Infection, Genetics and Evolution 23 (2014) 196–202

Contents lists available at ScienceDirect

Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid

Heteroresistance of Helicobacter pylori from the same patient prior to antibiotic treatment Cheng-Yen Kao a, Ai-Yun Lee b, Ay-Huey Huang c, Pin-Yi Song d, Yao-Jong Yang e, Shew-Meei Sheu f, Wei-Lun Chang f, Bor-Shyang Sheu f,⇑,1, Jiunn-Jong Wu d,g,⇑,1 a

Institute of Basic Medical Sciences, College of Medicine, National Cheng Kung University, Tainan, Taiwan Institute of Molecular Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan Department of Pathology, College of Medicine, National Cheng Kung University, Tainan, Taiwan d Department of Medical Laboratory Science and Biotechnology, College of Medicine, National Cheng Kung University, Tainan, Taiwan e Department of Pediatrics, College of Medicine, National Cheng Kung University, Tainan, Taiwan f Department of Internal Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan g Center of Infectious Disease and Signaling Research, College of Medicine, National Cheng Kung University, Tainan, Taiwan b c

a r t i c l e

i n f o

Article history: Received 24 October 2013 Received in revised form 7 February 2014 Accepted 10 February 2014 Available online 24 February 2014 Keywords: Helicobacter pylori Antibiotic Heteroresistance Treatment failure

a b s t r a c t Antibiotic resistance among Helicobacter pylori strains has been increasing worldwide and has affected the efficacy of current treatments. The aim of this study was to evaluate whether treatment failure was due to the presence of antibiotic-susceptible and -resistant H. pylori simultaneously within the same host before eradication. In order to discover H. pylori with antibiotic heteroresistance in the same patient, we examined the antibiotic susceptibility of H. pylori isolated from 412 patients without H. pylori eradication. The E-test was used to determine the minimal inhibitory concentration of these strains. The results showed 19 (4.6%) of patients harbored antibiotic heteroresistant H. pylori, resistant to levofloxacin (5/19), clarithromycin (1/19) and metronidazole (16/19). Among them, three patients’ isolates showed heteroresistance to two antibiotics. The genetic diversity of each isolate was evaluated by random amplified polymorphic DNA PCR and the results showed that only 1 patient’ isolate (5.3%) had a different pattern while the others showed identical or similar fingerprinting patterns. Mutations in the genes responsible for antibiotic resistance were investigated by direct sequencing and compared between strains within each pair. All 5 levofloxacin-resistant isolates had mutations in GyrA at the QRDR region (N87 or D91). Strain 1571R with clarithromycin resistance had a A2042G substitution in its 23S rRNA. There were 15 metronidazole-resistant strains (100%) with isogenic variation of RdxA, and 6 strains (40%) contained FrxA variation (excluded pair 1159). These results suggest that the treatment failure of heteroresistant H. pylori mostly develops from high genomic variation of pre-existing strains through long term evolution rather than mixed infection with different strains. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction Helicobacter pylori is a Gram-negative, spiral-shaped, and microaerophilic bacterium that infects 50% of the population worldwide (Pounder and Ng, 1995). Persistent infection with H. pylori increases the risk of developing gastroduodenal diseases, including peptic ulcer, duodenal ulcer and gastric adenocarcinoma

⇑ Corresponding authors. Address: Department of Medical Laboratory Science and Biotechnology, College of Medicine, National Cheng Kung University, No. 1, University Rd., Tainan 70101, Taiwan (J.-J. Wu). Tel.: +886 6 2353535x5775; fax: +886 6 2363956. E-mail address: [email protected] (J.-J. Wu). 1 Bor-Shyang Sheu and Jiunn-Jong Wu contributed equally to this article. http://dx.doi.org/10.1016/j.meegid.2014.02.009 1567-1348/Ó 2014 Elsevier B.V. All rights reserved.

(Ahmad et al., 2003; Graham et al., 1993; Marshall and Warren, 1984; Parsonnet et al., 1991; Rauws and Tytgat, 1990). Although treatment regimens containing a proton pump inhibitor (ppi) and combination of two or more antibiotics (amoxicillin, clarithromycin, metronidazole or tetracycline) for 7–14 days are considered to be the most efficacious previously (Graham, 2000), successful eradication of H. pylori remains a challenge to clinicians due to resistance to the commonly used antibiotics for H. pylori eradication increasing worldwide (Chang et al., 2009; Graham and Fischbach, 2010; Megraud, 2007). Several alternative therapies, including concomitant and sequential therapy, are suggested to treat patients initially with PPI triple therapy failure or infected with clarithromycin-resistance strain (Graham and Fischbach, 2010; Lin et al., 2002; Vaira et al., 2007).

