Rapid detection of gyrA and parC mutations in fluoroquinolone-resistant Neisseria gonorrhoeae by denaturing high-performance liquid chromatography

Rapid detection of gyrA and parC mutations in fluoroquinolone-resistant Neisseria gonorrhoeae by denaturing high-performance liquid chromatography

Journal of Microbiological Methods 59 (2004) 415 – 421 www.elsevier.com/locate/jmicmeth Rapid detection of gyrA and parC mutations in fluoroquinolone...

173KB Sizes 0 Downloads 8 Views

Recommend Documents

No documents
Journal of Microbiological Methods 59 (2004) 415 – 421 www.elsevier.com/locate/jmicmeth

Rapid detection of gyrA and parC mutations in fluoroquinolone-resistant Neisseria gonorrhoeae by denaturing high-performance liquid chromatography Katsumi Shigemuraa, Toshiro Shirakawab, Hiroshi Okadac, Kazushi Tanakaa, Tohru Udakad, Sadao Kamidonoa, Soichi Arakawaa, Akinobu Gotohb,* a

Division of Urology, Department of Organs Therapeutics, Faculty of Medicine, Kobe University Graduate School of Medicine, Japan b International Center for Medical Research, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-Cho, Chuo-Ku, Kobe 650-0017, Japan c Division of Urology, Teikyo University School of Medicine, Japan d Research Laboratory, Transgenomic Inc., Tokyo, Japan Received 5 June 2004; received in revised form 16 August 2004; accepted 19 August 2004 Available online 21 September 2004

Abstract The detection of DNA sequence variation is fundamental to the identification of the genomic basis of phenotypic variability. Denaturing high-performance liquid chromatography (DHPLC) is a novel technique that is used to detect mutations in human DNA. This is the first report that this technique is used as a tool to detect mutations in genes encoding fluoroquinolone resistance in Neisseria gonorrhoeae. Eighty-one strains of N. gonorrhoeae were used in this study. Genomic DNA from each strain was subjected to PCR amplification of 225 bp in gyrA and 166 bp in parC spanning the fluoroquinolone-resistance determining regions (QRDRs). After we performed DNA sequencing of these amplicons and identification of mutations in the QRDRs, DHPLC was undertaken to investigate whether its results correlate the distinctive chromatogram with their DNA mutations pattern. The profilings detected by DHPLC completely corresponded to the results of the DNA sequencing in mutation patters in gyrA and parC genes. They resulted in the following amino acid substitutions: Ser-91Phe, Asp-95Gly, and Asp-95Asn in gyrA; and Gly-85Asp, Asp-86Asn, Ser-87Arg, and Ser-88Pro in parC, respectively. These mutations existed alone or as combinations, and we identified five mutations patterns in gyrA and six in parC including wild-type. These mutations and their patterns could be rapidly and reproducibly identified from the PCR products using DHPLC, producing specific peak patterns that correlate with genotypes. This novel detection system facilitates the detection of resistance

* Corresponding author. Tel.: +81 78 382 5694; fax: +81 78 382 5693. E-mail address: [email protected] (A. Gotoh). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.08.004

416

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421

alleles, providing a rapid (5 min per sample), economic (96 sample per run), and reliable technique for characterizing fluoroquinolone resistance in N. gonorrhoeae. D 2004 Elsevier B.V. All rights reserved. Keywords: DHPLC; Neisseria gonorrhoeae; Fluoroquinolone resistance

