Evaluation of Phenotypic and Genotypic
Extended-Spectrum Beta-Lactamase
Detection methods
Louise Svärd
BSc Biomedical Science
May 2007
ACKNOWLEDGEMENTS
I would like to take this opportunity to thank all the people that contributed to the realization of this project and for making my stay in Dublin an unforgettable experience.
I would especially like to thank my supervisor Dr. Celine Herra for guiding me through this project and for being so helpful, supporting and encouraging during these three months.
Many thanks to Dr. Patrick McHale for being so friendly and making me feel welcome. Thanks are also due to Ms Brid Ann Ryan for extending a friendly welcome to Kevin Street College, DIT.
I would also like to thank Ted Doody and Patricia Taylor for being very friendly and for their good advice. And also big thank you for your kind assistance.
ABSTRACT
The emergence and spread of resistance in Enterobacteriaceae is a growing concern in human medicine today. Enterobacteriaceae producing extended-spectrum β-lactamases (ESBLs) have become efficient at inactivating β-lactam antibiotics especially the newer third generation cephalosporins. In addition, ESBL producing Enterobacteriaceae are frequently resistant to other groups of commonly used non-β-lactam antibiotics such as the fluoroquinolones. Reliable, rapid and low cost methods to detect ESBLs in clinical microbiology laboratories are therefore required.
The aim of this project was to evaluate the phenotypic and genotypic detection methods for ESBLs and to examine the optimum antimicrobial agent(s) for ESBL detection. A comparison with the CLSI susceptibility test and the ESBL screen test was performed using a number of clinical isolates of E. coli and Klebsiella spp. suspected to contain ESBLs. Two confirmatory tests, the double disc synergy test and the combination disc test for ESBLs were also compared. Single and multiplex PCR assays were established using primers for the TEM-, SHV- and OXA-type β-lactamases. The results of this study show that ESBL screening is required in routine laboratories for successful detection of ESBLs. The best indicator cephalosporin for detection of ESBLs in E. coli was cefpodoxime whilst the best indicators for detection of ESBLs in K. pneumoniae were cefpodoxime and ceftazidime. The combination disc confirmatory test demonstrated the highest rate of detection. The multiplex PCR assay was found to be a rapid and cost-effective method for ESBL detection. However, nucleotide sequencing is required to confirm ESBL production amongst these organisms.
CONTENTS
SCHOOL OF BIOLOGICAL SCIENCES, DUBLIN INSTITUTE OF TECHNOLOGY... 1
ACKNOWLEDGEMENTS ... 2
ABSTRACT ... 3
1. INTRODUCTION ... 6
1.1 β-lactam antibiotics ... 6
1.2 Antimicrobial resistance ... 8
1.3 Resistance to β-lactam antibiotics ... 10
1.4 β-lactamases ... 10
1.5 Classification of β-lactamases... 11
1.6 β-lactamase mediated β-lactam resistance in Gram-negative bacilli ... 13
1.7 Extended-spectrum β-lactamses, ESBLs ... 15
1.8 Epidemiology and clinical significance of ESBLs ... 17
1.9 Detection methods for ESBLs... 19
1.9.1 Phenotypic detection methods for ESBLs ... 20
1.9.2 ESBL disc diffusion screen ... 20
1.9.3 Confirmatory tests for ESBLs in Klebsiella spp and E. coli ... 21
1.9.3.1 Double disc synergy test ... 21
1.9.3.2 Cephalosporin/clavulanic acid combination disc tests ... 21
1.9.3.3 Broth microdilution... 22
1.9.4 Commercial Methods for ESBL detection... 22
1.9.4.1 E-test ESBL strips ... 22
1.9.4.2 VITEK ESBL test... 23
1.9.5 Molecular detections methods for ESBLs ... 23
1.10 The aim of this project ... 24
2. MATERIALS AND METHODS ... 25
2.1 Material ... 25
2.1.1 Culture media ... 25
2.1.4 Reagents... 25
2.1.3 Antimicrobial discs ... 26
2.1.5 PCR-reagents... 27
2.1.6 PCR-equipment... 27
2.1.7 Equipment for agarose gel electrophoresis ... 27
2.2 Methods ... 28
2.2.1 Bacterial isolates ... 28
2.2.2 API 20E identification ... 29
2.2.3 Antimicrobial susceptibility testing... 30
2.2.4 Screen for ESBL... 30
2.2.4.1 CLSI ESBL screen method ... 30
2.2.5 ESBL confirmatory tests ... 31
2.2.5.1 Double disc synergy test ... 31
2.2.5.2 The cephalosporin/clavulanic acid combination disc test... 32
2.2.6 DNA amplification of TEM-, SHV- and OXA-type β-lactamases... 33
2.2.6.1 Preparation of DNA template for multiplex PCR using the boiling method ... 37
2.2.6.2 Optimizing of SHV-primer in singleplex PCR... 37
2.2.6.3 Optimizing of TEM-primer in singleplex PCR... 38
2.2.6.4 Optimizing of OXA-primer in singleplex PCR ... 38
2.2.6.5 Optimizing of SHV, TEM and OXA-primer in multiplex PCR ... 39
2.2.7 Detection of amplified DNA-products using agarose gel electrophoresis... 39
3. RESULTS... 40
3.1 Antimicrobial susceptibility test... 40
3.2 CLSI ESBL screen method... 42
3.3 ESBL confirmatory tests (double disc synergy test and the combination disc test) ... 48
3.4 Combined result for disc susceptibility test, the ESBL screen and the combination disc confirmatory test ... 53
3.5 DNA amplification for detection of TEM-, SHV- and OXA- type β-lactamases... 57
4. DISCUSSION... 62
5. APPENDIX ... 68
1. INTRODUCTION
1.1 β-lactam antibiotics
Ever since Fleming discovered penicillin, the first β-lactam antibiotic, in 1928, there has been ongoing progress in their development. Today β-lactams such as broadspectrum penicillins, cephalosporins, monobactams and carabapenems are among the most frequently prescribed antimicrobial agents worldwide due to their efficacy and low toxicity.
β-lactams are a large group of antibiotics all containing the β-lactam ring (Figure 1.1). There are four major groups, penicillins, cephalosporins, carbapenems and monobactems, which differ from one another in the nature of the additional ring attached to this. In penicillins there is a five-membered thiazolidine ring, in cephalosporins a six-membered cephem ring, a double ring in carbapenems whereas in monobactams only the β-lactam ring is present. The various types of β-lactams within each group differ in the side chains attached to the core rings (Samaha-Kfoury and Araj, 2003).
break the already assembled PG chain so that the newly synthesised monomer can be inserted. Transglycodisases join the new monomer to the pre-existing PG chain. The structure is strengthened by transpeptidases cross-linking the peptide chains on the glycan chains. Beta-lactams bind to these transpeptidase enzymes i.e. PBPs and inhibit transpeptidation. This weakens the PG structure and the cell lyses (Black, 2002).
Penicillin was the first β-lactam introduced into clinical use for treatment of infection caused by Gram-positive bacteria. Later the development of broad-spectrum penicillins and first generation cephalosporins began. To increase the usefulness of β-lactam antibiotics to Gram-negative bacteria the pharmaceutical industry started the development of broad- and extended-spectrum β-lactams such as second, third and fourth generation cephalosporins, carbapenems and monobactams (Table 1.1).
1.2 Antimicrobial resistance
Β-lactams Antibiotics Penicillins
Penicillin G, Penicillin, Ampicillin, Ticarcillin, Piperacillin, Mezlocillin Narrow spectrum Broad spectrum Penicillin + β-lactamase inhibitors Amoxycillin-clavulanate, Piperacillin-tazobactam Cephalosporins First generation Second generation Extended-spectrum cephalosporins: Third generation
Cephalothin, Cefazolin, Cephalexin
Cefuroxime, Cefaclor, Cefamandole, Cephamycins (cefoxitin, cefotetan)
Cefotaxime, Ceftriaxone, Cefpodoxime, Ceftazidime, Cefoperazone, Ceftizoxime
Fourth generation Cefepime, Cefpirome
Carbapenems Meropenem, Imipenem, Ertapenem
Monobactams Aztreonam
Antibiotics do not induce mutations but they can create an environment that favours the survival of the resistant mutant.