197

C.-Y. Kao et al. / Infection, Genetics and Evolution 23 (2014) 196–202

Sequential therapy is a new treatment, instead of administering the antimicrobials all at once, they are administered in sequence. It begins with amoxicillin and a PPI followed by clarithromycin and metronidazole, again with a PPI or the four drugs prescribed concomitantly. Several studies in various countries have proven that concomitant and sequential therapies seem to maintain a high level of efficacy with eradication rates above 90%. (Gatta et al., 2009; Okada et al., 1999; Park et al., 2012; Vaira et al., 2007). Resistance to quinolones is mainly due to mutations in the quinolone-resistance-determining region (QRDR) at codons 86, 87, 88 and 91 of the gyrA gene, coding for the A subunit of the DNA gyrase (Gerrits et al., 2006). Mutations in gyrB are also reported in quinolone-resistant strains, but often occurred together with gyrA mutations (Miyachi et al., 2006). The bacteriostatic activity of clarithromycin depends on its capacity to inhibit protein synthesis by binding to the 50S bacterial ribosomal subunit. Clarithromycin resistance is mainly due to point mutations in the 23S ribosomal RNA (rRNA) gene, and nucleotides A2142G and A2143G are the most frequent mutations (Taylor et al., 1997). Metronidazole resistance among H. pylori strains has been related to alterations in gene products having metronidazole nitroreductase activities, mainly including oxygen-insensitive NAD(P)H nitroreductase (RdxA) and NAD(P)H flavin oxidoreductase (FrxA) (Aldana et al., 2005; Jenks et al., 1999; Jeong and Berg, 2000; Jeong et al., 2000). Although H. pylori from individual patients typically have either an antibiotic-susceptible or -resistant phenotype, both antibioticsusceptible and -resistant H. pylori can be present simultaneously (Arents et al., 2001; van der Ende et al., 2001). Heteroresistance can represent infection with a single strain or mixed-infection with several different H. pylori strains (Arents et al., 2001; Kwon et al., 2001; Lee et al., 2005; Matteo et al., 2006; van der Ende et al., 2001). Kennemann et al. indicated H. pylori genomic evolution during infection shows genome-wide recombination in H. pylori colonizing humans, especially with a high rate of mixed infections (Kennemann et al., 2011). These results suggest that continuing genomic diversities through long-term evolution in the same strain may play an important role in modulating antibiotic heteroresistant H. pylori in vivo and thus lead to treatment failure. In this study, we collected 19 pairs of H. pylori containing heteroresistant phenotypes from separate patients before H. pylori eradication and further investigated the mechanisms leading to antibiotic resistance. 2. Materials and methods 2.1. Bacterial strains and growth conditions The information on bacterial strains, antibiotic susceptibility and the clinical diseases of patients in this study are described in Table 2. Nineteen heteroresistant H. pylori pairs were identified from 412 individual patients enrolled in clinical trials at the National Cheng Kung University Hospital, Taiwan. No patients had received H. pylori eradication therapy before this study. H. pylori strains initially harvested as multiple colonies from individual gastric mucosal biopsy specimens were cultured on CDC anaerobic blood agar (BBL, Microbiology Systems, Cockeysville, MD) under microaerophilic conditions. All strains were stored at 80 °C in Brain–Heart Infusion (BHI) broth containing 30% glycerol until testing. 2.2. In vitro antibiotic MIC test The MIC values of each clinical isolate consecutively collected to four antibiotics (amoxicillin, clarithromycin, metronidazole and levofloxacin) were determined by Epsilometer test (E-test) (AB Biodisk, Solna, Sweden) as described previously (Chang et al., 2009; Hung et al., 2009). In brief, these H. pylori isolates were

