gene that allows integration of
cassettes.
Horizontal gene transfer is believed to be the essential mechanism for the attainment of genetic plasticity in many species of
bacteria. Thus during bacterial evolution the ability of bacteria to
adapt to new environments most often results from the acquisition of new genes through horizontal transfer rather than by
alteration of gene functions through numerous point mutations
and vertical transmission: a good example of early gene therapy.
In this article I want to review the development of antimicrobial resistance over the last 70 years and discuss how bacteria
have adapted to the wide range of inimical environments that
we have created. To illustrate how bacteria have adapted to
the onslaught of antibiotics I have chosen two organisms,
Staphylococcus aureus and Pseudomonas aeruginosa, the
former being a highly evolved human pathogen and the latter
an opportunistic nosocomial pathogen. I will restrict most
of the discussion to the interaction of these organisms with
b-lactam antibiotics. My intention is not to blind you with
science but to tell a story, and at times relate my own
involvement in the process.
S. aureus
Interaction with b-lactam antibiotics
S. aureus is a well-known major human pathogen commonly
found on the mucous membranes and skin of around one-third
of the population. At the time of the isolation of penicillin in the
late 1930s, S. aureus, along with Streptococcus pyogenes, was a
major cause of morbidity and mortality, hence penicillin, which
specifically targeted these organisms, was seen as a wonder
drug. However, the euphoria did not last for long as penicillinresistant strains of S. aureus soon started to appear. By 1955,
80% of all S. aureus strains in London hospitals were resistant
to penicillin.4
At this point we need to review the mode of action of penicillin and the mechanisms through which resistance to penicillin
can occur. The mechanism of antibiotic action in general can
be expressed as illustrated in Figure 2. Ideally, a given antibiotic
reaches its target site(s) unimpeded and intact, binds to and
1844
Antibiotic
Target(s)
Cell response
Amount of
antibiotic
reaching target
Nature of target
Nature of response
Figure 2. Mode of action of antibiotics and factors that influence their
activity.
inactivates the essential target(s) and, as a result, the cell
rapidly loses viability and dies. In the 1940s it was found that
penicillin had an effect on cell wall structure and morphology
and, by the 1950s, penicillin was shown to be a specific inhibitor
of cell wall synthesis.
In the 1960s the structure of bacterial cell walls and the
mechanisms of their synthesis were elucidated in detail. In
Gram-positive bacteria such as staphylococci, the cell wall is a
strongly linked peptidoglycan structure that is responsible for
the shape and integrity of the cell wall (Figure 3a). Early
studies recognized two major enzyme groups located on the
cytoplasmic membrane that were involved in peptidoglycan formation and that are susceptible to penicillin action, namely
transpeptidases and carboxypeptidases (Figure 4).5 It is now
recognized that there are multiple targets in the cell membrane,
referred to as penicillin-binding proteins (PBPs) that are all
involved in cell wall synthesis.6 Penicillin acts as a steric analogue
of the D-Ala-D-Ala peptide (which is a building block mimetic) and
thus binds to the cell-wall-synthesizing enzymes. As a result, the
enzymes are inactivated resulting in a weakened cell wall
structure causing the cell wall to rupture, resulting in cell
death. Penicillin has a high affinity for these enzymes and thus
is extremely effective in binding to this target site.
Resistance to b-lactam antibiotics
A major factor in bacterial resistance to antibiotics is the failure
of the antibiotic to reach its target (Figure 2). This may be due to
poor membrane permeability and/or antibiotic inactivation. For
b-lactam antibiotics to reach their target on the bacterial cytoplasmic membrane they must pass through the cell wall envelope. In staphylococci and other Gram-positive bacteria this
consists of one or more layers of peptidoglycan linked to an
anionic polymer (Figure 3a). Although the cell wall acts as a molecular sieve with an exclusion limit of 100000 Da, it is not an
effective barrier against penetration by b-lactam antibiotics
(300 Da). Thus permeability plays little or no role in resistance
to penicillin in staphylococci.
One of the major mechanisms of resistance to b-lactam antibiotics is enzymatic inactivation (Table 2). Even before penicillin
had been introduced, Abraham and Chain7 reported on the
identification of an enzyme that could inactivate the molecule,
which they termed penicillinase. This and related enzymes,
now referred to as b-lactamases, hydrolyse the cyclic amide
bond (the b-lactam ring) of susceptible b-lactams to produce
inactive products (Figure 5). These open ring molecules are no
longer recognized by the cell-wall-synthesizing enzymes.