1.3 Resistance to β-lactam antibiotics
β-lactam resistance in Gram-negative bacilli can occur by three mechanisms; alterations of porin proteins in the cell membrane causing reduced permeability and blocking entry of the drug, the use of an efflux mechanism to pump out the antibiotic as it crosses the membrane, alterations of the target PBPs to prevent β-lactam binding and the production of enzymes that inactivate the antimicrobial agent (Black, 2002). One such enzyme, β-lactamase, is the main cause of clinically significant resistance to β-lactams in Gram-negatives bacilli.
1.4 β-lactamases
β-lactams. To prevent the break down of β-lactam antibiotics due to β-lactamases, clavulanic acid can be used in combination with β-lactam compounds such as amoxycillin-clavulanate (Table 1.1). Clavulanic acid is a β-lactamase inhibitor that in combination with β-lactams acts by permanently inactivates the β-lactamase enzyme in the periplasmic space so that the antibiotic can reach its target, the PBPs.
1.5 Classification of β-lactamases
Table 1.2. Classification scheme for bacterial β-lactamases (Bush et al., 1995) Functional Major Molecular
Class (Ambler) Preferred substrate Inhibition by clavulanic acid Representative enzymes subgroup
Group (Bush)
1 C Cephalosporins - AmpC enzymes from gram-negative bacteria; MIR-1
2 3 4 2a 2b 2be 2br 2c 2d 2e 2f 3a, 3b, 3c A A A A A D A A B Not determined Penicillins Penicillins, cephalosporins Penicillins, narrow-spectrum and extended-spectrum cephalosporins, monobactams Penicillins Penicillins, carbenicillin Penicillins, cloxacillin Cephalosporins Penicillins, cephalosprins, carbapenems
Most β-lactams, including carbapenems Penicillins + + + +/- + +/- + + - -
Penicillinases from gram-positive bacteria TEM-1, TEM-2, SHV-1
TEM-3 to TEM-26, SHV-2 to SHV-6, Klebsiella oxytoca K1
TEM-30 to TEM-36, TRC-1 PSE-1, PSE-3, PSE-4
OXA-1 to OXA-11, (OXA-10) PSE-2
Inducible cephalosporinases from Proteus vulgaris
NMC-A from Enterobacter cloacae, Sme-1, from Serratia marcescens L1 from Xanthomonas maltophilia, CcrA from
Bacteroides fragilis
1.6 β-lactamase mediated β-lactam resistance in Gram-negative bacilli
Many Enterobacteriaceae have chromosomal-mediated β-lactamases. The most important are the AmpC β-lactamases which belong to class C β-lactamases (Bush group 1) (Table 1.2). These AmpC β-lactamases occur in P. aeruginosa, most enterobacteria except Salmonella spp. and some Klebsiella spp. E. coli produces small amounts of AmpC β-lactamases regardless of whether β-lactam antibiotics are present. This small amount of enzyme is insufficient to inhibit the action of the lactam. Hyperproduction of AmpC β-lactamase in E.coli can arise as a result of two mutational events which occur rarely. AmpC β-lactamases are clinically important when hyperproduced because they may confer resistance to a wide variety of β-lactams such as cefoxitin, aztreonam, third generation cephalosporins and β-lactam/clavulanic acid combinations (Shahid et al., 2004). Other species including P. aeruginosa and Enterobacter spp exhibit inducible expression of AmpC β-lactamases and show much greater potential for resistance. In recent years evidence of plasmid-mediated AmpC β-lactamases has emerged. Such plasmids have “escaped” from the chromosomes of Enterobacteriaceae and have reached K. pneumoniae (Livermore and Brown, 2004).
Many Enterobacteriaceae also possess mediated β-lactamases. The first plasmid-mediated β-lactamase in Gram-negative bacilli, TEM-1, was described in the early 1960s and was originally found in a single strain of Escherichia coli isolated from a patient in Greece. Today, TEM-1 is the most common β-lactamase in Gram-negative bacilli and 90% of ampicillin resistance in E. coli is due to the production of TEM-1 (Bradford, 2001). The fact that TEM-1 is plasmid-mediated has facilitated the spread to other bacteria and is now found worldwide in many species of members of the family Enterobacteriaceae, Pseudomonas aerguinosa, Haemophilus influenzae and Neisseria gonorrhoeae. TEM-2 was the first derivative of TEM-1 with only one single amino acid substitution. Another common plasmid-mediated β-lactamase found in both Klebsiella pneumoniae and E. coli is SHV (for sulphydryl variable). In K. pneumoniae the SHV-1 β-lactamase is responsible for 20% of the plasmid-mediated ampicillin resistance (Bradford, 2001). TEM-1, TEM-2 and SHV-1 belong to Amblers class A (Bush group 2b). Other plasmid-mediated β-lactamases found in Gram-negative bacilli include the OXA-type β-β-lactamases. They belong to Ambler class D (Bush group 2d) (Table 1.2). The OXA-type β-lactamases are mainly found in P. aeruginosa but in recent years their incidence in Enterobacteriaceae has increased. The most common OXA-type β-lactamase, OXA-1 has been found in 1 to 10% of E.coli (Paterson and Bonomo, 2005).
the classical plasmid-mediated β-lactamases are susceptible to the β-lactamase inhibitor clavulanic acid though, hyperproduction of these β-lactamases can overcome the action of the inhibitor and permit hydrolysis of the β-lactam.
1.7 Extended-spectrum β-lactamses, ESBLs
Resistance to lactam antibiotics due to chromosomal- and plasmid-mediated β-lactamases in Gram-negative bacilli has become one of the major problems in human medicine. Many β-lactam antibiotics have been developed during the past 20 years to overcome this problem. However, with every new class of antibiotic used to treat patients, new lactamases emerged that caused resistance to that class. The extended-spectrum β-lactam antibiotics (Table 1.1) were specifically designed to resist the hydrolytic action of β-lactamases and were widely used for treatment of serious infections caused by Gram-negative bacilli (Bradford, 2001). But unfortunately, β-lactamase mediated resistance to these extended-spectrum agents quickly emerged in K. pneumoniae and E.coli. This resistance was caused by point mutations in the amino sequence of the classical plasmid-mediated TEM-1, TEM-2, OXA-1 and SHV-1 β-lactamases.
SHV-2 was found in 1983 and the first β-lactamase enzyme able to hydrolyze the third generation cephalosporins and because of its increased spectrum of activity this
seem to be more frequently found in clinical isolates than other types (Paterson and Bonomo, 2005).
The difference between the classical TEM- and SHV-enzymes and ESBL enzymes, result from 1-7 basepair substitutions that alter the configuration and enlarger the active site of the enzyme. Enlarging the active sites makes it possible to hydrolyze bulky side chains such as those used in third generation cephalosporins. Consequently, ESBLs have the ability to hydrolyze antibiotics of the third generation cephalosporins such as ceftazidime, cefotaxime, cefpodoxime, ceftriaxone and aztreonam which belongs to the monobactams. Furthermore, ESBLs have become resistant to fourth generation cephalosporin. They are not active against cephamycins and carbapenems and they are generally inhibited by β-lactamase inhibitors, e.g. clavulanic acid (Al-Jasser, 2006).
The majority of strains expressing TEM- and SHV- type ESBLs belong to the family of Enterobacteriaceae, such as E. coli and K. pneumoniae but ESBLs of the SHV-type seems to be more common in K. pneumoniae (Bradford, 2001). ESBLs have also been found in other species of Enterobacteriaceae and the fact that ESBLs are plasmid-mediated increases the risk of dissemination to other families (Ali Shah et al, 2004).
include OXA-11, OXA-14, OXA-17. These confer resistance to cefotaxime, ceftazidime and aztreonam (Paterson and Bonomo, 2005).
While the majority of ESBLs are derived from TEM or SHV β-lactamases, a few other ESBLs have also been reported that are not related to any of these types of β-lactamases. They are plasmid-mediated but are not simple point mutations of any known β-lactamase, like TEM and SHV. One of them is PER which shares about 25% homology to the TEM and SHV-type ESBLs. This type is very rare and has only been found in South America. Another type that is closely related to PER is VEB which was first found in an isolate of E. coli from Vietnam (Al-Jasser, 2006). PER and VEB-type of ESBLs confer resistance to third generation cephalosporins, especially ceftazidime and aztreonam. BES, SFO, TLA, GES and IBC are other types of the expanding family of ESBLs (Ali Shah et al, 2004).