suspended in 0.85% NaCl solution and adjusted to a 2.0 McFarland standard. A blood agar plate was streaked in three directions with a swab dipped into each bacterial suspension to produce a lawn of growth, an E-test strip was placed onto each separate plate, which was immediately incubated in a microaerophilic condition at 37 °C for 72 h. Then the MIC was interpreted according to the manufacturer’s instructions. H. pylori American Type Culture Collection (ATCC) J99 was used as the quality control strain. Using recommendations of the European Committee on Antimicrobial Susceptibility Testing (EUCAST), resistance to levofloxacin, metronidazole, clarithromycin and amoxicillin was defined as MIC > 1 lg/mL, >8 lg/mL, MIC > 0.5 lg/mL, and MIC > 0.12 lg/mL, respectively. The MIC values of nineteen heteroresistant H. pylori pairs were further confirmed by the agar dilution method according to the Clinical and Laboratory Standards Institute guidelines (CLSI, 2010). 2.3. Random amplified polymorphic DNA (RAPD)–PCR amplification Mini Qiagen columns and a QiaAmp DNA extraction kit (Qiagen, Valencia, CA) were used for chromosomal DNA extraction. Three different primers for RAPD–PCR analysis used in this study are listed in Table 1. The RAPD–PCR reaction conditions were described previously with modification (Akopyanz et al., 1992; Sheu et al., 2009). In brief, the PCR mixtures were made in a volume of 50 lL containing 100 ng of DNA, 10 pmol of primer, 0.15 mM each deoxynucleoside triphosphate, reaction buffer with MgCl2, and 1 U of Taq DNA polymerase (Promega, Madison, WI). PCR was carried out in a Perkin–Elmer 2720 thermal cycler (Perkin–Elmer; Applied Biosystems, Foster City, CA, USA) through 8 cycles of low-stringency amplification and 35 cycles of high-stringency amplification. The PCR products (15 lL) were electrophoretically separated in 1% agarose gels. The classes of fingerprint patterns were (i) highly homogeneous RAPD banding patterns in the isolates of one patient were defined as single strain infection, (ii) more than one different RAPD banding pattern was defined as mixed-infection by different strains (Sheu et al., 2009). 2.4. Gene mutation analysis by direct sequencing PCR was carried out according to the manufacturer’s instructions using Taq polymerase (Promega). Primers used in this study are described in Table 1. Using these primers, a 428 base pair fragment containing the quinolone resistance-determining region of gyrA gene was amplified. Reaction conditions were an initial 5 min at 95 °C, and 45 s at 95 °C, 45 s at 55 °C, and 45 s at 72 °C for 26 cycles, followed by a final extension for 10 min at 72 °C. Table 1 Oligonucleotide primers used in this study. Primer

Sequence (50 -30 )

PCR-Sequencing gyrA-1 TTTAGCTTATTCAATGAGCGT gyrA-2 GCAGACGGCTTGGTAGAATA gyrB-1 TGCAAAAGCCAGAGAAGCCA gyrB-2 ACATGCCCTTGTTCAATCAGC rdxA-1 GCAGGAGCATCAGATAGTTCT rdxA-2 GGGATTTTATTGTATGCTACAA 23s rRNA-1 ATGGGAGCTGTCTCAACCAG 23s rRNA-2 TGTCCTGCCTGTGGATAACA 50bf-frxA-Fw ATTGGATATGGCAGCCGTTTATCA frxA-3-Rv AGCGTTTTTATTCAATCACTTCATAA p5-1 CGAATTGGATATGGCAGCCG RAPD–PCR 226 227 228

AAGAGCCCGT NNNAACAGCTATGACCATG GAGCGGCCAAAGGGAGCAGAC

Reference Tankovic et al. (2003) Tankovic et al. (2003) Jeong and Berg (2000) This study This study Marais et al. (2003) Akopyanz et al. (1992) Akopyanz et al. (1992) Akopyanz et al. (1992)