In 1955 all resistance to penicillin among staphylococci was
due to enzymatic inactivation by b-lactamase, which is a
single protein coded for by a single gene: these genes were
JAC
Garrod Lecture
(a)
LTA
Transpeptidase
Carboxypeptidase
Peptidoglycan
Protein
Cytoplasmic
membrane
Phospholipids
Cytoplasm
(b)
O-antigen
repeat
n
n
n
n
n
n
Outer
core
Porin
Outer
membrane
LPS
Inner
core
Lipid A
Efflux pump
Lipoproteins
Peptidoglycan
Periplasm
MDO
(periplasmic
b-glucans)
Undecaprenyl
phosphate
sugars
Inner
membrane
Phospholipids
Cytoplasm
Protein
Figure 3. Structure of the cell walls of S. aureus (a) and P. aeruginosa (b). LTA, lipoteichoic acid; LPS, lipopolysaccharide.
shown to be plasmid mediated with cell-to-cell transfer occurring by transduction (that is by bacteriophage). Most staphylococci are lysogenic, which means they carry phages, and
staphylococcal R-plasmids are transmitted from cell to cell by
these lysogenic phages. The b-lactamase produced by staphylococci is an inducible extracellular enzyme, thus production occurs
only when necessary in the presence of b-lactam antibiotics.8
The number of plasmid copies can be quite high and, along
with the population effect, enzyme levels can represent up to
1% of the dry weight of the cell. Such large quantities of
enzyme pass into the external milieu to inactivate penicillin
before it has an opportunity to reach the target site, producing
high levels of resistance. It is important to recognize here that
there are two opposing enzymatic mechanisms at work that
both have an affinity for the antibiotic: the transpeptidase or
cell-wall-synthesizing enzymes and the b-lactamase. Therefore,
in order to be effective, the b-lactamase must have an affinity
for the b-lactam antibiotic at or around the MIC for the
1845
Garrod Lecture
GlcNAc
GlcNAc
MurNAc
GlcNAc
MurNAc
MurNAc
L-Ala
L-Ala
L-Ala
D-Glu
D-Glu
D-Glu
meso-DAP
meso-DAP-?-NH2
meso-DAP
D-Ala
D-Ala
D-Ala
D-Ala
LD-Carboxypeptidase
D-Ala
DD-Carboxypeptidase
D-Ala
meso-DAP
meso-DAP
D-Glu
D-Glu
L-Ala
MurNAc
Transpeptidase
L-Ala
GlcNAc
MurNAc
GlcNAc
Figure 4. Penicillin-sensitive enzymes involved in cell wall biosynthesis. GlcNAc, N-acetylglucosamine; MurNAc, N-acetylmuramic acid; DAP,
diaminopimelic acid.
Table 2. Mechanisms of antibiotic resistance in S. aureus and P. aeruginosa
Organism
Resistance mechanism
Antibiotics affected
S. aureus
enzymatic inactivation/modification
alteration of target
antibiotic trapping
efflux pumps
b-lactams, aminoglycosides
methicillin, macrolides, vancomycin
vancomycin, daptomycin
fluoroquinolones, tetracyclines
P. aeruginosa
enzymatic inactivation/modification
alteration of target
efflux pumps
porin loss
b-lactams, aminoglycosides
aminoglycosides, fluoroquinolones
most antibiotics
carbapenems
organism. In the case of staphylococci the MIC of penicillin
is 1027 M, similar to the affinity of the b-lactam for the
b-lactamase. Thus the enzyme is an extremely potent
weapon against penicillin particularly when produced in high
concentrations.
Obviously the plasmids carrying genes for b-lactamase production did not just appear. Plasmids are common in bacteria
and there is little doubt that horizontal gene transfer through
bacteriophages was a common event. Only on the introduction
of penicillin was such a b-lactamase-producing gene useful to
the bacterium.
already developed mechanisms and adapting them to cope
with a new environment. It is interesting to compare this adaptive process with that seen in methicillin-resistant organisms,
which we shall come to shortly. Soil samples from the British
Museum collected in the seventeenth century were shown to
contain strains of Bacillus spp. producing b-lactamase. Once
penicillin appeared on the scene those organisms carrying the
b-lactamase gene would have a significant advantage and
could transfer that advantage quickly to other organisms by
the horizontal route thus quickly producing resistant strains. All
of this happened in a short period of time and is a great
example of adaptive survival.