1.8 Epidemiology and clinical significance of ESBLs
ESBL-producing strains were first detected in Western Europe, most likely because extended-spectrum β-lactam antibiotics were first used there clinically. Therefore, Europe has the highest level of ESBL-producing bacteria in the world. It is interesting that specific ESBLs appear to be unique to a certain country or region. Comparing the United States to Europe, TEM-3 ESBL producing strains have only been found in France but not in the United States while TEM-10 is common in both countries (Ali Shah et al, 2004). SHV-5 β-lactamase is very common worldwide and has been found in France, Greece, Poland, Hungary, South Africa, the United Kingdom and the United States. SHV-12 is the most common ESBL found in Korea (Bradford, 2001). Unfortunately, there are few epidemiologic data on the geographical spread of OXA-type ESBLs. According to a survey by the Clinical and Laboratory Standards Institute (CLSI) the prevalence of ESBLs is underestimated (Colodner, 2005). This may be due to the difficulty in detecting ESBL-producing bacteria in routine laboratories and the inconsistencies in reporting ESBL outbreaks.
Infections caused by ESBL-producers are associated with high rates of morbidity and mortality, prolonged hospital stay and heavy expenses. This is due to the increased rate of treatment failure of these infections. Infections caused by ESBLs can not be treated with cephalosporins but since ESBLs are difficult to detect and many clinical laboratories are not aware of them, cephalosporins are often used anyway. Several studies have reported that carabapenems are the best alternative for treating serious infections caused by ESBL-positive isolates (Colodner, 2005). These antibiotics are highly stable to β-lactamases, however, they are very expensive and the usage should be limited. It has been reported that ESBL-positive isolates can become resistant to carbapenems due to loss of porin proteins in the outer membrane (Sturenburg and Mack, 2003). Plasmids responsible for ESBL production frequently carry genes encoding resistance to other groups of commonly used antibiotics and therefore, the quinolones and aminoglycosides may also be unsuitable for use. Multidrug resistance is most common in K. pneumoniae and E. coli.
1.9 Detection methods for ESBLs
carriage of ESBL-encoding genes (Sturenburg and Mack, 2003). It is important that the detection of ESBLs reflects the level of resistance that would be demonstrated by strains expressing these enzymes in vivo. The fact that ESBL-producing bacteria can be acquired in the community also increases the need for reliable detection methods. Clinical laboratories may not be aware of the importance of screening for ESBL when dealing with infections originating in the community such as urinary tract infections (Paterson, 2006). The lack of sensitivity and specificity in traditional susceptibility tests to detect ESBLs has prompted the search for an accurate test to detect the presence of ESBLs in clinical isolates (Bradford, 2001).
1.9.1 Phenotypic detection methods for ESBLs
Detection methods for ESBLs usually include an initial screening test for reduced susceptibility to a range of extended spectrum cephalosporins such as cefpodoxime, cefotaxime, ceftriaxone, ceftazidime and aztreonam. This is then followed by a phenotypic confirmatory testing where the ability of a β-lactamase inhibitor, usually clavulanic acid, to reduce the resistance is tested. The CLSI (Clinical and Laboratory Standards Institute) has provided guidelines for detection of ESBL-production using an initial screening method and a phenotypic confirmatory test based on synergy between an indicator cephalosporin and clavulanic acid. If an ESBL is detected, the strain should be reported as resistant to all extended-spectrum cephalosporins and aztreonam regardless of the susceptibility testing result.
1.9.2 ESBL disc diffusion screen
antibiotic susceptibility testing can screen for ESBL-production by noting specific zone diameters which indicate high level of suspicion for ESBL-production. In this test susceptibility to ceftriaxone, cefotaxime, ceftazidime, cefpodoxime and aztreonam is tested in accordance with CLSI susceptibility guidelines and interpretations made using specifically devised breakpoints. If any of the zone diameters are reduced and indicate suspicion for ESBL production, phenotypic confirmatory test should be used for verifying. The sensitivity for this test improves if all these five antibiotics are used.
1.9.3 Confirmatory tests for ESBLs in Klebsiella spp and E. coli 1.9.3.1 Double disc synergy test
This test was described by Jarlier et al. and one of the first detection tests for ESBL- production used (Jarlier et al, 1988). The test is based on the synergy between a cephalosporin and clavulanic acid. The synergy effect is detected when a disc of amoxicillin/clavulanic acid (20/10μg) is placed 30mm apart (center to center) from a disk containing a third generation cephalosporin. Extension of the edge of the cephalosporin zone on the side exposed to the disc containing clavulanic acid caused by synergy, indicate the presence of an ESBL. The major advantage of this test is its low cost and ease of use. Unfortunately, interpretation could be very subjective because the optimal distance between the discs for the demonstration of synergy may vary with the individual strain.
1.9.3.2 Cephalosporin/clavulanic acid combination disc tests
discs. However, the use of both antibiotic discs is advisable. A ≥ 5mm increase in a zone diameter for either antimicrobial agent tested in combination with clavulanic acid versus its zone when tested alone, indicate the presence of and ESBL.
1.9.3.3 Broth microdilution
Phenotypic confirmatory testing can also be performed by broth microdilution assays using extended spectrum cephalosporins in absence and presence of a fixed concentration of clavulanic acid e.g. ceftazidime 0.25 to 128μg/mL, ceftazidime plus clavulanic acid 0.25/4 to 128/4, cefotaxime 0.25μg to 64μg/mL and cefotaxime plus clavulanic acid 0.25/4 to 64/4. Phenotypic confirmation is considered as a ≥ two-threefold-serial dilution decrease in MIC of either cephalosporin in the presence of clavulanic acid compared to its MIC when tested alone (Paterson and Bonomo, 2005).
1.9.4 Commercial Methods for ESBL detection 1.9.4.1 E-test ESBL strips
1.9.4.2 VITEK ESBL test
The VITEK 2 is an automated system that performs bacterial identification and antibiotic susceptibility testing. The instrument includes an expert software system designed to interpret antibiotic resistance profiles such as ESBL production. In this is system ESBL prediction is based on the simultaneous assessment of the inhibitory effects of cefepime, cefotaxime and ceftazidime, alone and in presence of clavulanic acid (Spanu et al., 2006).
Although the phenotypic ESBL confirmatory tests are generally sensitive and specific there are a number of instances where these confirmatory tests may generate false positive or false negative results. AmpC β-lactamases are not inhibited by clavulanic acid and may therefore mask the synergistic effect of clavulanic acid against ESBLs. The coexistence of both ESBLs and plasmid-mediated AmpC β-lactamases in K. pneumoniae may therefore result in false negative ESBL results. However, although AmpC β-lactamases are very resistant to third generation cephalosporins as well as cefoxitin they may remain susceptible to fourth generation cephalosporins. The use of fourth generation cephalosporins such as cefepime and cefepime plus clavulanic acid may therefore be used to detect ESBL in the presence of AmpC β-lactamases. Problems with confirmatory test may also arise in K. oxytoca. It is possible to misinterpret a K1- producer as an ESBL-producer if ceftazidime is omitted in the antibiotic test panel (Essack, 2000).
1.9.5 Molecular detections methods for ESBLs
oligonucleotide primer sets for TEM-, SHV- and OXA-type β-lactamases have been used in multiplex PCR to amplify internal regions in the target β-lactamase genes. Therefore, because the oligonucleotide primer sets used do not amplify the specific point mutations in the ESBL gene both classical and ESBL derivates will be amplified and detected. Nucleotide sequence analysis of the resulting amplicons is therefore required to confirm ESBL production. Other PCR-based detection methods involve the design of specific primers targeting the specific ESBL mutations. However, in this case, for each new point mutation that occurs in ESBLs new primers must be designed.
