Previous Article | Next Article ![]()
Journal of Clinical Microbiology, May 2009, p. 1436-1442, Vol. 47, No. 5
0095-1137/09/$08.00+0 doi:10.1128/JCM.02380-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio,1 Research Service, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio,2 Departments of Pharmacology,3 Genetics, Case Western Reserve University School of Medicine, Cleveland, Ohio,4 Departments of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia,5 Ibis Biosciences, Carlsbad, California,6 Molecular Biology and Microbiology, Case Western Reserve University School of Medicine, Cleveland, Ohio7
Received 11 December 2008/ Returned for modification 3 February 2009/ Accepted 11 March 2009
|
|
|---|
|
|
|---|
An unwelcome feature of A. baumannii is the frequency in which isolates manifest resistance to the most commonly used antibiotics (12, 14). When resistance to three or more classes of antibiotics is demonstrated, A. baumannii is referred to as multidrug resistant (MDR) (21). Among MDR A. baumannii isolates, resistance to quinolones (e.g., ciprofloxacin and levofloxacin), aminoglycosides (e.g., gentamicin and amikacin), ampicillin-sulbactam, extended-spectrum cephalosporins (e.g., ceftazidime), and carbapenems (e.g., imipenem, meropenem, and doripenem) is escalating (21, 23). Concern exists that, if the incorrect antimicrobial agents are chosen to treat MDR A. baumannii infection, the outcome of patients may be poor (9, 22). Thus, a rapid assessment of antimicrobial susceptibility could have a significant impact on patient care.
In this report, we employ a rapid high-throughput method to identify unique DNA changes associated with quinolone resistance in a collection of MDR A. baumannii isolates. The method was originally developed to identify and genotype various species of bacteria from complex mixtures in respiratory samples (6). This process uses electrospray ionization mass spectrometry (ESI-MS) and base composition analysis of PCR amplification products derived from highly conserved genes ("housekeeping genes"). The primers that are designed for PCR/ESI-MS yield amplicons with unique mass signatures that can be measured by high-performance mass spectrometry and identified by base composition analysis (7, 13).
The best-described mechanism of resistance to quinolones in Acinetobacter spp. is mutations in the genes encoding DNA gyrase A (i.e., gyrA) and subunit A of topoisomerase IV (i.e., parC). The most important mutations, resulting in changes at codon 83 for gyrA and at codon 80 for parC, have been mapped to a unique location in each of these genes, the quinolone resistance-determining region (QRDR) (27, 28). Additional mutations in the QRDR of gyrA thought to enhance quinolone resistance include mutations at codons for amino acids Gly81, Ala84, and Glu87 (1, 11, 17, 27, 28, 30). Another previously described mutation (at the Val101 codon) does not appear to have an effect on the susceptibility profiles (1). Ancillary mutations in parC, at codons for amino acids Gly78 and Glu84, when combined with the Ser83 and Ser80 codon mutations, also contribute to high-level quinolone resistance (11, 27, 30).
Appreciating that mutations in the housekeeping genes gyrA and parC also define the quinolone resistance phenotype, we reasoned that PCR/ESI-MS and base composition analysis can be applied to determine the sequence variability of the QRDRs of Acinetobacter spp. and rapidly predict a quinolone-resistant phenotype. Therefore, we chose to study 73 well-characterized clinical isolates of A. baumannii, one Acinetobacter genome species strain 3 isolate, and one Acinetobacter johnsonii isolate. This paper describes the novel application of this highly sensitive and specific method to the analysis of resistance phenotypes in this clinically important pathogen.
|
|
|---|
Determination of susceptibility to quinolones.