Evolution of b-lactamase
Where did this enzyme come from in the first place? It now turns
out that many soil microbes (including bacteria, streptomycetes
and fungi) are capable of producing b-lactam antibiotics, and
they in turn have developed defence mechanisms to protect
themselves against the antibiotics produced. The staphylococcal
b-lactamase enzyme has been shown to have close homology
with that from the soil organism Bacillus licheniformis.9 In turn,
this enzyme is closely related to the transpeptidase involved in
cell wall biosynthesis,10 a clear process of evolution using
1846
Semi-synthetic penicillins
This situation put enormous pressure on the pharmaceutical
industry to develop more agents that could resist the action
of b-lactamases. The isolation of the penicillin nucleus
[6-amino-penicillanic acid (6-APA)] in 1959 led to the development of b-lactamase-stable semi-synthetic penicillins such as
cloxacillin and methicillin (Figure 6).11 These compounds
showed a high degree of resistance to hydrolysis by staphylococcal b-lactamase. However, following the introduction of
JAC
Garrod Lecture
R
Ser
NH
S
CH3
N
OH
R
Ser
S
CH3
CH3
N
O
O
O
O
O
HO
R
NH
CH3
HO
R
NH
S
HO
O
NH
CH3
H2O
Ser
NH
S
O
CH3
O
CH3
NH
O
HO
CH3
O
HO
acyl intermediate
Figure 5. Inactivation of penicillin by b-lactamase.
methicillin into clinical practice in 1961, methicillin-resistant
strains [methicillin-resistant S. aureus (MRSA)] soon started to
appear. Over the past 50 years these organisms have acquired
multidrug-resistant traits and have spread worldwide to
become one of the most important causative agents of
hospital-acquired infections. Then in the late 1990s these organisms began to show up in the community. This is a great
example of a dangerous human pathogen continually evolving
to deal with a changing environment. These globally spread epidemic clones of MRSA are the product of the convergence of
several unusual evolutionary processes.
Methicillin resistance
The development of resistance to methicillin is completely different from that seen with the early penicillins. First, resistance to
penicillins was rapid whereas resistance to methicillin was slow
to develop. Staphylococcal b-lactamase has a poor affinity for
methicillin, which therefore reaches the target site intact and
inactivates the organism. Thus methicillin resistance is not due
to permeability or enzymatic inactivation but is primarily due
to production of a modified target PBP to which the antibiotic
cannot bind. However, as was the case with the early penicillins,
resistance emerged following the acquisition of foreign DNA, in
this case through the mecA gene (Figure 7).12 This gene is the
determinant of a unique PBP (PBP2A) that has a low affinity
for b-lactam antibiotics but can still function as a surrogate
for the native staphylococcal PBPs, the cell-wall-synthesizing
enzymes.
The mecA resistance gene appears to have evolved from a
domestic gene of the fully b-lactam-susceptible Staphylococcus
sciuri, a frequent colonizer of the skin of both wild and domestic
animals and one of the most abundant staphylococcal species
on this planet.13 However, in order for the mecA gene to establish
itself in S. aureus, it had to be incorporated into a unique molecular vector called the staphylococcal chromosome cassette (SCC),
which has the ability to deliver a variety of different resistance or
virulence mechanisms to S. aureus. The successful grafting of the
SCCmec gene into the methicillin-susceptible S. aureus (MSSA)
cell is not a straightforward process and requires additional
factors, particularly a unique and apparently rare genetic background that allows maintenance and expression of the mecA
gene. Hence, the relatively few successful SCCmec acquisition
events that give rise to an MRSA strain appear to require the
absence of a genetic barrier that is very common in many
S. aureus lineages.14 This explains why we see only a handful
of lineages around the world. As with the gene for b-lactamase
production, the SCC cassette is transferred by transduction.
The origin of the mecA gene is a fascinating event and could
not possibly have been a result of the appearance of methicillin.
Penicillin was used extensively in animals from 1949 and S. sciuri
strains have remained susceptible to penicillin as they carry no
plasmids carrying the b-lactamase gene. The gene homologue
of mecA is a PBP transpeptidase and so the gene codes for a
cell-wall-synthesizing enzyme in S. sciuri.15 However, if this
gene is up-regulated then the organism shows resistance to
penicillin. Therefore it is reasonable to assume that the extensive
use of antibiotics in animals led to the up-regulation of the gene,
which was subsequently transferred to MSSA strains thus converting them to methicillin resistance. Once again we see that
a cell-wall-synthesizing enzyme has been hijacked for resistance
purposes.