1.10 The aim of this project
The aim of this project is to evaluate the phenotypic and genotypic detection methods for extended-spectrum β-lactamases, ESBLs. The study was designed to:
• confirm the need for screening for ESBL production
• examine the optimum antimicrobial agent(s) for ESBL detection
• compare two phenotypic ESBL confirmatory tests, the double disc synergy test and the combination disc test
2. MATERIALS AND METHODS
2.1 Material
2.1.1 Culture media
• Columbia agar base, LabM Lot: 088997 • Mueller Hinton agar, LabM Lot: 087118/207
• Horse blood defibrinated, Charles River Laboratories Biolabs Europe Lot: 070323
2.1.4 Reagents
• API 20E, bioMérieux Lot: 798892001
• McFarland standard 0.5, bioMérieux Lot: 799104401 • Microbeads, Pro-Lab Diagnostics Lot: 2697R
2.1.2 Bacterial isolates
• ATCC 25922 E. coli (negative for ESBL production)
• ATCC 700603 K. pneumoniae (positive for SHV-type β-lactamase) • ATCC 35218 E. coli (positive for TEM-type β-lactamase)
• Strain of Enterobacter cloacae positive for AmpC β-lactamase, Microbiology Department, St. James´s Hospital, Dublin
• Strain of E. coli positive for OXA-type β-lactamase, Microbiology Department, St. James´s Hospital, Dublin
2.1.3 Antimicrobial discs
OXOID Antimicrobial susceptibility test discs
Cephalosporins:
• Ceftazidime (CAZ) 30μg Lot: 499421 • Cefotaxime (CTX) 30μg Lot: 468511 • Ceftriaxone (CRO) 30μg Lot: 487145 • Cefpodoxime (CPD) 10μg Lot: 349371 • Cefoxitin (FOX) 30μg Lot: 497249 • Cephalothin (KF) 30μg Lot: N/A
Penicillins:
• Amoxycillin/Clavulanic acid (AUG) 20/10μg Lot: 494166 • Piperacillin (PRL) 30μg Lot: 370531
• Ampicillin (AMP) 10μg Lot: 481007
Monobactams:
• Aztreonam (ATM) 30μg Lot: 397531
Carbapenems:
• Meropenem (MEM) 10μg Lot: N/A
Aminoglycosides:
• Amikacin (AK) 30μg Lot: 388110 • Gentamicin (CN) 10μg Lot: 461297
Quinolones:
• Nalidixic acid (NA) 30μg Lot: 459341
Fluoroquinolones:
Tetracyclines:
• Tetracycline (TE) 30μg Lot: 478353
Others:
• Chloramphenicol (C) 30μg Lot: 471676
• Trimethoprim/sulfamethoxazole (SXT) 5μg Lot: 392405
MAST Diagnostics ESBL Combination discs
• Cefpodoxime 10μg / Cefpodoxime 10μg + Clavulanic acid 1μg Lot: 210186 • Ceftazidime 30μg / Ceftazidime 30 μg + Clavulanic acid 10μg Lot: 210185 • Cefepime 30μg / Cefepime 30μg + Clavulanic acid 10μg Lot: 203322 • Cefotaxime 30μg / Cefotaxime 30μg + Clavulanic acid 10μg Lot: 210506
2.1.5 PCR-reagents
• 100mM dNTP, Invitrogen Cat no: 10297-018 Lot: 1384321
• Taq DNA-polymerase recombinant 500 U, Invitrogen Cat no: 10342 Lot: 1368149 • 10XPCR Buffer (-MgCl2), Invitrogen Cat no: 10342-020 Lot: 1361222
• 50mM MgCl2, Invitrogen Cat no: 10342-020 Lot: 1355421
• TEM, SHV and OXA-primers, Sigma Genosys
2.1.6 PCR-equipment
• Hybaid Omn-E thermal cycler
2.1.7 Equipment for agarose gel electrophoresis
• Autoclaved MiliQ-water
• 1 X Tris-borate-EDTA (TBE) buffer
• Gel loading buffer X 6 (0.25% xylene cyanole, 0.25% bromophenol blue, 30% glycerol in water)
• Ethidium Bromide (0.5μg/mL)
• Molecular weight ladder 100 bp, Invitrogen
2.2 Methods
2.2.1 Bacterial isolates
Site E. coli K. pneumoniae
Table 2.1. Clinical isolates from St. James´s Hospital, Dublin
K. oxytoca
Bloodculture 13 1 -
Sputa 9 9 1
Wounds 7 1 3
Urine 1 1 1
On receipt of the isolates, the bacteria were refrozen on protect beads for long term storage. Blood agar (Columbia agar base plus 7% defibrinated horse blood) was prepared according to the manufacturer’s instructions and was inoculated with the test organism and incubated at 37°C for 16-20 hours. After, incubation 3-5 large colonies of the isolate were inoculated into the cryopreservative solution in the protect tube. The solution was removed after 5 minutes with a sterile pasteur pipette and the protect beads were stored at -20°C. The isolates were removed from -20°C and cultured from the beads onto blood agar on the day prior to testing.
2.2.2 API 20E identification
result of the API 20E test and this profile was processed in a software program for full identification of the bacteria.
2.2.3 Antimicrobial susceptibility testing
Antimicrobial susceptibility testing was performed on all 47 isolates against 13 commonly used antibiotics using the CLSI disc diffusion method. Bacterial colonies were suspended in sterile water to a turbidity equivalent to a McFarland 0.5 standard. The suspension was inoculated on Mueller Hinton agar with a sterile swab and within 15 minutes of inoculation 5 antibiotic discs were applied on each plate. The 13 antibiotic discs were cefoxitin (30μg), cephalothin (30μg), amoxycillin/clavulanic acid (20/10μg), piperacillin (30μg), ampicillin (10μg), amikacin (30μg), gentamicin (10μg), meropenem (10μg), nalidixic acid (30μg), ciprofloxacin (5μg), tetracycline (30μg), chloramphenicol (30μg) and trimethoprim/sulfamethoxazole (5μg). The plates were incubated at 37°C for 16-20 hours. The diameters of the zones of inhibition were measured using a ruler and compared to zone diameter interpretive standards (Performance Standards for Antimicrobial Susceptibility Testing; M100-S16, CLSI, 2006).
2.2.4 Screen for ESBL
2.2.4.1 CLSI ESBL screen method
suspension was inoculated onto Mueller Hinton agar with a sterile swab and within 15 minutes of inoculation the antibiotic discs were applied. The five antibiotic discs used were aztreonam (30μg), cefotaxime (30μg), cefpodoxime (10μg), ceftazidime (30μg) and ceftriaxone (30μg). The plates were incubated at 37°C for 16-20 hours. The diameters of the zones of inhibition were measured using a ruler and compared to the CLSI breakpoints for ESBL screen (Performance Standards for Antimicrobial Susceptibility Testing; M100-S16, CLSI). The diameters of the zones of inhibition were also measured using the disc susceptibility zone diameter interpretive standards.
2.2.5 ESBL confirmatory tests 2.2.5.1 Double disc synergy test
16-20 hours. An enhanced zone of inhibition between either of the cephalosporin antibiotics and the amoxycillin/clavulanic acid disc was indicative of synergistic activity with clavulanic acid and the presence of an ESBL.
2.2.5.2 The cephalosporin/clavulanic acid combination disc test
2.2.6 DNA amplification of TEM-, SHV- and OXA-type β-lactamases
DNA amplification for TEM-, SHV- and OXA-encoding genes was performed on all 47 strains using a PCR method according to Colom et al. (Colom et al., 2003). In this study, 3 primer sets were used to amplify an internal region of the 3 β-lactamase genes TEM, SHV and OXA-1 (Figure 2.2 and 2.3). The primers were designed using the CLUSTALW multiple sequence alignment program and were selected from the most conserved regions of the β-lactamase genes (Table 2.2).
Initially, PCR for the individual β-lactamase genes was performed in singleplex. Then following optimisation, the 3 sets of primers were combined in a single multiplex assay.
♦ Bp numbering according to Lal P., Kapil A. and Das BK. * Bp numbering according to Sutcliffe, 1978 ● Bp numbering according to Ouellette et al., 1987
Primer Tm (°C) Sequence (5´- 3´) Bp position Size of amplicons (bp) SHV-F 59.3 AGGATTGACTGCCTTTTTG 425-444♦ 392 SHV-R 61.5 ATTTGCTGATTTCGCTCG 801-818 TEM-H 60.6 CCCCGAAGAACGTTTTC 384-401* 516 TEM-C 53.2 ATCAGCAATAAACCAGC 884-901 OXA-F 56.4 ATATCTCTACTGTTGCATCTCC 83-104● 619 OXA-R 58.7 AAACCCTTCAAACCATCC 685-702
2.2.6.1 Preparation of DNA template for multiplex PCR using the boiling method
A large colony were suspended in 50μL of TE-buffer and boiled for 10 minutes. The suspension was centrifuged at 12000 rpm for 5 minutes. The supernatant was used as a source of extracted DNA and was stored on ice to preserve the DNA.