1. Introduction Fluoroquinolones have shown excellent antiNeisseria gonorrhoeae activity. However, for nearly 10 years, in Japan, we have had difficulty in treating N. gonorrhoeae urethritis with fluoroquinolones which show increasing resistance to fluoroquinolones. One study reports a resistance rate of nearly 70% (Taylor et al.). It is necessary to diagnose the fluoroquinolone resistance of N. gonorrhoeae rapidly and accurately to determine the best clinical treatment. Several amino acid mutations within short fluoroquinolone-resistance determining regions (QRDRs) of DNA gyrase (gyrA) and topoisomerases (parC) can confer clinical resistance (Hannachi-M’Zali et al., 2002). The most reliable method of identifying mutations associated with drug resistance is a DNA sequence analysis of PCR products that contain regions where these mutations may be found. There are differences between the electrophoretic variations of heteroduplex analysis such as constant– gradient gel electrophoresis, denaturing–gradient gel electrophoresis, and temperature–gradient gel electrophoresis. They result in melting points of duplex DNA strands associated with differences in their nucleotide base composition that in turn affect their mobilities in gels (Kristensen et al., 2001). Denaturing high-performance liquid chromatography (DHPLC) is used to compare mixtures of PCR amplicons for polymorphisms by the differential retention of homo- and heteroduplex DNA on a reverse-phase chromatography support under partial heat denaturation conditions (Oefner and Underhill, 1998). Transgenomic (Omaha, Nebraska, USA) has adapted a special-purpose DNA binding column to reversephase HPLC for the analysis and purification of nucleic acids. In the present study, we evaluated the Transgenomic DHPLC system (WAVE) as a rapid screening

and identification method for variations of DNA mutations in the gyrA and parC genes that are associated with fluoroquinolone resistance in N. gonorrhoeae. In this study, we performed the DHPLC analysis of the QRDR of gyrA and parC gene of N. gonorrhoeae after amino acid sequencing and found that the DHPLC system allowed us to detect and identify completely the mutations in these genes, which are related significantly to fluoroquinolone-resistant N. gonorrhoeae strains. This rapid method will allow us to prevent the useless treatment and the increase of resistant strains to many other kinds of antimicrobial agents.

2. Materials and methods 2.1. Bacterial strains and susceptibility testing Gonococci strains isolated from 81 male patients with urethritis, who consulted urology clinics in Hyogo, Japan in 2002, were referred to the urology department laboratory of Kobe University Hospital. The isolates were maintained in 10% skim milk at 80 8C for long-term storage. Quantitative tests of susceptibility to ciprofloxacin were performed according to our previous method (Shigemura et al., 2004). The ciprofloxacin susceptibilities and MIC ranges for strains were categorized by the NCCLS guideline (NCCLS, 2002). 2.2. DNA extraction and PCR DNA extraction was performed using a Qlamp DNA Extraction Kit (QIAGEN, Tokyo, Japan). PCR amplification was performed according to our previous method. PCR primers were also used according to our previous ones (Shigemura et al., 2004).

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421

2.3. DNA sequencing and computer analysis The PCR products were sequenced directly using 0.5 Al (25 ng) of template DNA, 0.75 Al (7.5 pmol) of primer, and 2Al of dye terminator premix (Amersham Bioscience, Piscataway, NJ, USA) in 16.75 Al of distilled water. Following thermal cycling, used sequencer and database search methods were performed according to our previous method (Shigemura et al., 2004). 2.4. Primer selection and PCR conditions for DHPLC We selected the same primers for PCR amplification of genes as to the forward primer in QRDR of gyrA and parC, as described above. However, in gyrA, we added 20-nucleotide guanine- and cytosine(GC)-containing linker (tb-15) on the 5V terminus of the reverse primer, 5V–CGGCCCGCCGCCCCCGCCCCATTTCGGTATAGCGCATGGCTG–3, and in parC, we chose a different primer, 5V–AACGCACCATCGCCTCATAGG–3V, for the purpose of excluding the 104- or 131-silent mutation seen in our sequencing results from PCR range. Because these silent mutations are not amino acid mutations, they did not affect the activities of antimicrobial agents of strains. The preparations of the PCR products and the conditions for the reaction were the same as those described above. 2.5. DHPLC analysis We performed DHPLC analysis of all 81 strains. The PCR products were mixed with an equal amount of DNA which was amplified from a corresponding wild-type reference sample. The mixtures were heated to 95 8C for 5 min and then cooled gradually to 25 8C at a rate of 1.5 8C/min to form hetero- and homoduplex molecules. The heteroduplexed PCR products were analyzed using the WAVE system with a DNA sep HT cartridge (Transgenomic Japan, Tokyo, Japan). The buffers used for DHPLC were 0.1 M triethylammonium acetate, pH 7.0 (buffer A), and 0.1 M triethylammonium acetate in 25% acetonitrile (buffer B). The DNA fragment was eluted at a flow rate of 0.9 ml/min with a gradient of buffers A and B, and it was detected spectrophotometrically by UV absorption at 260 nm. The gradient condition and