In the late 1990s MRSA-infected patients from the community who lacked any of the usual risk factors for MRSA infections
began to make their appearance. These organisms were rapidly
transmitted from person to person causing skin structure
1847
Garrod Lecture
H
H
N
OMe
H
S
H
H
N
Me
H
S
Me
N
Me
OMe
O
N
O
Methicillin
Me
O
CO2H
CO2H
Cl
6-Aminopenicillanic acid
N
O
NH
H3C
S
CH3
N
O
CH3
Cloxacillin
O
HO
H2N
H
N
HOOC
S
S
N
O
NH2
O
N
CH3
O
CH3
O
O
O
COOH
O
Cephalosporin C
H
O
COOH
7-Aminocephalosporanic acid
COOH
O
H
N
NH
OH
CH2OH
O
H
HN
H
H
Clavulanic acid
S
N
O
+
OH
NH3
H
H
OH
O
S
Imipenem
N
O
O
S
–
NH2
O
N
Thiapenem
N
HN
O
CH3
O
HN
O
CH3
OH
N
O
Monobactam
Figure 6. Structures of a range of b-lactams.
1848
N
SO3H
O
Aztreonam
SO3H
JAC
Garrod Lecture
MSSA
orfX
b-Lactamase inhibitors
Excision
Integration
mecA
SCCmec
MRSA
mecA
SCCmec
MSSA
orfX
mecA
SCCmec
Figure 7. Excision and integration of SCCmec containing the mecA gene.
16
infections, osteomyelitis and pneumonia.
They were also
responsible for fatal diseases such as severe sepsis and necrotizing fasciitis. In contrast to the healthcare-associated MRSA,
these community-associated MRSA showed diverse genetic
backgrounds indicating the loss of a genetic barrier to the acquisition of the mecA gene. Subsequently, a few highly epidemic
clones have emerged and spread in the community. Many of
these strains are associated with the Panton –Valentine leucocidin toxin gene (pvl), which seems to correlate with the high frequency of skin and soft tissue infections caused by these
organisms.17 In the mid-1990s ,0.1% of individuals in the community carried MRSA strains whereas .50% of S. aureus hospital
isolates were resistant to methicillin in the same period.18 Now
we see high carriage rates of 6%– 8% in underserved groups in
the community.19
Once b-lactamase had been overcome with the development
of b-lactamase-resistant penicillins, if the organisms were to
survive they had to find other defence mechanisms. Modification
of the target site is an obvious mechanism of defence but is
much more complicated than antibiotic inactivation. This
extended genetic reservoir accessible to staphylococci (mobile
genetic elements plus DNA transfer mechanisms afforded by
horizontal gene flux) is fundamental to the acquisition, maintenance and dissemination of resistance in these organisms.
Resistance to cephalosporins
In 1954 the isolation of cephalosporin C from a sewage outlet in
Sardinia by Brotzu, and the subsequent identification of the
cephalosporins as b-lactam antibiotics by Abraham and
Newton in Oxford, caused a great deal of excitement, as they
were shown to be resistant to staphylococcal b-lactamase. The
cephalosporins were another example of naturally occurring
b-lactam antibiotics produced by a soil microorganism. The isolation of 7-aminocephalosporanic acid (7ACA) (Figure 6) in the
early 1960s allowed the development of semi-synthetic cephalosporins, the first of which were cefalotin and cefaloridine.
Although these early semi-synthetic cephalosporins were active
against staphylococci, they were considered to be of more
value against Gram-negative organisms, which were beginning
to become a serious problem in the hospital environment. Staphylococcal b-lactamase inactivated cefaloridine but not cefalotin so cefalotin was used widely at the time.
For many years there had been an ongoing search in the
industry for naturally occurring inhibitors of b-lactamase in
the hope that they could be used in combination with the
early penicillins. In 1976 the b-lactamase inhibitor clavulanic
acid (Figure 6) was isolated from Streptomyces clavuligerus and
this was shown to be a potent inhibitor of staphylococcal
b-lactamases.20 This compound was combined initially with amoxicillin to produce co-amoxiclav (Augmentin), which is active against
non-methicillin-resistant staphylococci. To date, no staphylococcal
enzymes have been seen that show resistance to clavulanic acid.
Resistance of staphylococci to other antibiotics
Over the last few years, the widespread development of resistance
to methicillin has led to the wider use of glycopeptide antibiotics
such as vancomycin. In 2000, MRSA strains carrying the transposon Tn1546-based enterococcal vancomycin resistance mechanism were identified.21 This resistance appears to be due to
modification of the target site, with the D-alanine-D-alanine
moiety of the peptidoglycan precursor (to which glycopeptides
bind) being replaced with D-alanine-D-lactate. These organisms
are also resistant to clindamycin, aminoglycosides, trimethoprim/sulfamethoxazole, rifampicin and the fluoroquinolones.