2.2.6.2 Optimizing of SHV-primer in singleplex PCR
PCR amplification of the SHV β-lactamase gene was performed in a singleplex PCR assay using the SHV-forward and SHV-reverse primer. ATCC 25922 E. coli (SHV-negative), ATCC 700603 K. pneumoniae (SHV-positive), ATCC 35218 E. coli (TEM-positive) and a strain of E. coli (OXA-positive) were used as controls and were included in each batch of tests. The initial optimization of the SHV PCR was performed using 13 of the clinical isolates (7 E. coli and 6 K. pneumoniae). Optimization of the assay was performed by increasing the recommended annealing temperature from 54°C to 56°C, decreasing the primer concentration from 1μM to 0.5μM and increasing the amount of DNA template from 1μl to 2μl. A titration of MgCl2 covering the concentration range of 0.5mM to 4.0mM
was also performed. When the optimum conditions for the SHV-primer were established, PCR amplification was performed in 25μl reaction mixtures containing 0.2mM of dNTP,
1.5mM MgCl2, 0.5μM of SHV-primers, 2 U of Taq DNA-polymerase and 1 X PCR
and primer extension at 72°C for 1 minute. A final extension step at 72°C for 10 minutes and a final hold step at 4°C were included.
2.2.6.3 Optimizing of TEM-primer in singleplex PCR
The TEM PCR assay was optimized using the same 13 isolates used for the SHV assay. ATCC 25922 E. coli (TEM-negative), ATCC 700603 K. pneumoniae (SHV-positive), ATCC 35218 E. coli (TEM-positive) and a strain of E. coli (OXA-positive) were used as controls and included in each batch of tests. The reaction was then optimised by decreasing the primer concentration from 1.0μM to 0.5μM and increasing the annealing temperature from 54°C to 55°C. In the optimised TEM assay PCR amplification was performed in 25μl reaction mixtures containing 0.2mM of dNTP, 1.5mM MgCl2, 0.5μM of TEM-primers,
2 U of Taq DNA-polymerase and 1 X PCR Buffer. As before, autoclaved MiliQ-water was added to a total volume of 25μl, mastermix was prepared for each batch of tests, was aliquoted into individual tubes and 2μl of DNA template was added. A negative amplification control containing 2μl of autoclaved MiliQ-water was included in each batch of tests. DNA was amplified with the Hybaid Omn-E thermal cycler and the PCR programe consisted of a pre-denaturation step at 94°C for 3 minutes followed by 35 cycles of DNA denaturation at 94°C for 45 sec, primer annealing at 55°C for 30 sec, primer extension at 72°C for 1 minute, a final extension step at 72°C for 10 minutes and a final hold step at 4°C.
2.2.6.4 Optimizing of OXA-primer in singleplex PCR
included as a negative control. The OXA assay was optimised using 0.5μM primer and the PCR was performed as previously outlined using 55°C as the primer annealing temperature.
2.2.6.5 Optimizing of SHV, TEM and OXA-primer in multiplex PCR
The 3 primer sets were optimized using the same 13 clinical isolates and control strains used in the singleplex assay. Initially, primer sets for SHV and TEM were optimised together in a single reaction (primer concentration of 0.5μM and annealing temperature at 55°C). Finally the OXA primer set was added and the multiplex assay was optimised at a primer concentration of 0.5μM and an annealing temperature at 55°C. The PCR reaction mixture was prepared as previously described. Multiplex PCR was performed on all 47 isolates.
2.2.7 Detection of amplified DNA-products using agarose gel electrophoresis
3. RESULTS
3.1 Antimicrobial susceptibility test
The results of the antimicrobial suscetibility test are summarized in Figure 3.1 and Figure 3.2. The figures demonstrate the percentage of resistant, intermediate and sensitive isolates to a range of common groups of antibiotics for the 30 isolates of E. coli and 17 isolates of Klebsiella spp. tested. The breakpoints used were the interpretive standards recommended by the CLSI. The quality control isolates used in this test were within the allowed control limits. E. coli 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Peni cillins Ceph alos porins Monob actam s Carba pene ms Aminog lycosi des Tetrac yclin es Phen icols Quino lones Fluo roqui nolo nes Sulfon amide s Resistant Intermediate Sensitive
Figure 3.1. Antimicrobial susceptibility testing of 30 isolates of E. coli. The figure shows the percentage of resistant, intermediate and sensitive isolates to a range of commonly groups of antibiotics.
there was also a high level of resistance ~50% against the tetracyclines, quinolones and fluoroquinolones. Few isolates demonstrated resistance to the aminoglycosides.
Klebsiella spp 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Penic illins Cep halos porins Monob actam s Carb apen ems Aminog lyco sides Tetra cyclin es Phen icols Quino lones Fluo roqu inolone s Sulfo nam ides Resistant Intermediate Sensitive
Figure 3.2. Antimicrobial susceptibility testing of 17 Klebsiella spp. The figure shows the percentage of resistant, intermediate and sensitive isolates to a range of commonly groups of antibiotics.
3.2 CLSI ESBL screen method
Cefpodoxime 10μg E. coli 0 2 4 6 8 10 12 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 17 mm Breakpoint susceptibility test ≤ 20 mm
Figure 3.3CLSI ESBL screen for 30 E. coli using cefpodoxime 10μg. The breakpoint for cefpodoxime according to CLSI ESBL screen is ≤ 20 mm . The breakpoint for cefpodoxime according to CLSI susceptibility test is ≤ 17 mm.
Cefpodoxime 10μg Klebsiella spp 0 2 4 6 8 10 12 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint susceptibility test ≤ 20 mm Breakpoint ESBL screen ≤ 17 mm
Ceftazidime 30μg E. coli 0 1 2 3 4 5 6 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 22 m m Breakpoint susceptibility test ≤ 17 m m Ceftazidime 30μg Klebsiella spp 0 1 2 3 4 5 6 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 22 m m Breakpoint susceptibility test ≤ 17 m m
Figure 3.5 CLSI ESBL screen for 30 E. coli using ceftazidime 30μg. The breakpoint for ceftazidime according to CLSI ESBL screen is ≤ 22 mm. The breakpoint for ceftazidime according to CLSI susceptibility test is ≤ 17 mm. Compared with the ESBL screen test the number of isolates interpreted as non ESBL producers using the breakpoints for CLSI susceptibility test is 5 (isolates lying between the two arrows).
Cefotaxime 30μg E. coli 0 1 2 3 4 5 6 7 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 27 m m Breakpoint susceptibility test ≤ 22 mm
Figure 3.7 CLSI ESBL screen for 30 E. coli using cefotaxime 30μg. The breakpoint for cefotaxime according to CLSI ESBL screen is ≤ 27 mm. The breakpoint for cefotaxime according to CLSI susceptibility test is ≤ 22 mm. Compared to the ESBL screen test the number of isolates interpreted as non ESBL producers using the breakpoints for CLSI susceptibility test is 1 (isolate lying between the two arrows).
Cefotaxime 30μg Klebsiella spp 0 1 2 3 4 5 6 7 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 27 mm Breakpoint susceptibility test ≤ 22 mm
Ceftriaxone 30μg E. coli 0 1 2 3 4 5 6 7 8 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint susceptibility test ≤ 20 mm Breakpoint ESBL screen ≤ 25 m m
Figure 3.9 CLSI ESBL screen for 30 E. coli using ceftriaxone 30μg.The breakpoint for ceftriaxone according to CLSI ESBL screen is ≤ 25 mm. The breakpoint for cefotaxime according to CLSI susceptibility test is ≤ 20 mm. Compared to the ESBL screen test the number of isolates interpreted as non ESBL producers using the breakpoints for CLSI susceptibility test is 1 (isolates lying between the two arrows).