417

DHPLC analysis temperatures were determined by use of the Transgenomic WaveMaker utility software. The software predicts the optimal temperature for resolving heteroduplex and homoduplex fragments based on the wild-type amplicon sequence. DHPLC analysis was performed at the predicted temperature of 64.2 8C for parC. However, for gyrA, the analysis temperature was optimized to 65.1 8C which was specific to the known mutation site, although the predicted average melting temperature over the whole amplicon was 63.6 8C. The chromatographic peaks were compared with the wild-type homoduplex peak, which was generated by reannealing the wild-type reference products with either no additional PCR product or other products known to be wild-type (Xiao and Oefner, 2001).

3. Results 3.1. Mutation detection by direct DNA sequencing analysis We found 77 amino acid mutations in gyrA and 59 in parC. The strains contained each type of single or multiple combinations of gyrA or parC mutations and all these sequencing patterns are shown in Table 1. The most common alternations in the gyrA QRDR resulted in serine exchanged for the phenylalanine residue at position 91 (Ser91-to-Phe). The other common alternations were at position 95: aspartic acid exchanged for an asparagine or glycine (Asp95to-Asn or Asp95-to-Gly). Mutations at this position were usually accompanied by the presence of mutations at position 91. ParC point mutations encoded amino acid changes for glycine to aspartic acid at position 85 (Gly85-to-Asp), for aspartic acid to asparagine at position 86 (Asp86-to-Asn), for serine to arginine at position 87 (Ser87-to-Arg), and for serine to proline at position 88 (Ser88-to-Pro). In these mutations, serine to arginine at position 87 was the most common (Table 1). We classified these mutations into five amino acid mutation patterns in gyrA and six patterns in parC, including the wild-type, and we performed all 81 samples as representative examples for the DHPLC study. Table 2 shows their ciprofloxacin MIC ranges in every mutation pattern including wild-type.

418

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421

Table 1 Amino acid mutations of gyrA and parC gene of N. gonorrhoeae Strain No.

Amino acid mutation gryA Ser91

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe Phe

Strain No. parC

Asp95

Gly85

Asp86

Asn

gryA Ser87 Arg

Ser88 Pro

Asn Gly Gly Gly Gly Gly Asn Asn

Arg Arg Arg Arg Arg

Asn Gly Gly Gly Asn Gly Gly Asn

Gly Gly Gly Gly Asn Gly Gly Gly Asn Gly Gly Gly Gly Gly

Pro

Arg Arg Arg Arg Arg Arg Asn

Gly Gly Gly Asn Gly

Pro Arg Arg Arg Arg Arg

Asp

Amino acid mutation

Pro

Arg Arg Arg Arg Arg Arg Arg Arg Arg Arg Arg Arg Arg

3.2. Mutation detection by DHPLC analysis We considered the DHPLC conditions to be optimal when wild-type homoduplexes eluted as a single peak and all mixtures of mutant and wild-type amplicons produced patterns that were clearly distinguishable from the reference patterns. Elution times

Ser91 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

parC Asp95

Phe Gly Phe Gly no mutation Phe Phe Gly Phe Gly Phe Gly Phe Phe Gly Phe Gly Phe Gly Phe Gly Phe Phe Gly Phe no mutation Phe Gly Phe Gly Phe Gly Phe Phe Gly no mutation Phe Gly