New compounds introduced over the last few years such as linezolid and daptomycin have also selected out resistant strains.22
It is now clear that in some instances staphylococcal resistance is due to a multiplicity of factors including altered PBPs,
large amounts of b-lactamase production, antibiotic trapping
and efflux pumps (Table 2). These can also be associated with
virulence factors that encourage the maintenance of plasmids
carrying antibiotic resistance genes.
P. aeruginosa
P. aeruginosa is a non-fermenting Gram-negative organism commonly found in the natural environment. However, it is also
a nosocomial pathogen, an opportunistic organism that has a
propensity for infecting critically ill and immunocompromised
patients. It can also colonize medical equipment such as
catheters and prostheses through its ability to form biofilms.
It represents a phenomenon of bacterial resistance since
practically all known mechanisms of antimicrobial resistance
can be seen in it.
P. aeruginosa has a large genome with 6.26 Mb equivalent to
5567 genes, whereas S. aureus has 2.81 Mb equivalent to 2594
genes. Such a large genome provides P. aeruginosa with much
greater adaptability allowing it to survive in most hostile environments and even dissolve oil slicks.
Interaction with b-lactam antibiotics
Like most of the Enterobacteriaceae, P. aeruginosa showed little
susceptibility to the early penicillins due to the inability of the
drug to reach its target site, the PBPs on the cytoplasmic membrane. The cell wall of Pseudomonas is very different from that of
S. aureus (Figure 3b). On the outside of the cell there is often a
layer of alginate polysaccharide that is important in biofilm formation. The outer membrane is the main barrier to entry of
1849
Garrod Lecture
b-lactam antibiotics, which can only cross this barrier through
aqueous channels called porins. P. aeruginosa possesses
several different porins including OprF, which is present in all
strains, and OprD, which is a specialized porin involved in the
uptake of amino acids and which has become adapted for the
specific uptake of the carbapenems.23 These porins are also
closely associated with efflux systems that are responsible for
rejecting molecules from the bacterial cell. In evolutionary
terms these systems have developed to allow the organism to
survive in inimical environments and have now adapted themselves to deal with antibiotics. There are four genetically different
efflux systems in P. aeruginosa, all of which are composed of
three components, namely an energy-dependent pump, an
outer membrane porin and a linker protein that couples these
two membrane proteins together.24
All classes of antibiotics are susceptible to extrusion by these
systems. Hence, there are significant permeability barriers to the
entry of b-lactam antibiotics into P. aeruginosa, unlike the situation in staphylococci. Molecules such as b-lactams have great
difficulty penetrating the outer membrane and even if they do
there is a high probability that they will be extruded. Even
compared with Escherichia coli, P. aeruginosa is 100 times
more impermeable to b-lactams. However, in common with
S. aureus, these organisms produce b-lactamase and over the
last 45 years there has been a constant battle between the production of new and more potent antibiotics and the microorganisms’ ability to destroy them. This is an incredible example of the
eternal battle that Huxley described as ‘Nature, red in tooth and
claw’ (Huxley was himself quoting from Tennyson).
b-Lactamases
All wild-type strains of P. aeruginosa produce an inducible chromosomally encoded b-lactamase designated AmpC, which is
produced constitutively at low levels and is capable of hydrolysing cephalosporins.25 However, in the presence of b-lactam antibiotics the level of enzyme can increase up to 1000-fold. This
enzyme, along with the permeability barrier, provides high-level
resistance to the early penicillins and cephalosporins. Unlike
the situation in staphylococci, the enzyme in P. aeruginosa is
located strategically in the periplasmic space.
