| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Microbiology, Medical University of Sofia, 2 Zdrave Street, 1431 Sofia, Bulgaria
2 Laboratory of Clinical Microbiology, Alexander University Hospital, Medical University of Sofia, 1 Georgi Sofiiski Blvd, 1431 Sofia, Bulgaria
3 Laboratory of Molecular Pathology, University Hospital of Obstetrics and Gynecology, Medical University of Sofia, 2 Zdrave Street, 1431 Sofia, Bulgaria
Correspondence
Tanya Strateva
dr.strateva{at}abv.bg
Received 4 October 2006
Accepted 13 March 2007
Abbreviations: AAC, aminoglycoside acetyltransferase; ANT, aminoglycoside nucleotidyltransferase; ESBL, extended-spectrum ß-lactamase; ICU, intensive care unit; LRTI, lower respiratory tract infection; MBL, metallo-ß-lactamase; URTI, upper respiratory tract infection.
The GenBank/EMBL/DDBJ accession nos for the P. aeruginosa blaVEB-1 and blaPSE-1 gene sequences are DQ333895 and M69058, respectively.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Multidrug-resistant strains of P. aeruginosa (resistant to at least three of the following antimicrobials: ceftazidime, imipenem, gentamicin and ciprofloxacin) are often isolated among patients suffering from nosocomial infections, particularly those receiving intensive care treatment (Tassios et al., 1997). The increasing rate of P. aeruginosa strains in a wide spectrum of clinical settings determines them as emerging pathogens, especially in intensive care units (ICUs), and justifies the necessity for antimicrobial-resistance surveillance.
The aim of this study was to assess the current levels of antimicrobial susceptibility and to evaluate the resistance mechanisms to antipseudomonal antimicrobial agents among problematic clinical isolates of P. aeruginosa collected from five university hospitals in Sofia, Bulgaria.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Antimicrobial-susceptibility testing. The susceptibility of the investigated P. aeruginosa isolates to 17 antimicrobial agents was determined by the disc diffusion method on Mueller–Hinton II agar plates (Becton Dickinson) using antibiotic-containing discs provided by Becton Dickinson, Mast Diagnostics and Bul Bio, and was interpreted according to the National Committee for Clinical Laboratory Standards (NCCLS) (now the Clinical and Laboratory Standards Institute) 2004 recommendations (NCCLS, 2004). Control strains included P. aeruginosa ATCC 27853 and Escherichia coli ATCC 25922.
Phenotypic methods for detection of resistance mechanisms to antimicrobial agents
Detection of group 1 inducible ß-lactamases.
The prevalence of inducible AmpC ß-lactamase (molecular class C, functional group 1) (Bush et al., 1995) in the studied strains of P. aeruginosa was investigated using a disc approximation test method (Sanders & Sanders, 1992). A ceftazidime (30 µg) disc was placed at a distance of 20 mm (centre to centre) from an imipenem (10 µg) disc on a Mueller–Hinton II agar plate inoculated with a suspension of the test organism, adjusted to a McFarland no. 0.5 tube. After overnight incubation, distinct flattening of the inhibitory zone around the ceftazidime-containing disc on the side nearest to the imipenem disc was taken to indicate the presence of inducible AmpC ß-lactamase.
Screening for extended-spectrum ß-lactamases (ESBLs). The presence of ESBLs was investigated by the double disc synergy test (Jarlier et al., 1988). Ceftazidime (30 µg), cefepime (30 µg), cefpirome (30 µg) and aztreonam (30 µg) discs were placed next to an amoxicillin/clavulanic acid (20/10 µg)-containing disc at a distance of 20 mm (centre to centre) on a Mueller–Hinton II agar plate inoculated with the test organism. After overnight incubation at 37 °C, an enhancement of the inhibition zone around at least one of these discs toward the clavulanate-containing disc indicated the presence of ESBLs. All studied strains were tested additionally by a disc diffusion method with imipenem (10 µg) and ceftazidime (30 µg) discs for the presence of synergism (Weldhagen et al., 2003).