Ceftriaxone 30μg Klebsiella spp 0 1 2 3 4 5 6 7 8 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 25 m m Breakpoint susceptibility test ≤ 20 mm
Figure 3.10 CLSI ESBL screen for 17 Klebsiella spp using ceftriaxone 30μg. The
Aztreonam 30μg E. coli 0 1 2 3 4 5 6 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 27 m m Breakpoint susceptibility test ≤ 21 mm Aztreonam 30μg Klebsiella spp 0 1 2 3 4 5 6 0 3 6 9 12 15 18 21 24 27 30 33 36 Zone diameter (mm) Number of isolates Breakpoint ESBL screen ≤ 27 m m Breakpoint susceptibility test ≤ 21 m m Figure 3.11 CLSI ESBL screen for 30 E. coli using aztreonam 30μg.
The breakpoint for aztreonam according to CLSI ESBL screen is ≤ 27 mm. The breakpoint for aztreonam according to CLSI susceptibility test is ≤ 21 mm. Compared to the ESBL screen test the number of isolates interpreted as non ESBL producers using the breakpoints for CLSI susceptibility test is 5 (isolates lying between the two arrows).
In total, 85% of clinical isolates of E. coli which had been previously collected as suspect ESBL producers were positive in the ESBL screen test using either cefotaxime or aztreonam (Table 3.1). A lower percentage of E. coli isolates were positive in the ESBL screen using cefpodoxime (80%), ceftriaxone (75%) and ceftazidime (65%). For Klebsiella spp. 94% were positive in the ESBL screen using aztreonam. A lower percentage of Klebsiella spp. were positive in the ESBL screen using cefpodoxime (88%), cefotaxime (88%), ceftriaxone (88%) and ceftazidime (71%) (Table 3.2).
3.3 ESBL confirmatory tests (double disc synergy test and the combination disc test)
The CLSI ESBL screen was followed by two ESBL confirmatory tests, double disc synergy test and combination disc test. The results from each test are summarized in Figure 3.13 and 3.14.
The control isolates used in these test were as expected (Figure 3.15 and 3.16 negative control, Figure 3.17 and 3.18 positive control). Figure 3.13 demonstrates the overall percentage of positive, negative and indeterminate isolates obtained in the double disc synergy test. Figure 3.14 demonstrates the overall percentage of positive (≥ 5mm) and negative (no change in mm) isolates obtained in the combination disc test. The percentage of isolates demonstrating an increase in zone size below the 5mm cut off is also included.
synergistic effect between the clavulanic acid and the cephalosporin in the combination disc test. However, those isolates were not confirmed as positive since the difference in the zones of inhibition was < 5 mm.
ESBL confirmatory test Double disc synergy test
0% 10% 20% 30% 40% 50% 60% 70%
Positive Negative Indeterminate
E. coli
Klebsiella spp
Figure 3.13 Double disc synergy test. Percentage of positive, negative and indeterminate isolates for 30 E. coli and 17 Klebsiella spp.
ESBL confirmatory test Combination disc test
0% 10% 20% 30% 40% 50% 60% 70% ≥ 5 mm < 5 mm No change E. coli Klebsiella spp
Figure 3.15 Double disc synergy test Figure 3.16 Combination disc test ATCC 25922. Cefpodoxime 10μg (left), cefpodoxime 10μg + clavulanic acid 1μg (right).
ATCC 25922. Ceftriaxone 30μg (left), Cefpodoxime 10μg (right), amoxycillin/ clavulanic acid 20/10μg (in the middle).
Figure 3.18 Combination disc test ATCC 700603. Ceftazidime 30μg (right),
ceftazidime 30μg + clavulanic acid 10μg (left).
Figure 3.17 Double disc synergy test ATCC 700603. Ceftriaxone 30μg (left), Cefpodoxime 10μg (right), amoxycillin/ clavulanic acid 20/10μg (in the middle).
E. coli. Cefotaxime 30μg (right), cefotaxime 30μg + clavulanic acid 10μg (left).
Figure 3.20 Combination disc test Figure 3.19 Double disc synergy test
E. coli. Ceftriaxone 30μg, cefpodoxime 10μg, ceftazidime 30μg, cefotaxime 30μg, amoxycillin/clavulanic acid 20/10μg (in the middle).
3.4 Combined result for disc susceptibility test, the ESBL screen and the combination disc confirmatory test
Table 3.1 and 3.2 represents the combined results of the cephalosporin disc susceptibility test, the ESBL screen and the confirmatory test. Table 3.1 summarises the results of the 20 E. coli isolates originally selected as presumptive ESBL positive isolates. The remaining 10 isolates of E. coli that were selected as sensitive controls were all negative in the ESBL screen and confirmed as non ESBL producers in the combination disc confirmatory test.
In the ESBL screen 65% of E. coli indicated ESBL production using ceftazidime. Five of these isolates would not have been detected using the CLSI susceptibility test. These five isolates were confirmed as ESBL producers in the combination disc test (Table 3.1). Of the 35% that failed to indicate the presence of an ESBL in the ESBL screen test, three false negative isolates were found when tested in the combination disc test. In addition, three further isolates detected in the ESBL screen failed to confirm in the combination disc test i.e. false positives.
In the case of Klebsiella spp. (Table 3.2) 71% of the isolates were considered as ESBL producers in the ESBL screen but only 59% were confirmed containing ESBLs when using ceftazidime. One additional isolate was identified as an ESBL producer in the combination disc test. When using cefpodoxime and cefotaxime 88% of isolates indicated ESBL production in the ESBL screen test, but only 47% were confirmed with cefpodoxime and 53% with cefotaxime in the combination disc test (Table 3.2).
CAZ 30μg CPD 10μg CTX 30μg CRO 30μg ATM 30μg CPM 10μg Susceptibility test
* 20 E. coli isolates originally selected as presumptive ESBL producers. The results of the 10 susceptible E. coli isolates are excluded from this table. CAZ = ceftazidime, CPD = cefpodoxime, CTX = cefotaxime, CRO = ceftriaxone, ATM = aztreonam, CPM = cefepime
N/A = not applicable Susceptible Intermediate Resistant 65 % 10 % 25 % 10 % 10 % 80 % 25 % 25 % 50 % 30 % 15 % 55 % 45 % 15 % 40 % N/A
ESBL disc diffusion screen
65% (n =13) 35% (n =7) 75 % (n=15) 25 % (n=4) 85 % (n=17) N/A Positive Negative 80% (n =16) 20% (n =4) 85 % (n=17) 15 % (n=3) 15 % (n= 3)
ESBL confirmatory test
65 % (n=13) 70 % (n=14)
N/A N/A (Combination disc diffusion)
Confirmed as ESBL producer 65 % (n=13) 50 % (n=10)
CAZ 30μg CPD 10μg CTX 30μg CRO 30μg ATM 30μg CPM 10μg Susceptibility test
CAZ = ceftazidime, CPD = cefpodoxime, CTX = cefotaxime, CRO = ceftriaxone, ATM = aztreonam, CPM = cefepime N/A = not applicable
Susceptible Intermediate Resistant 35 % 12 % 53 % 6 % 6 % 88 % 41 % 35 % 24 % 18 % 53 % 29 % 18 % 35 % N/A 47 % ESBL disc diffusion screen
88 % (n=15) 12 % (n=2) 88 % (n=15) 12 % (n=2) 94 % (n=16) 71 % (n=12) N/A Positive Negative 88 % (n=15) 12 % (n=2) 29 % (n=5) 6 % (n= 1)
ESBL confirmatory test
59 % (n= 10) 53 % (n=9)
N/A N/A (Combination disc diffusion)
Confirmed as ESBL producer 47 % (n=8) 35 % (n= 6)
3.5 DNA amplification for detection of TEM-, SHV- and OXA- type β-lactamases
In the initial amplification with SHV-primers, non-specific PCR products were observed for the TEM and OXA-positive control isolates. In order to optimize the conditions for the SHV amplification the primer concentration was decreased, the amount of template DNA was increased, the annealing temperature was increased and a MgCl2 titration was
performed. Figure 3.23 shows the result of this optimization. Decreasing the primer concentration and increasing the template DNA failed to remove the non-specific products (figure 3.23 panel A, lane 7-11). When the annealing temperature was increased to 56°C there was no non-specific amplified product seen in (figure 3.23 panel B, lane 7) compared to (figure 3.23 panel A, lane 2). The MgCl2 concentration of 1.5mM generated a single
Panel A 1 2 3 4 5 6 7 8 9 10 11
Panel B 1 2 3 4 5 6 7
Panel C 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Panel A Optimisation of SHV primers by modification of SHV primer and extracted DNA concentration. Lane 1
100bp standard size reference marker, lanes 2- 6: ATCC 25922, SHV positive control, TEM positive control, OXA positive control and NTC using 1μL of extracted DNA and1μM SHV-prime. Lanes 7-11: ATCC 25922, positive control SHV, positive control TEM, positive control OXA and NTC using 2μL of extracted DNA and 0.5μM SHV-primer.