The introduction of carbenicillin in 1965 provided the first
penicillin with activity against P. aeruginosa. Carbenicillin is able
to enter the cell through the porin system, shows poor affinity
for the AmpC cephalosporinase and is thus able to inactivate
the cell-wall-synthesizing enzymes. Carbenicillin was used
widely for the treatment of P. aeruginosa infections particularly
in burns patients. In 1969 a strain of P. aeruginosa was isolated
in the Burns Unit at the Birmingham Accident Hospital with MICs
of carbenicillin of .250 mg/L—very different from the low-level
resistance that had been seen previously. At the time I was
working in Bristol with Mark Richmond doing my PhD thesis on
the b-lactamases of P. aeruginosa. We asked Dr Edward
Lowbury for a culture of his organism. On examination the
strain produced two b-lactamases, one the normal AmpC cephalosporinase but the other was an enzyme shown to be indistinguishable from the TEM b-lactamase that at the time was
commonly found in strains of E. coli and known as RP-1.26
This was the first report of such an enzyme in P. aeruginosa. It
was shown subsequently that the gene was carried on a
1850
transposon with the capability of moving between the chromosome and a non-chromosomal location. Within a few months
we were contacted by Bill Newsome at Papworth Hospital, Cambridge who also had a strain of P. aeruginosa highly resistant to
carbenicillin. We showed that it was a different enzyme from
RP-1 even though it was transposon mediated.27 It is now
known as PSE-1 (where PSE stands for Pseudomonas-specific
enzyme). These were both rare occurrences at that time. There
are now four carbenicillin-hydrolysing enzymes that are Pseudomonas specific and all closely related, with the exception of
PSE-4, which appears to have been acquired from another
species.28
The next development came with the introduction of thirdgeneration cephalosporins with activity against Pseudomonas
including cefoperazone and ceftazidime. Ceftazidime came out
of a programme at Glaxo based on the oxyimino cephalosporins
with which I was closely associated in the 1970s. These compounds initially showed resistance to both the chromosomal
enzyme and the transposon-mediated enzymes and were used
extensively in Pseudomonas infections. Thienamycin, a naturally
occurring carbapenem antibiotic isolated from Streptomyces cattleya, led to the synthesis of imipenem29 (Figure 6), a novel
b-lactam introduced in 1985, again showing high levels of
activity against P. aeruginosa. Subsequently, while working at
the Squibb Institute for Medical Research in the late 1970s, we
isolated a novel group of molecules called monobactams
(Figure 6), which led to the development of aztreonam,
another compound active against Pseudomonas.30
This continuous attack on P. aeruginosa with new and powerful drugs has resulted in a powerful response, most of which has
been concentrated on drug inactivation. One feat of adaptability
occurred in cystic fibrosis patients who at the time were
being bombarded with ceftazidime because of its resistance
to the AmpC and TEM-type b-lactamases. This resulted in a
spontaneous mutation in the regulatory gene AmpR leading to
the derepression of enzyme production. Now P. aeruginosa
began behaving more like staphylococci, producing large quantities of enzyme bleeding out into the environment and
producing a population effect and high resistance. The gene
for this enzyme is still restricted to a chromosomal location in
P. aeruginosa but it will only be a matter of time before it gets
incorporated into a mobile genetic element similar to the
situation seen in the Enterobacteriaceae. Organisms carrying
this derepressed gene are highly resistant to third-generation
cephalosporins.
Since 1990, the constant exposure of P. aeruginosa to antibiotics in the hospital environment has resulted in a plethora
of b-lactamase types that exhibit differing substrate profiles.
The number of discrete enzymes in 1972 was 13, in 1999 it
was 282 and by 2004, 532. These enzymes, capable of inactivating b-lactam antibiotics, can be divided into four types based
upon nucleotide and amino acid sequences (Table 3). The
class A enzymes include the carbenicillinases (Table 3), which
are transposon mediated and are represented by the PSE
enzymes.31 Also in class A are the enzymes referred to as
extended-spectrum b-lactamases (ESBLs), which made their
appearance from 1990. They include the new TEM-type
enzymes, SHV enzymes, PER and VEB (South-East Asia) and
GES (Guinea). These enzymes are all closely related but with differing substrate profiles that have obviously evolved due to
JAC
Garrod Lecture
Table 3. b-Lactamases found in P. aeruginosa
Class
Type
A
carbenicillinases
B
C
D
metallo-b-lactamases
cephalosporinases
oxacillinases
Examples
Comments
PSE
ESBLs including TEM, SHV, PER, VEB and GES
VIM, IMP, GIM and SIM
AmpC
five groups
increased selection pressure. Most of the variations take place by
amino acid substitutions at the active site. Dissemination of
these class A ESBL-encoding genes by plasmids and integrons
provides a very important role in the resistance and dissemination of antibiotic resistance.