Screening for metallo-ß-lactamases (MBLs). The presence of Ambler class B MBLs (Bush et al., 1995) was studied using the modified Hodge test (Lee et al., 2001).
Detection of presumptive aminoglycoside-modifying enzymes. This test was performed according to the substrate profile, as described by the Aminoglycoside Resistance Study Groups (1994).
PCR amplification and sequencing of ß-lactamase genes.
Total DNA from P. aeruginosa isolates was extracted by boiling. The detection of blaVEB-1, blaPER-1, blaPSE-1, blaOXA-groupI, blaOXA-groupII, blaIMP-like and blaVIM-like genes in the investigated strains was performed by PCR with the specific primers (Alpha DNA) listed in Table 1
. PCR was carried out with 2 µl template DNA, 0.25 µM each primer, 0.2 mM deoxyribonucleoside triphosphates, 1x reaction buffer, 2 mM MgCl2 and 1.5 U Prime Taq DNA polymerase (GENET BIO) in a total volume of 25 µl. The DNA was amplified in a Techgen PCR thermocycler (Techne) using the following protocol: initial denaturation (94 °C for 5 min), followed by 30 cycles of denaturation (94 °C for 45 s), annealing (50–64 °C, from 45 s to 1 min) and extension (72 °C, from 45 s to 1 min), with a single final extension of 7 min at 72 °C. PCR products were separated in 1 % agarose gel for 50 min at 150 V, stained with ethidium bromide (0.5 µg ml–1) and detected by a UV transillumination (wavelength 312 nm). The amplified genes were identified on the basis of fragment size (shown in Table 1). Selected VEB-1 and PSE-1 PCR products were purified with ExoSAP-IT reagent (Amersham Biosciences). Sequencing reactions were performed using the same blaVEB-1-and blaPSE-1-specific primers and a BigDye terminator v3.1 kit (Applera) in an automated sequencer (ABI 310 sequence genetic analyser; Applied Biosystems). The nucleotide and deduced amino acid sequences were analysed with software available from the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov).
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. The established antimicrobial resistances, in increasing order, were to: polymyxin B<imipenem<meropenem<ceftazidime<cefepime<aztreonam<piperacillin/tazobactam. Polymyxin B remained active against all isolates. The antimicrobial resistance to antibiotics of the investigated problematic strains of P. aeruginosa was higher than the mean P. aeruginosa resistance found in Bulgaria in 2003, according to data from the national program BulSTAR: 45.8 vs 24.5 % to ceftazidime, 42.3 vs 8.3 % to imipenem, 59.1 vs 24.9 % to amikacin, 79.7 vs 38.7 % to gentamicin and 80.3 vs 30.7 % to ciprofloxacin (Petrov et al., 2005). Approximately half of our isolates (49.8 %) were multidrug resistant.
|
. The strains isolated during 2004–2006 were significantly more resistant (P<0.001) than those obtained during 2001–2003 to the following antibiotics: ceftazidime (79.5 vs 20.0 %), cefepime (75.0 vs 26.9 %), cefpirome (81.6 vs 7.6 %), aztreonam (73.9 vs 31.0 %), imipenem (59.1 vs 31.3 %), meropenem (62.5 vs 27.8 %) and amikacin (72.7 vs 47.8 %). Thus, these results demonstrated increasing resistance rates to extended-spectrum antipseudomonal cephalosporins, carbapenems, monobactams and amikacin in P. aeruginosa isolates from the university hospitals in Sofia during the last 3 years.
|
). The ICU isolates were significantly more resistant to meropenem (61.4 %) than all investigated isolates as a whole (45.5 %, P<0.05), which is related to the widespread use of meropenem for the treatment of life-threatening infections in ICUs.