Gly85

Asp86

Ser87

Ser88

Arg

Arg Arg Arg Arg Arg Arg Arg

Pro

Asn Arg

Arg Arg Arg

Asn Pro

Phe Phe Phe Phe Phe

Asn Gly Gly Gly Gly Asn

Phe Phe Asn Phe Phe Asn no mutation Phe Phe Asn Phe Asn Phe Gly Phe Gly

Arg Arg Arg Arg Arg Asn Asn Asn Asn Arg Arg

Pro Pro

Arg Arg

for the diagnostic peaks ranged from 3 to 7.5 min. Mixtures of mutant and wild-type amplicons (i.e., mixtures of homoduplexes and heteroduplexes) eluted as two or more peaks (Figs. 1 and 2). DHPLC detected some different profiles including the wildtype in gyrA and parC. These amplicons (gyrA/parC) cannot be combined in a single run because of the

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421 Table 2 DHPLC profiling patterns and their amino acid mutations, and ciprofloxacin MICs of each patterns in gyrA and parC gene of N. gonorrhoeae DHPLC profiling pattern

Number (n)

gyrA gyrA gyrA gyrA gyrA

5 14 1 47 14

a b c d e

gyrA mutation Ser91

Asp95

wild-type Phe Phe Phe

Asn Gly Asn

Ciprofloxin MIC range (A g/ml) b0.002–0.031 0.125–0.5 0.0004 2– N8 0.031– N8

Gly85 f g h i j k

22 7 10 42 1 2

Asp

Asp86

DHPLC profiling strain had the same mutation pattern in both gyrA and parC, respectively. From this, it is apparent that the DHPLC profiling corresponded with their amino acid sequence in each case. Table 2 shows the DHPLC profiling patterns and their amino acid mutations, and ciprofloxacin MIC ranges of each pattern in gyrA and parC of N. gonorrhoeae. We can easily get the ciprofloxacin MIC ranges from DHPLC profilings (Figs. 1 and 2; Table 2).

4. Discussion

parC mutation parC parC parC parC parC parC

419

Ser87

wild-type Asn Arg Arg Arg Asn

Ser88

Pro

Pro

b0.002–4 0.125– N8 0.5– N8 0.063– N8 N8 4 – N8

different elution conditions. We tried and found that there were five profiling patterns in gyrA and six in parC. The DHPLC profiling peaks tended to be plural and became more complicated in the multiple mutant strains. These profilings could be reproducible repeatedly at the optimal condition. We compared differences in the DHPLC profiling to the sequencing results and found that the same

N. gonorrhoeae resistance to fluoroquinolones has been increasing during the past several years (Fox et al., 1997), and the resistant rate has been reported to be 70–80% in a particular region in the world (Chaudhry et al., 2002). In Japan, there has also been a gradual increase in level of resistance (Tanaka et al., 2000), and thus it is important to diagnose resistant strains more quickly and properly, and to identify the appropriate therapy. This study is the first to report the successful use of DHPLC for the screening of DNA sequence variants in QRDRs found in the gyrA and parC genes in N. gonorrhoeae. DHPLC has been used to analyze gene mutations linked to cancer in human cells (Yokomizo et al., 1998) and resistance to fluoroquinolones in staph-

Fig. 1. DHPLC traces corresponding to base changes, resulting in the following gyrA amino acid substitutions: (a) wild-type; (b) Asp-95Asn; (c) Ser-91Phe; (d) Ser-91Phe and Asp-95Asn; (e) Ser-91Phe and Asp-95Gly. The axes (not shown) represent retention times in minutes (x) and absorbance at 260 nm ( y). The direction of elution is left to right.