MICs of ciprofloxacin and levofloxacin were determined by the broth microdilution method using cation-adjusted Mueller-Hinton medium according to the Clinical and Laboratory Standards Institute (CLSI) standard criteria (3). Specific panels were "custom-made" by Trek Diagnostics (Cleveland, OH) to better measure the level of quinolone resistance. Ciprofloxacin was tested in the range of 0.06 to 64 mg/liter, whereas levofloxacin was tested in the range of 0.06 to 32 mg/liter. ATCC control strains, including A. baumannii isolates 9955 and 17961, Escherichia coli 25922, Pseudomonas aeruginosa 27853, and Klebsiella pneumoniae 700603, were used. Susceptibility results were interpreted according to the guidelines recommended by CLSI (4). In this paper, A. baumannii isolates were defined as quinolone resistant when the ciprofloxacin or levofloxacin MIC or both MICs were in the nonsusceptible range (MIC
1 mg/liter).
PCR amplification for RE digestion and DNA sequencing. All primers used for PCR amplification along with the product sizes are listed in Table 1. The restriction endonuclease (RE) digestions with HinfI of the gyrA and parC QRDRs have been previously described (14).
|
View this table: [in a new window] |
TABLE 1. Primers used to amplify the gyrA and parC genes for sequencing and PCR/ESI-MS
|
DNA preparation and PCR amplification for PCR/ESI-MS. Template DNA was prepared by making a 1:100 dilution of a lysogeny broth overnight culture in TE buffer (10 mM Tris, 1 mM EDTA [pH 8.0]) and boiling for 15 min. All PCRs were performed in 96-well plates. The PCR mixture consisted of 3 U of AmpliTaq Gold polymerase (Applied Biosystems, Foster City, CA); 20 mM Tris (pH 8.3); 75 mM KCl; 1.5 mM MgCl2; 0.4 M betaine; 800 µM equal mixture of dCTP, dTTP, dGTP, and dATP; and 250 nM of each primer. PCR cycling under the following conditions was performed on an MJ Dyad 96-well thermocycler (Bio-Rad Inc., Hercules, CA): 95°C for 10 min, followed by 8 cycles of 95°C for 30 s, 48°C for 30 s, and 72°C 30 s, with the 48°C annealing temperature increasing 0.9°C each cycle. PCR was then continued for 37 additional cycles of 95°C for 15 s, 56°C for 20 s, and 72°C for 20 s. The PCR cycle ended with a final extension of 2 min at 72°C followed by a 4°C hold. Five microliters of the template DNA was used in each PCR.
Primer design for PCR/ESI-MS and base composition analysis. The primer pairs used for mass spectrometry and base composition analysis of the QRDRs of gyrA and parC in Acinetobacter spp. are also listed in Table 1. To completely cover the QRDR, three sets of overlapping primers were constructed for both topoisomerase genes. Two sets of primers encompass the flanking regions as well as the nucleotides in parC that encode amino acid Ser80 or those in gyrA that encode Ser83 (Fig. 1). The third set of primers specifically targeted only the nucleotides that encode the Ser83 or Ser80 amino acid. The length of each amplified region is listed in Table 1.
|
View larger version (18K): [in a new window] |
FIG. 1. Priming strategy used for PCR/ESI-MS of the QRDRs of gyrA and parC in Acinetobacter baumannii. Six primer pairs (BCT2852 to BCT2854 and BCT2846 to BCT2848) were developed to cover the entire length of the gyrA and parC QRDRs, respectively. The principal mutations captured by these primer pairs are indicated below the wild-type sequences. Colored boxes indicate the scope of these primer pairs, all of them covering the structurally homologous gyrA codon for Ser83 and parC codon for Ser80.