The antimicrobial resistance in P. aeruginosa varied among different clinical specimens (Table 2). The P. aeruginosa isolates from in-patients with lower respiratory tract infections (LRTIs) were more resistant to piperacillin than the isolates obtained from wounds and drainages (P<0.05). The strains of P. aeruginosa isolated from in-patients with LRTIs and upper respiratory tract infections (URTIs) were more resistant to piperacillin/tazobactam than those from urine and wounds (P<0.01 and P<0.001, respectively). The observed resistance rate to ceftazidime in P. aeruginosa from wounds and drainages was lower than that in URTI isolates (P<0.05). URTI strains of P. aeruginosa were more resistant to cefpirome than LRTI isolates (P<0.05). The antibiotic resistance of P. aeruginosa from urine samples towards imipenem and meropenem was lower than that among URTI strains (P<0.02 and P<0.01, respectively). URTI isolates were significantly more susceptible than the P. aeruginosa isolates from urine, LRTIs, and wounds and drainages to the following antimicrobials: amikacin (P<0.001), gentamicin (P<0.001), tobramycin (P<0.001), netilmicin (P<0.01, P<0.001 and P<0.01, respectively) and ciprofloxacin (P<0.001).
A total of 104 of the 203 investigated problematic P. aeruginosa isolates (51.2 %) showed a ‘penicillinase production phenotype’ (resistance to carboxypenicillins and ureidopenicillins, and susceptibility to ceftazidime) (Bert et al., 2003). These strains were presumptive producers of narrow-spectrum ß-lactamases. Additionally, they expressed an inducible AmpC ß-lactamase (cephalosporinase).
The ceftazidime resistance rate was 45.8 %. A total of 57 of the strains of the 203 studied P. aeruginosa isolates (28.1 %) were resistant to extended-spectrum cephalosporins, including ceftazidime, and were characterized as presumptive producers of ESBLs according to the double disc synergy test. These strains also displayed in vitro synergism between imipenem and ceftazidime, typical of the producers of clavulanic acid- and tazobactam-inhibited ESBLs from molecular class A, such as VEB-, PER- or GES-type (Weldhagen et al., 2003).
Twelve strains of P. aeruginosa (5.9 %) were resistant to all ß-lactams except carbapenems, and showed a negative result in the double disc synergy test. In these isolates, resistance to extended-spectrum cephalosporins was related mainly to the overproduction of a chromosomal AmpC cephalosporinase from molecular class C.
Of the 203 P. aeruginosa isolates, 12 (5.9 %) possessed a OprD– mutant phenotype (resistant only to imipenem and meropenem, but susceptible to other ß-lactams). Carbapenem resistance is mostly due to OprD deficiency and is independent of susceptibility towards other ß-lactam agents (Livermore, 2001). This resistance mechanism demands continued expression of the chromosomal AmpC ß-lactamase (Livermore, 1992).
Overproduction of active efflux systems with wide substrate profiles was the prevailing resistance mechanism in eight P. aeruginosa isolates (3.9 %). In our study, the presumptive efflux systems were: MexA–MexB–OprM associated with decreased susceptibility or resistance to all ß-lactams, except imipenem, and with decreased susceptibility or resistance to quinolones (nalB or nalC mutants) (Llanes et al., 2004), and MexC–MexD–OprJ conferring resistance to fourth-generation cephalosporins (cefepime and cefpirome) and resulting from mutation in nfxB (Poole et al., 1996).
Sixty strains of P. aeruginosa (29.6 %) were resistant to all ß-lactams, including carbapenems, and thus could be related to a phenotype of Ambler class B MBL-producing strains (Nordmann & Poirel, 2002). All carbapenem-resistant strains of P. aeruginosa showed a negative Hodge test and therefore were not producers of MBLs. Most probably, the resistance to ß-lactams resulted from a combination of different mechanisms: OprD deficiency, derepression of chromosomal AmpC cephalosporinase, ESBL production and overexpression of active efflux systems.