Class B enzymes are the metallo-b-lactamases (Table 3),
which are dependent upon zinc ions for activity and are referred
to as carbapenemases,25 because of their resistance to the carbapenems such as imipenem but sensitivity to the monobactam
aztreonam. These enzymes contain the VIM, IMP, SPM, GIM and
SIM enzymes all named on the basis of their initial geographical
location. Many of these enzymes are closely related but again
differ by changes at the active site. Many of the genes for
these enzymes are carried on mobile elements called gene cassettes that are inserted into class 1 integrons. Such integronlocated resistance genes provide increased potential for
expression and dissemination. Several of these class 1 integrons
have been found in transposons, allowing integrons to be transposed, which increases the threat of genes being disseminated
among diverse groups of bacteria. Resistance to imipenem is
also characterized by a deficiency of the OprD porin. Loss of
this porin in the presence of derepressed AmpC provides a high
level of resistance.
The class C enzymes (Table 3) are the cephalosporinases of
the AmpC type, which are naturally occurring inducible
enzymes in P. aeruginosa.25 The Class D enzymes (Table 3) are
the oxacillinases of which there are five different groups in
P. aeruginosa.26 Initially these enzymes were active against the
early penicillins and cephalosporins but now, like many of the
other enzyme types, they have broadened their profile to
include third-generation cephalosporins such as ceftazidime.
These enzymes are all closely related as a result of amino acid
substitutions. They can be chromosome-, plasmid- or integronlocated genes and so spread easily. These enzymes are not
susceptible to b-lactamase inhibitors.
Alteration of the target site is a rare occurrence in resistance
to antibiotics in P. aeruginosa as compared with staphylococci.
There are examples of low-affinity PBPs for certain antibiotics
but they are not a major cause of resistance. From an evolutionary point of view, the ability of Pseudomonas to produce a
plethora of b-lactam-hydrolysing enzymes located behind a
lipid barrier has been so successful in overcoming the onslaught
of b-lactam antibiotics that the necessity and complexity of an
altered target site has not proved necessary. This is in complete
contrast to the situation seen with the staphylococci. The mechanisms of resistance seen in P. aeruginosa are illustrated in
Table 2.
PSE-1 to PSE-4, transposon mediated
ESBLs, disseminated by plasmids and integrons
carbapenemases
may be encoded by chromosome, plasmids or integrons
Summary
Following the publication of On the Origin of Species in 1859 and
the subsequent debate at Oxford in 1860, the concept of evolution through natural selection and consequently the ability to
adapt slowly started to be accepted. However, today there are
still significant numbers of people around the world who continue to question the theory of evolution. Because of their
ability to reproduce rapidly and their enormous genetic flexibility,
bacteria have provided a perfect example of Darwin’s principles
when faced with the onslaught of a plethora of antibiotics.
The increasing availability of antibiotics in the 1950s and
1960s led many in authority to predict the ‘beginning of the
end’ for infection. Nothing could be further from the truth! Our
quests to counter the inevitable emergence of microbial resistance are puny compared with the power of Nature to defeat
our feeble attempts! Resistance is increasing remorselessly and
the output of novel antimicrobial compounds is decreasing.
However hard we try and however clever we are, there is no
question that organisms that have been around for 3 billion
years, and have adapted to survive under the most extreme
conditions, will always overcome whatever we decide to throw
at them.
Transparency declarations
None to declare.
References
1 Lipps G, ed. Plasmids: Current Research and Future Trends. Caister:
Academic Press, 2008 (ISBN 978-1-904455-35-6).
2 McClintock B. The origin and behaviour of mutable loci in maize. Proc
Natl Acad Sci USA 1990; 36: 344– 55.
3 Hall RM, Collins CM. Mobile gene cassettes and integrons: capture and
spread of genes by site-specific recombination. Mol Microbiol 1995; 15:
593–600.
4 Ridley M, Barrie D, Lynn R et al. Antibiotic-resistant Staphylococcus
aureus and hospital antibiotic policies. Lancet 1970; i: 230–3.
5 Tipper DJ, Strominger TL. Mechanism of action of penicillins: a proposal
based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl
Acad Sci USA 1965; 54: 1133– 41.
6 Blumberg PM, Strominger JL. Five penicillin-binding components occur
in Bacillus subtilis membranes. J Biol Chem 1972; 247: 8107 –13.
7 Abraham EP, Chain E. An enzyme from bacteria able to destroy
penicillin. Nature 1940; 146: 837.
1851
Garrod Lecture
8 Richmond MH. Purification and properties of the exo-penicillinase from
Staphylococcus aureus. Biochem J 1963; 88: 452–9.
relation to community-associated disease activity. J Infect Dis 2005;
192: 811–8.
9 Ambler RP. The amino acid sequence of Staphylococcus aureus
penicillinase. Biochem J 1975; 151: 197–218.