|
M > 100,000 full-width half-maximal) mass spectra were acquired for each sample by using high-throughput ESI-MS protocols described previously (13). For each sample, 1.5 µl of analyte solution was consumed during the 74-s spectral acquisition. Raw mass spectra were converted to monoisotopic molecular masses. Unambiguous base compositions were derived from the exact mass measurements of the complementary single-stranded oligonucleotides (18). Quantitative results were obtained by comparing the peak heights with an internal PCR calibration standard present in every PCR well at 100 molecules (13). The ESI-MS measurements required approximately 45 s per PCR, so analysis of each Acinetobacter sp. isolate required less than 5 min of mass spectrometry time (338 min for 75 samples). |
|
|---|
2 mg/liter). High-level ciprofloxacin resistance (i.e.,
64 mg/liter) was seen in 74.7% of the isolates. Similarly, 84.9% (62/73) of the isolates were resistant (i.e., MIC
4 mg/liter) to levofloxacin. |
View this table: [in a new window] |
TABLE 2. Quinolone MICs of the 75 Acinetobacter sp. isolates in this study
|
We further determined that isolates AB0056, AB0057, and AB0059 possess silent mutations in the codons of parC encoding Pro98, Leu99, Glu101, Gly102, and Gln103 (the importance of these mutations is still unknown). Comparing our results with the reference sequence, we discovered that silent mutations were found in gyrA for isolate AG0073 (Acinetobacter genome sp. strain 3) and AJ0075 (A. johnsonii) at codons for amino acids Ala67, Val69, Pro79, Val90, Gln94, and Val103.
PCR/ESI-MS and base composition analysis. In Table 3, we present the target base composition analysis for all PCR/ESI-MS primer sets resulting from the generated monoisotopic masses for each forward and reverse PCR amplicon strand. Table 4 lists the ESI-MS data for both the gyrA and parC primer sets for each Acinetobacter sp. isolate analyzed. For ease of interpretation, the various base compositions detected by ESI-MS are color coded and correspond to substitutions from the wild-type sequence (Table 3). In every case, when base substitutions were detected using PCR/ESI-MS, standard DNA sequencing confirmed these findings (100% correlation). In addition, MIC determination identified 67 out of 75 (89.3%) strains as quinolone resistant. By PCR/ESI-MS 88% of the isolates are predicted to be quinolone resistant, and overall the accurate prediction of quinolone susceptibility or resistance was 98.7% (74/75 isolates).
|
View this table: [in a new window] |
TABLE 3. Target base composition analysis for all PCR/ESI-MS primer sets based on the generated monoisotopic mass for each forward and reverse PCR amplicon strand
|
|
View this table: [in a new window] |
TABLE 4. Summary for PCR/ESI-MS base composition analysis for gyrA and parC QRDR ampliconsa
|
Primer set BCT2853 amplifies the nucleotide sequence of the gyrA QRDR that encodes amino acids 83 to 104. This primer set also yielded three distinct mass signature profiles. As with primer set BCT2852, the wild-type signature (turquoise), the second distinct signature representing the Ser83Leu mutation (gold), and the third signature (green) represent sequence variations that correspond to changes in codons for amino acids Val90, Gln94, and Val103 in Acinetobacter genome sp. strain 3 and A. johnsonii.
Primer set BCT2854 specifically targets the gyrA-encoded Ser83 mutation and allows us to discern with confidence that there is a C
T change at the codon for amino acid 83, not in the surrounding sequence, and that this encodes an amino acid change from Ser to Leu (i.e., from TCA
TTA).
Like the complementary primer set in gyrA, primer set BCT2846 corresponds to the 5' flanking region, including the Ser80 codon in parC (Fig. 1). Five distinct mass signature profiles were determined and confirmed by DNA sequencing. As explained for gyrA, the first profile is the expected wild-type sequence (i.e., A33G26C29T33; turquoise) in Tables 3 and 4. The second base composition (i.e., A33G26C28T34; gold) indicates a base change from C to T at the codon for amino acid 80. The third (i.e., A29G26C27T39; green) and fifth (i.e., A32G28C26T35; sky blue) mass signatures reflect sequence variations or polymorphisms in the parC gene of Acinetobacter genome sp. strain 3 and A. johnsonii, respectively (Fig. 1). The fourth unique mass signature (pale blue) (Table 3) reflects the Glu84Lys mutation in isolates AB0042, AB0048, and AB0058. This G
A nucleotide change was uncovered by the base composition analysis and confirmed by DNA sequencing (i.e., A34G25C29T33; Table 4).