One hundred and twenty isolates out of the studied strains of P. aeruginosa (59.1 %) were resistant to amikacin and 69.6–89.6 % to the other aminoglycosides (netilmicin, gentamicin and tobramycin). The most widespread mechanism of resistance to these antimicrobials involves enzymic modification by aminoglycoside acetyltransferases (AACs), aminoglycoside nucleotidyltransferases (ANTs) or aminoglycoside phosphotransferases (Poole, 2005). The prevailing phenotypes of aminoglycoside resistance in our strains were: (i) amikacin+gentamicin+tobramycin+netilmicin (49.9 %), associated with AAC (6')-I±ANT (2''); (ii) gentamicin+tobramycin+netilmicin (14.1 %), associated with AAC (3)-V or ANT (2'')+AAC (3)-Ia; and (iii) amikacin+gentamicin+tobramycin (14.1 %), related to aminoglycoside phosphotransferase (3')-VI+ANT (2'') (Aminoglycoside Resistance Study Group, 1994).
One hundred and sixty three (80.3 %) of the isolates were resistant to ciprofloxacin. The most important mechanisms of quinolone resistance are structural alterations of the primary or secondary targets because of chromosomal point mutations in gyrA/gyrB or parC/parE genes, respectively, followed by an active efflux of these antimicrobial agents (Hooper, 2001).
A molecular genetic study was carried out for the presence of ß-lactamases belonging to different molecular classes. A total of 160 isolates were investigated, and 53 (33.1 %) were found to be VEB-1 producers. The sequence of blaVEB-1 amplified from different selected isolates was identical for all isolates and 100 % identical to the known veb-1 sequence (GenBank accession no. DQ333895). The frequency of VEB-1 ESBLs among the ceftazidime-resistant P. aeruginosa was 57.0 % (53/93). A total of 36 of the 160 isolates (22.5 %) produced PSE-1 enzyme. Selected PSE-1 PCR products showed 100 % identity to blaPSE-1 (GenBank accession no. M69058). The frequency of OXA group I and OXA group II ß-lactamases was 41.3 % (66/160) and 8.8 % (14/160), respectively.
The distribution of the Ambler class A and D ß-lactamases among the investigated strains of P. aeruginosa is presented in Table 3. As shown, the relative proportion of ß-lactamase-producing strains of P. aeruginosa (66.8 %) was higher than the proportion of ß-lactamase-non-producing strains (33.1 %). An analogous study carried out recently in Korea established that ß-lactamase-non-producing strains of P. aeruginosa were more widespread than producers (74.6 vs 25.4 %; Lee et al., 2005).
|
A high frequency of distribution of ESBLs in the ceftazidime-resistant isolates of P. aeruginosa was established in our study. In all the university hospitals monitored in Sofia, widespread dissemination of blaVEB-1 in clinical isolates of P. aeruginosa was found. Recently, Bachvarova et al. (2005) reported a significantly lower (P<0.01) rate of prevalence of VEB-1-type ß-lactamases among ceftazidime-resistant strains of P. aeruginosa obtained from distinct regions of Bulgaria during 1998–2003 than that determined in our study (36.8 vs 57.0 %). Thus, the observed trend towards an increasing rate of VEB-1-producing P. aeruginosa strains in Bulgaria relates to the last 2 years. VEB-1 and VEB-1-like enzymes are widespread in Asia (Thailand, Kuwait, India and China) (Weldhagen et al., 2003; Girlich et al., 2002; Poirel et al., 2001), but in European countries have been detected only in France (Naas et al., 1999).
A total of 42 (26.3 %) of the 160 isolates studied possessed both VEB-1 and OXA group I enzymes. It is likely that the strains produced the narrow-spectrum OXA-10 from OXA group I (Sanschagrin et al., 1995), encoded by a gene located on class 1 integron In50, as well as blaVEB-1 (Girlich et al., 2002). Lee et al. (2005) reported that OXA-10 was the most prevalent enzyme (13.5 %) in Korea in 2002.