420

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421

Fig. 2. DHPLC traces corresponding to base changes, resulting in the following parC amino acid substitutions: (f) wild-type; (g) Asp-86Asn; (h) Ser-87Arg and Ser-88Pro; (i) Ser87Arg; (j) Gly-85Asp and Ser-87Arg; (k)Asp-86Asn and Ser-88Pro. The axes (not shown) represent retention times in minutes (x) and absorbance at 260 nm ( y). The direction of elution is left to right.

ylococcus aureus (M’ Zali et al., 2001). It has been reported that it is possible to analyze DNA fragments of up to 1.5 kb (Nickerson et al., 2000). In the present study, the size of the PCR product used was 225 bp in gyrA and 166 bp in parC gene. In general, because the recommended size is 150–450 bp (Taylor et al.), we were successful in carrying out the DHPLC analysis. The mutations detected by nucleotide sequence analysis could be rapidly and reproducibly identified in the PCR products using DHPLC, producing specific DNA wave patterns that correlate with the genotype in the QRDRs of gyrA and parC gene. Fluoroquinolone resistance, unlike bacterial resistance to many other antimicrobials, relies upon mutations in endogenous genes rather than on the acquisition of additional genetic information. This makes genetic characterization of fluoroquinolone resistance more demanding the use of current technology because it is sometimes necessary to identify a single nucleotide change rather than detect the acquisition of substantial amounts of exogenous DNA. The development and application of DHPLC will considerably simplify the identification of mutations associated with fluoroquinolone resistance (Hannachi-M’Zali et al., 2002). DHPLC analysis proved advantageous for the detection of multiple mutations, although its utility for recognizing more than three mutations towards the end of a DNA fragment may be limited. DHPLC also provided a

rapid, high-throughput alternative to Lightcycler and single-strand conformation polymorphism (SSCP) for screening frequently occurring mutations. The temperature-dependent, ion pair chromatography required only a 5-min gradient per sample (Eaves et al., 2002). Our samples showed various mutations and patterns. In particular, in gyrA gene at position 95, two kinds of alternation patterns were seen: aspartic acid for either asparagine or glycine. Only the mutation to asparagine at that position could be seen alone and another mutation was accompanied by the alternation of Ser91 to Phe. However, we could distinguish these same position 95 mutations clearly and rapidly with DHPLC, as was previously done by Eaves et al. (2002). In parC gene, novel mutations were located in the QRDR at nearby codons, Gly-85, Asp-86, Ser-87, and Ser-88, and included multiple combinations of amino acid substitutions. Those seen in the DHPLC profiling were all clearly distinguishable (Figs. 1 and 2). Our sequencing results showed that multiple mutations tended to be present where there was increased resistance to ciprofloxacin (Table 2). Silent mutations in positions 104 and 131 in parC mostly were seen in the QRDR sequence, and then the amino acid mutations were limited in the 85, 86, 87, and 88 codons. Thus, we made a different PCR reverse primer for DHPLC for the sequencing and so we could perform DHPLC for only important amino acid mutations for fluoroquinolone resistance. Because in gyrA gene the mutations positions 91

K. Shigemura et al. / Journal of Microbiological Methods 59 (2004) 415–421

and 95 were in the region that included a relatively high amount of guanine and cytosine, we needed to add the GC clamp to the reverse primer, as described above. We used a single optimal temperature in both gyrA and parC to discriminate all kinds of profilings, which was not done in previous studies (Eaves et al., 2002). A rapid-PCR protocol, followed by high-throughput DHPLC, could potentially allow the clinician to identify sensitivity patterns in less than half a day (in fact, optimization and streamlining of the protocol may allow results to be obtained in as little as 2 h). This would represent a several-fold reduction in the typical laboratory turn-around time. We recommend this technique as a rapid (5 min/sample), highcapacity (96 samples per run) method for the detection of resistance alleles and for characterizing antibiotic resistance in bacteria. This is the first report to apply to antimicrobialresistant-related N. gonorrhoeae of DHPLC, and in the papers related to DHPLC, we are the first to perform and establish DHPLC method of two kinds of gene in a single paper. The present results suggest that DHPLC profiling will allow us to diagnose fluoroquinolone resistance in N. gonorrhoeae and associate the extent of resistance so that we may optimize the therapy in clinical isolates in a single screening. We have now a design that this method will be applied to not only clinic but also to other gene-related diseases. References Chaudhry, U., Ray, K., Bala, M., Saluja, D., 2002. Mutation patterns in gyrA and parC genes of ciprofloxacin resistant isolates of Neisseria gonorrhoeae from India. Sexually Transmitted Infections 78, 440 – 444. Eaves, D.J., Liebana, E., Woodward, M.J., Piddock, L.J., 2002. Detection of gyrA mutation in quinolone-resistant Salmonella enterica by denaturing high-performance liquid chromatography. Journal of Clinical Microbiology 40, 4121 – 4125. Fox, K.K., Knapp, J.S., Holmes, K.K., Hook III, E.W., Judson, F.N., Thompson, S.E., Washington, J.A., Whittington, W.L.,