Isolates AB0056, AB0057, and AB0059 have silent mutations in the nucleotides encoding amino acids Pro98, Leu99, Glu101, Gly102, and Gln103 in parC, which were also identified by PCR/ESI-MS using primer set BCT2847 (i.e., A29G29C23T33; orange in Tables 3 and 4). This primer set encompasses the 3' flanking region of the codon for the Ser80 amino acid (Table 4). Six unique mass profiles were obtained for the amplicons generated by primer set BCT2847. As with gyrA, a specific primer set for parC, BCT2848, targets only the nucleotides that encode amino acid Ser80, and therefore the wild-type and mutated genes can be identified.
|
|
|---|
PCR/ESI-MS accurately identified quinolone-resistant isolates. PCR/ESI-MS predicted that 88% of the isolates would be resistant to one or both of the quinolones, whereas MIC determinations identified 89.3% of the isolates as quinolone resistant. Overall, the PCR/ESI-MS method identified the correct phenotype in 74/75 (98.7%) isolates. Although it is possible that upregulation of efflux pumps such as AdeABC or AbeM may also contribute to quinolone resistance in these isolates, when mutations were found in both the gyrA and parC QRDRs by PCR/ESI-MS, high-level ciprofloxacin resistance was also detected by traditional susceptibility testing.
In the case of A. baumannii isolates possessing the GyrA Ser83 mutation but not the corresponding ParC Ser80 mutation (i.e., isolates AB0041, AB0042, AB0043, AB0044, AB0045, AB0046, AB0047, AB0048, AB0049, AB0050, and AB0058), the MICs show that the gyrA-encoded Ser83 mutation is enough to confer quinolone resistance in these isolates. Although levofloxacin is thought to better target ParC than ciprofloxacin, only one isolate in this group (i.e., AB0041) remained susceptible to levofloxacin when the primary Ser83Leu mutation was encoded by gyrA.
We also demonstrate that PCR/ESI-MS accurately identified the correct phenotype in the quinolone-susceptible isolates (i.e., isolates AB0051, AB0064, AB0068, AB0071, AB0072, AG0073, AB0074, and AJ075). For only one isolate, AB0069, PCR/ESI-MS and phenotypic tests were not in agreement. The MICs show a reduced susceptibility to quinolones (intermediate for ciprofloxacin, susceptible to levofloxacin), although, by DNA sequencing, RE digestion, and PCR/ESI-MS, AB0069 was predicted to have a susceptible phenotype.
In conclusion, we report the discriminating ability of PCR/ESI-MS for detecting quinolone resistance by targeting the QRDRs in 75 isolates of Acinetobacter spp. Furthermore, our results show that the observed variations in base composition of gyrA and parC can easily be detected even among isolates of related species. Although the informational content provided by base composition is less than that generated by sequencing large regions or entire genes, PCR/ESI-MS has the same practical utility in detecting resistance mutations when the amplified target sites are "information rich" and properly chosen (7).
Two important limitations of PCR/ESI-MS must be recognized. First, it remains uncertain if detecting a resistance gene always indicates that a resistance phenotype will be present. Second, the interpretation of mutations in resistance genes that exist in multiple copies is potentially problematic. The levels of clinical accuracy and predictability remain a significant challenge, as gene dosage, level of transcription, and expression are not yet measured by these methods. This further refinement will require other types of studies. Nevertheless, assessment of quinolone susceptibility by a rapid molecular method (in this case, less than 6 h) could have a significant impact on clinical outcome by aiding in the proper selection of antibiotics to treat infections.
This series of experiments served as a "proof of principle" for the application of PCR/ESI-MS to detect and evaluate resistance genes not only in A. baumannii but also in other gram-negative pathogens. Using this platform for the rapid determination of resistance phenotypes in hospital-associated infections, biodefense, and molecular epidemiology can have a profound impact on clinical microbiology.
Published ahead of print on 18 March 2009. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»