PSE-1 ß-lactamases belonging to Ambler class A and functional group 2c (Bush et al., 1995) were detected in 22.5 % of the investigated strains. In 2002, Nordmann (2002) reported 11 % CARB-producing strains of P. aeruginosa in France, 90 % of which were PSE-1.
Our study did not reveal blaPER-1, in contrast to the widespread detection of these genes in Europe. Epidemics caused by PER-1-producing P. aeruginosa have been reported previously in Turkey and Italy (Vahaboglu et al., 1997; Luzzaro et al., 2001). Isolates of P. aeruginosa with PER-1 enzymes have also been observed in France, Belgium and Poland (De Champs et al., 2002; Claeys et al., 2000; Empel et al., 2005).
The established frequency of OXA group II ß-lactamases comprising OXA-2, -3, -15 and -20 (Sanschagrin et al., 1995) was the lowest in our research. It is likely that the oxacillinases from group II were predominantly narrow spectrum, such as OXA-2 or -3, as these enzymes were detected mainly in ceftazidime-susceptible strains of P. aeruginosa. In comparison, OXA group II enzymes were disseminated among 2.3 % of the studied P. aeruginosa isolates in Korea and all strains were determined as OXA-2 producers (Lee et al., 2005). The frequency of blaOXA-groupII in our strains of P. aeruginosa (8.8 %) was similar to the dissemination rate of OXA group II ß-lactamases in France during 1994–1999 (9.9 %) (Bert et al., 2002).
Carbapenem-hydrolysing IMP- and VIM-type metalloenzymes belonging to Ambler class B were not detected in this study. The investigated carbapenem-resistant strains of P. aeruginosa from Sofia did not harbour blaVIM-like genes, although the detection of these genes is widespread, especially in neighbouring countries such as Greece and Turkey (Tsakris et al., 2000; Mavroidi et al., 2000; Pournaras et al., 2002; Bahar et al., 2004). The carbapenem resistance was related to non-enzymic mechanisms such as OprD deficiency and active efflux.
The comparative antimicrobial resistances of ß-lactamase-producing and ß-lactamase-non-producing P. aeruginosa are summarized in Table 4. The ß-lactamase producers were significantly more resistant than non-producers to ceftazidime, cefepime, cefpirome, aztreonam, amikacin, tobramycin and ciprofloxacin (P<0.001) and to gentamicin (P<0.01). The susceptibilities to extended-spectrum cephalosporins, aztreonam and carbapenems among ß-lactamase-producing strains of P. aeruginosa were lower than those in non-ß-lactamase producers and were similar to the susceptibilities in analogous P. aeruginosa isolates from Korea in 2002 (Lee et al., 2005). Moreover, in our study and the Korean study, the cross-class resistance to aminoglycosides and ciprofloxacin was significantly higher in class A and D ß-lactamase-producing P. aeruginosa. As described previously, VEB-1 was the first class A enzyme found to be encoded by an integron-located gene cassette (Poirel et al., 1999). In the blaVEB-1-containing integrons of P. aeruginosa, the veb-1 cassette is often associated with aminoglycoside resistance gene cassettes (Girlich et al., 2002).
|
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
, 2001–2003 (n=115); 
, 2004–2006 (n=88). (a) Resistance to ß-lactams. CAR, Carbenicillin; AZL, azlocillin; PIP, piperacillin; TZP, piperacillin+tazobactam; CAZ, ceftazidime; CPZ, cefoperazone; FEP, cefepime; CPO, cefpirome; ATM, aztreonam; IMP, imipenem; MEM, meropenem. (b) Resistance to aminoglycosides and fluoroquinolones. AMK, Amikacin; GEN, gentamicin; TOB, tobramycin; NET, netilmicin; CIP, ciprofloxacin.