421

1997. Antimicrobial resistance in Neisseria gonorrhoeae in the United states, 1988–1994: the emergence of decreased susceptibility to fluoroquinolones. Journal of Infectious Diseases 175, 1396 – 1403. Hannachi-M’Zali, F., Ambler, J.E., Taylor, C.F., Hawkey, P.M., 2002. Examination of single and multiple mutations involved in resistance to quinolones in Staphylococcus aureus by a combination of PCR and denaturing high-performance liquid chromatography (DHPLC). Journal of Antimicrobial Chemotherapy 50, 649 – 655. Kristensen, V.N., Kelefiotis, D., Kristensen, T., Borresen-Dale, A., 2001. High-throughput methods for detection of genetic variation. BioTechniques 30, 318 – 332. M’ Zali, F.H., O’ Neill, A.J., Hawkey, P.M., Dalhof, A., Ambler, J.E., 2001. Examination of the spectrum of mutations involved in resistance to fluoroquinolones in Staphylococcus aureus by a combination of polymerase chain reaction (PCR) and denaturing high performance liquid chromatography (DHPLC). Journal of Antimicrobial Chemotherapy 47 (Suppl. S1), 79. National Committee for Clinical Laboratory Standards (NCCLS), 2002. Performance standards for antimicrobial susceptibility testing. Informational supplement. M100-S12. National Committee for Clinical Laboratory Standards, Wayne, PA. Nickerson, M.L., Weirich, G., Zbar, B., Schmidt, L.S., 2000. Signature-based analysis of MET proto-oncogene mutations using DHPLC. Human Mutation 16, 68 – 76. Oefner, P.J., Underhill, P.A., 1998. DNA mutation detection using denaturing high-performance liquid chromatography (DHPLC). Current Protocols Human Genetics 19, 7.10.1 – 7.10.12. (Suppl.). Shigemura, K., Shirakawa, T., Okada, H., Hinata, N., Acharya, B., Kinoshita, S., Kofuku, T., Kawabata, M., Kamidono, S., Arakawa, S., Gotoh, A., 2004. Mutations in the gyrA and parC genes and in vitro activities of fluoroquinolones in 91 clinical isolates of Neisseria gonorrhoeae in Japan. Sexually Transmitted Diseases 31, 180 – 184. Taylor, P., Munson, K., Gjerde, D., Transgenomic, Inc. Detection of Mutations and polymorphisms on the WAVEk DNA fragment analysis system. TRANSGENOMIC Application Note 101. Tanaka, M., Nakayama, H., Haraoka, M., Saika, T., 2000. Antimicrobial resistance of Neisseria gonorrhoeae and high prevalence of ciprofloxacin-resistant isolates in Japan, 1993– 1998. Journal of Clinical Microbiology 38, 521 – 525. Yokomizo, A., Tindall, D.J., Drabkin, H., Gemmill, R., Franklin, W., Yang, P., Sugio, K., Smith, D.I., Liu, W., 1998. RPTEN/ MMACI mutations identified in small cell, but not in non-small cell lung cancers. Oncogene 17, 475 – 479. Xiao, W., Oefner, P.J., 2001. Denaturing high-performance liquid chromatography: a review. Human Mutation 17, 439 – 474.