Panel B Optimisation of SHV primers by increasing primer annealing temperature. Lane 1 100bp standard size
reference marker lanes 2-7: ATCC 25922, SHV positive control, TEM positive control, OXA positive control, clinical isolate K. pneumoniae and NTC using a primer annealing temperature of 56ºC.
Panel C Optimisation of SHV primers by performing a MgCl2 titration curve. Lane 1, 100bp standard size
reference marker, lanes 2-10 MgCl2 titration of SHV positive control: 0mM MgCl , 0.5mM MgCl2, 1.0mM
MgCl2, 1.5mM MgCl2, 2.0mM MgCl2. 2.5mM MgCl2 , 3.0mM MgCl2, 3.5mM MgCl2 and 4.0mM MgCl2, Lanes
11-19 MgCl2 titration of OXA positive control: 0mM MgCl , 0.5mM MgCl2, 1.0mM MgCl2, 1.5mM MgCl2,
2.0mM MgCl2. 2.5mM MgCl2 , 3.0mM MgCl2, 3.5mM MgCl2 and 4.0mM MgCl2
The optimised conditions for the SHV, TEM and OXA primers in singleplex PCR are demonstrated in Figure 3.24. Using the SHV-primers, a PCR product of the expected size (392 bp) was amplified from the positive control isolate ATCC 700603 K. pneumoniae (Figure 3.24, panel A, lane 3). This figure also demonstrates the presence of a SHV specific product in many clinical isolates (Figure 3.24, panel A, lane 8, 10, 11, 12 and 13). Using the TEM-primers, a PCR product of the expected size (516 bp) was amplified from the positive control isolate ATCC 35218 E. coli (Figure 3.24, panel B, lane 4). However, PCR products of the expected size for TEM were also demonstrated in the SHV- and OXA-positive control isolates although PCR bands were fainter in appearance (Figure 3.24 panel B, lanes 3 and 5). These cross-reactive PCR products were not evident in the multiplex assay. Using the OXA-primer, a PCR product of the expected size (619 bp) was amplified from the control isolate of E. coli positive for OXA and in other clinical isolates (Figur 3.24, panel C, lane 5 and 6).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 500bp 400bp Panel D 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 400bp 600bp 500bp Panel C 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 600bp
Lanes 1 and 20: 100bp standard size reference marker, lane 2. ATCC 25922, lane 3. SHV positive control, lane 4. TEM positive control, lane 5. OXA positive control, lane 6 clinical isolate of E. coli, lane 7. clinical isolate of K. oxytoca, lane 8. clinical isolate of K. pneumoniae, lane 9. clinical isolate of E. coli, lanes 10-13. clinical isolates of K. pneumoniae, lanes 14- 18. clinical isolates of E. coli and lane 19. NTC.
Panel D Multiplex PCR performed using SHV, TEM and OXA-primers at 0.5μM and an annealing temperature of 55 ºC
Panel C Singleplex PCR performed using OXA-primers at 0.5μM and an annealing temperature of 55 ºC Panel A Singleplex PCR performed using SHV-primers at 0.5μM and an annealing temperature of 56ºC. Panel B Singleplex PCR performed using TEM-primers at 0.5μM and an annealing temperature of 55 ºC.
4. DISCUSSION
In Gram-negative bacilli, lactamases remain the most important contributing factor to β-lactam resistance and their increasing prevalence and evolution represents a serious challenge for clinical microbiology laboratories. In addition to increasing resistance to cephalosporins, resistance to other commonly used antibiotics such as fluoroquinolones is increasing. The results of this study indicate that almost 40-50% of E. coli and Klebsiella spp. isolates are resistant to third generation cephalosporins. However, whilst the high rate of resistance to other agents such as the fluoroquinolones is a concern, the aminoglycosides remain active. The high rate of susceptibility to aminoglycosides demonstrated in this study supports the continued use of combination therapy involving cephalosporins and aminoglycosides for the treatment of resistant Gram-negative bacilli.
According to the guidelines provided by CLSI it is recommended to use one of the five extended cephalosporins for ESBL screening i.e. cefpodoxime, ceftazidime, ceftriaxone, cefotaxime or aztreonam. However, the use of more than one of these indicator cephalosporins improves the sensitivity (Performance Standards for Antimicrobial Susceptibility Testing; M100-S16 2006).
The results of this study show that the best indicator cephalosporin for detection of ESBLs in E. coli is cefpodoxime. This cephalosporin demonstrated the highest rate of ESBL detection in E. coli in the susceptibility test and the ESBL screen which were subsequently confirmed. Ceftazidime, on the other hand, demonstrated the highest percentage of misinterpreted isolates in the susceptibility test and the lowest detection rate in the ESBL screen.
In the case of Klebsiella spp. the best indicator cephalosporins are ceftazidime and cefpodoxime. Cefpodoxime demonstrated the highest rate of detection both in the susceptibility test and the ESBL screen. However, the high degree of false positivity associated with this indicator cephalosporin indicates the need for a second indicator cephalosporin such as ceftazidime which also demonstrated a high detection rate. In accordance with other studies, these results indicate the need to screen with a combination of indicator cephalosporins such as ceftazidime and cefpodoxime.
Plasmid-mediated AmpC β-lactamases mask the synergistic effect of the clavulanic acid and the cephalosporin against ESBLs and may thus lead to false negative ESBL results . In order to detect ESBLs in the presence of AmpC β-lactamases, the use of fourth generation cephalosporins such as cefepime is required.
Screen-positive ESBL isolates must be confirmed using an ESBL confirmatory test and in this study the combination disc test appeared to be the most reliable test compared to the double disc synergy test. Only 30% of E. coli and Klebsiella spp. were confirmed as ESBLs in the double disc synergy test whereas the combination disc test detected 50% of E. coli and 65% of Klebsiella spp. as ESBL producers. However, the combination disc test demonstrated some isolates where the zones of inhibition were enhanced by only 3 and 4mm. As this could be due to an inoculum effect, isolates demonstrating enhanced zones of inhibiton below the recommended 5mm cut off should be re-tested using a lighter inoculum. There was a high level of indeterminate isolates in the double disc synergy test. The distance between the discs may not have been optimal for each individual strain that was tested. It is also possible that inoculum variation may have contributed to the interpretation problem.
can be difficult to interpret. Although the test is low cost and easy to perform the problems of interpretation make it unsuitable for use in the routine laboratory. More reliable confirmatory tests such as the combination disc test, although more costly should be the first choice for ESBL detection.
Other commercial ESBL detect systems such as the E-test is also available. Although this system is reliable and easy to perform, the high cost may prohibit use in the routine laboratory. As well as low cost and reliable detection methods for ESBL, rapid detection is desirable in clinical laboratories for successful infection management. The VITEK 2 may represent a good choice instead of manual methods. In a recent study performed by Spanu et al, 2006, it was showed that the VITEK 2 ESBL identification is a reliable, time-saving tool for routine identification of ESBL-producing organisms. The VITEK 2 correctly detected 306 of 312 ESBL-producing organisms with a sensitivity of 98 % and a specificity of 99 %. It delivered results in 6-13 hours, compared to at least 18 hours for the CLSI screening and confirmatory manual methods.
A number of molecular detection methods are available for the detection of ESBLs. The easiest and most common molecular method used to detect the presence of ESBLs is PCR with oligonucleotide primers targeted for TEM, SHV and OXA β-lactamase genes.
In this study singleplex and multiplex PCR assays were performed for detection of TEM, SHV and OXA-type β-lactamases. Even though a standard protocol for the multiplex assay was previously established some problems arose (Colom, 2003). Analysis of the DNA sequence in the internal region of the TEM and SHV β-lactamase gene targeted by the SHV and TEM primers showed very little difference in bp sequence (Figure 2.2 and 2.3). The high sequence homology in the target DNA sequence therefore resulted in cross-reactivity in the TEM and SHV singleplex PCR. However, when the primer sets were combined in the multiplex assay the individual primers appeared to select for their specific target sequence and no cross-reactive products were observed.