20 Howarth TT, Brown AG, King TJ. Clavulanic acid, a novel b-lactam
isolated from Streptomyces clavuligerus. J Chem Soc Commun 1976;
266–7.
10 Herzberg O, Moult J. Bacterial resistance to b-lactam antibiotics:
crystal structure of b-lactamase from Staphylococcus aureus PC1 at
2.5 Å resolution. Science 1987; 236: 694– 701.
11 Batchelor FR, Doyle FP, Nayler JHC et al. Synthesis of penicillin:
6-aminopenicillanic acid in penicillin fermentations. Nature (London)
1959; 183: 257–8.
21 Chang S, Sievert DM, Hageman JC et al. Infection with
vancomycin-resistant Staphylococcus aureus containing the vanA
resistance gene. N Engl J Med 2003; 348: 1342 –7.
22 Tsakris A, Pillai SK, Gold HS et al. Persistence of rRNA operon mutated
copies and rapid re-emergence of linezolid resistance in Staphylococcus
aureus. J Antimicrob Chemother 2007; 60: 649– 51.
12 Fuda CC, Fisher JF, Mobashery S. b-Lactam resistance in
Staphylococcus aureus: the adaptive resistance of a plastic genome.
Cell Mol Life Sci 2005; 62: 2617– 33.
23 Livermore DM. Of Pseudomonas, porins, pumps and carbapenems.
J Antimicrob Chemother 2001; 47: 247–50.
13 Couto I, de Lencastre H, Severina E et al. Ubiquitous presence of a
mecA homologue in natural isolates of Staphylococcus sciuri. Microb
Drug Resist 1996; 2: 377–91.
24 Livermore DM. Multiple mechanisms of antimicrobial resistance in
Pseudomonas aeruginosa: our worst nightmare? Clin Infect Dis 2002;
34: 634–40.
14 Katayama Y, Zhang HZ, Hong D et al. Jumping the barrier to b-lactam
resistance in Staphylococcus aureus. J Bacteriol 2003; 185: 5464–72.
25 Nordmann P, Guibert M. Extended-spectrum b-lactamases in
Pseudomonas aeruginosa. J Antimicrob Chemother 1998; 42: 128–31.
15 Wu S, Piscitelli C, de Lencastre H et al. Tracking the evolutionary
origin of the methicillin resistance gene: cloning and sequencing
of a homologue of mecA from a methicillin susceptible strain of
Staphylococcus sciuri. Microb Drug Resist 1996; 2: 435–41.
26 Sykes RB, Richmond MH. Intergeneric transfer of a b-lactamase gene
between Ps. aeruginosa and E. coli. Nature 1970; 226: 952–4.
16 Daum RS. Clinical practice. Skin and soft-tissue infections caused by
methicillin-resistant Staphylococcus aureus. N Engl J Med 2007; 357:
380–90.
28 Sanschagrin F, Bejaoui N, Levesque RC. Structure of CARB-4 and AER-1
carbenicillin-hydrolyzing b-lactamases. Antimicrob Agents Chemother
1998; 42: 1996– 72.
17 Tristan A, Bes M, Meugnier H et al. Global distribution of
Panton-Valentine leukocidin-positive methicillin-resistant Staphylococcus
aureus, 2006. Emerg Infect Dis 2007; 13: 594– 600.
29 Kahan JS, Kahan FM, Goegelman R et al. Thienamycin, a new
b-lactam antibiotic. I. Discovery, taxonomy, isolation and physical
properties. J Antibiot (Tokyo) 1979; 32: 1– 12.
18 Sa-Leao R, Sanches IS, Couto I et al. Low prevalence of
methicillin-resistant strains among Staphylococcus aureus colonizing
young and healthy members of the community in Portugal. Microb
Drug Resist 2001; 7: 237–45.
30 Sykes RB, Cimarusti CM, Bonner DP et al. Monocyclic b-lactam
antibiotics produced by bacteria. Nature 1981; 291: 489–91.
19 Pan ES, Diep BA, Charlebois ED et al. Population dynamics of nasal
strains of methicillin-resistant Staphylococcus aureus—and their
1852
27 Sykes RB, Richmond MH. R-factors, b-lactamase and carbenicillinresistant Pseudomonas aeruginosa. Lancet 1971; ii: 342– 4.
31 Bert F, Branger C, Lambert-Zechovsky N. Identification of PSE and OXA
b-lactamase genes in Pseudomonas aeruginosa using PCR-restriction
fragment length polymorphism. J Antimicrob Chemother 2002; 50:
11– 8.