It was also shown that the negative control strain ATCC 25922 E. coli demonstrated a weak positive reaction in the TEM PCR. This “false” positive reaction may have resulted from amplification of a generic β-lactamase-type gene originally involved in cell wall structure and function rather than β-lactamase-mediated resistance. This negative control was replaced with a clinical isolate that was also sensitive for ampicillin, which was negative in the TEM PCR.
of the susceptible isolates of E. coli tested were positive in the TEM PCR. In order to identify if an ESBL is present in a clinical isolate nucleotide sequencing must thus be performed. However, the high cost and high technical expertise required for nucleotide sequencing outrules this option for most routine laboratories. Nevertheless the use of such technology is necessary to confirm ESBL production when phenotypic tests prove indeterminate. In addition this technology also facilitates the classification of ESBLs which is essential for epidemiological studies. Classification of ESBLs is of course possible by performing PCR assays where a large number of primers are designed to target each individual variant ESBL. However the design and use of many primers is expensive and time consuming and does not represent a practical approach for ESBL detection in a small laboratory. The provision of antimicrobial resistance reference laboratories where these technologies are in practice is therefore recommended.
5. APPENDIX
Isolate AMP 10 mg PRL AUG 30 mg SXT AK 30 mg CN TE C FOX KF MEM NA 30 mg CIP 5 mg
Isolate AMP 10 mg PRL 100 mg AUG 30 mg SXT 5 mg AK 30 mg CN 10 mg TE 30 mg C 30 mg FOX 30 mg KF 30 mg MEM 10 mg NA 30 mg CIP 5 mg ATM 30 mg CTX 30 mg CPD 30 mg CAZ 30 mg CRO 30 mg 41. E. coli 0mm R 0mm R 17mm I 0mm R 21mm S 20mm S 26mm S 25mm S 26mm S 16mm I 33mm S 27mm S 34mm S 32mm S 32mm S 27mm S 28mm S 30mm S 42. E. coli 0mm R 0mm R 15mm I 0mm R 21mm S 19mm S 26mm S 25mm S 27mm S 16mm I 34mm S 26mm S 36mm S 34mm S 36mm S 30mm S 30mm S 32mm S 43. E. coli 0mm R 0mm R 14mm I 0mm R S 21mm 20mmS 26mmS 27mmS 27mmS 15mm I 32mmS 27mmS 34mmS S 32mm 36mmS 28mmS 30mm S 30mm S 44. E. coli 0mm R 0mm R 13mm R 0mm R 20mm S 19mm S 25mm S 26mm S 27mm S 15mm I 34mm S 26mm S 36mm S 30mm S 32mm S 28mm S 28mm S 30mm S 45. E. coli 0mm R 0mm R 11mm R 0mm R S 20mm 19mmS 0mm R 22mmS 26mmS 9mm R 30mmS 26mmS 32mmS S 30mm 34mmS 26mmS 26mm S 29mm S 46. E. coli 0mm R 0mm R 12mm R 0mm R 22mm S 0mm R 0mm R 18mm S 23mm S 10mm R 34mm S 0mm R 0mm R 33mm S 32mm S 26mm S 30mm S 30mm S 47. E. coli 19mm S 24mm S 21mm S 27mm S 23mm S 22mm S 25mm S 24mm S 28mm S 21mm S 32mm S 26mm S 40mm S 32mm S 32mm S 26mm S 28mm S 32mm S
Table 2. Results from combination disc test
Isolate CAZ CAZ
47. E. coli 26. K. pneumoniae 0 18 18 + 0 23 23 + 0 0 0 - 0 8 8 + 27. E.coli 10 26 16 + 0 29 29 + 0 16 16 + 11 21 10 + 28. E.coli 23 23 0 - 27 28 1 - 10 10 0 - 31 31 0 - 29. K. pneumoniae 25 24 1 - 29 30 1 - 9 10 1 - 36 36 0 - 30. K. pneumoniae 15 26 11 + 9 30 21 + 0 23 23 + 14 29 15 + 31. E. coli 0 10 10 + 13 11 2 - 0 0 0 - 23 23 0 - 32. K. oxytoca 23 23 0 - 21 24 3 - 10 12 2 - 18 18 0 - 33. E.coli 23 28 5 + 20 35 15 + 0 26 26 + 25 32 7 + 34. E. coli 19 30 11 + 12 34 22 + 0 22 22 + 18 29 11 + 35. K. pneumoniae 24 29 5 + 26 35 9 + 14 28 14 + 28 32 4 - 36. K. pneumoniae 12 24 12 + 20 27 7 + 9 21 12 + 21 26 5 + 37. E. coli 29 29 0 - 32 32 0 - 28 29 1 - 32 32 0 - 38. E. coli 29 29 0 - 34 34 0 - 27 27 0 - 32 32 0 - 39. K.pneumoniae 28 28 0 - 32 32 0 - 28 28 0 - 31 31 0 - 40. E. coli 29 29 0 - 35 35 0 - 29 29 0 - 32 32 0 - 41. E. coli 30 30 0 - 34 34 0 - 28 28 0 - 33 33 0 - 42. E. coli 30 30 0 - 35 35 0 - 29 29 0 - 33 33 0 - 43. E. coli 30 30 0 - 36 36 0 - 29 29 0 - 33 33 0 - 44. E. coli 31 31 0 - 35 35 0 - 29 29 0 - 33 33 0 - 45. E. coli 29 29 0 - 34 34 0 - 28 28 0 - 30 30 0 - 46. E. coli 29 29 0 - 34 34 0 - 26 26 0 - 31 31 0 - 30 30 0 - 35 35 0 - 27 27 0 - 34 34 0 -
6. REFERENCES
Al-Jasser AM. Extended-Spectrum Beta-Lactamases (ESBLs): A global problem. Kuwait Medical Journal 2006; 38: 171-185.
Ali Shah A, Hasan F, Ahmed S, Hameed A. Characteristics, epidemiology and clinical importance of emerging strains of gram-negative bacilli producing Extended-Spectrum Beta-Lactamases. Research in Microbiology 2004; 155: 409-421.
Black JG. Antimicrobial therapy: the resistance of microorganisms In Microbiology principles and explorations. Ed Roesch B, Swain E. J Wiley & Sons, New York. 5th edition
2002: 342-343.
Bradford PA. Extended-Spectrum Beta-Lactamases in the 21st century: Characterization, epidemiology, and detection of this important resistance threat. Clinical Microbiology Reviews 2001; 14: 933-951.
Bush K, Jacoby GA, Medeiros AA. A functional classification scheme for beta-lactamases and its correlation with molecular structure. Antimicrobial Agents and Chemotherapy 1995: 1211-1233.
Colodner R, Reznik B, Gal V, Yamazaki H, Hanaki H, Kubo R. Evaluation of a novel kit for the rapid detection of extended-spectrum beta-lactamses. European Journal of Microbiology and Infection diseases 2005.
Colom K, Pérez J, Alonso R, Fernández-Aranguiz A, Lariño E, Cisterna R. Simple and reliable multiplex PCR assay for detection of blaTEM, blaSHV and blaOXA-1 genes in
Enterobacteriaceae. FEMS Microbiology Letters 2003; 223: 147-151.
Essack SY. Laboratory detection of extended-spectrum beta-lactamases (ESBLs) - The need for a reliable, reproducible method. Diagnostic Microbiology and Infection Disease 2000; 37: 293-295.
Jarlier V, Nicolas MH, Fournier G, Philippon A. Extended broad-spectrum beta-lactamases conferring transferable resistance to newer beta-lactam agents in Enterobacteriaceae: Hospital prevalence and susceptibility patterns. Reviews of Infections Diseases 1988; 10: 867-877.
Livermore DM. β-lactamases in laboratory and clinical resistance. Clinical Mircobiology Reviews 1995; 8: 557- 584.
Livermore DM. Beta-lactamase-mediated resistance and opportunities for its control. Journal of Antimicrobial Chemotherapy 1998; 41: 25-41.