198

C.-Y. Kao et al. / Infection, Genetics and Evolution 23 (2014) 196–202

For gyrB sequencing, a 444 base pair fragment was amplified. Reaction conditions were an initial 5 min at 95 °C, and 45 s at 95 °C, 45 s at 56 °C, and 45 s at 72 °C for 26 cycles, followed by a final extension for 10 min at 72 °C. For 23S rRNA sequencing, a 245 base pair fragment was amplified. Reaction conditions were an initial 5 min at 95 °C, and 45 s at 95 °C, 45 s at 64 °C, and 30 s at 72 °C for 26 cycles, followed by a final extension for 7 min at 72 °C. For rdxA sequencing, a 886 base pair fragment was amplified. Reaction conditions were an initial 5 min at 95 °C, and 45 s at 95 °C, 45 s at 50 °C, and 60 s at 72 °C for 26 cycles, followed by a final extension for 10 min at 72 °C. For frxA sequencing, a 711 base pair fragment was amplified. Reaction conditions were an initial 5 min at 95 °C, and 45 s at 95 °C, 45 s at 57 °C, and 60 s at 72 °C for 26 cycles, followed by a final extension for 10 min at 72 °C. The products were then directly sequenced by an automated Applied Biosystems 3730 DNA Analyzer (Applied Biosystems, Foster, CA, USA). 3. Results 3.1. Antibiotic-heteroresistance of H. pylori We collected 19 pairs of H. pylori with antibiotic heteroresistance from 412 patients prior to antibiotic therapy. The number of naïve H. pylori pairs showing heteroresistance to amoxicillin, levofloxacin, clarithromycin and metronidazole were 0, 5 (26.3%), 1 (5.3%), and 16 (84.2%), respectively (Table 2). Interestingly, three pairs (pairs 1147, 1528 and 1615) showed heteroresistance to two antibiotics. Strain 1147R1 was resistant to metronidazole (MIC > 256 lg/mL), and 1147R2 was resistant to levofloxacin (MIC > 32 lg/mL). Strain 1528S was sensitive to four antibiotics tested in this study, whereas 1528R1 was resistant to levofloxacin (MIC > 32 lg/mL), and 1528R2 was resistant to metronidazole (MIC > 256 lg/mL). Furthermore, strains 1615R1 and 1615R2 were resistant to levofloxacin (MIC = 1.5 lg/mL) and metronidazole (MIC > 256 lg/mL), respectively (Table 2). Nineteen pairs showing antibiotic heteroresistance were isolated from 9 patients with gastritis, 8 patients with gastro-duodenal ulcer, and 2 patients with gastric cancer (Table 2).

3.2. RAPD genotype analysis of antibiotic-heteroresistant H. pylori To understand whether antibiotic-heteroresistant H. pylori develop from a pre-existing strain or from different strains, we compared fingerprinting genotypes of the isolates in 19 pairs. The results showed that all pairs except 1 (pair 1159) had similar or identical profiles (Fig. 1), and the RAPD patterns were different between each pair. Pair 1159 showed different fingerprinting patterns between the susceptible and the resistant strains, suggesting that the patient might have a mixed infection (Fig. 1). 3.3. gyrA, gyrB and 23S rRNA mutations In order to clarify the mechanisms leading to antibiotic resistance, PCR sequencing was used to analyze the gene mutations. The nucleotide and amino acid sequence comparison between strains isolated from the same patient showed antibiotic heteroresistance. Table 3 shows the association between gyrA and gyrB mutations and MIC levels for levofloxacin in H. pylori. There were three amino acid mutation sites found in the GyrA of levofloxacin-resistant strains, including positions 87, 91, and 143 (Table 3). A A406G amino acid substitution in GyrB was only found in strain 1527R. In addition, a high nucleotide variation rate in gyrA was identified in group 1527 compared to other groups showing heteroresistance to levofloxacin (Table 3). In a previous study, mutations in domain V of 23S rRNA were directly associated with clarithromycin-resistance (Taylor et al., 1997). Strain 1571R showed resistant to clarithromycin (MIC = 16 lg/mL) with a substitution in the 23S rRNA gene at nucleotide position 2142 from A to G. 3.4. rdxA and frxA mutations The rdxA sequencing results of the 15 metronidazole-resistant strains (except 1159R) showed that 6 strains contained a nucleotide insertion or deletion. Among them, 3 strains (1504R, 1528R2 and 1675R) had a frameshift mutation, 2 strains (874R and 1009R) had prematurely truncated RdxA due to an early stop

Table 2 Clinical disease and antibiotic-heteroresistance of 19 pairs of H. pylori from individual patients. Strains

798S 798R 874S 874R 922S 922R 996S 996R 1009S 1009R 1147R1b 1147R2b 1158S 1158R 1159S 1159R 1192S 1192R 1232S 1232R

Clinical diseasea

DU Gastritis DU Gastritis Gastritis Gastritis Gastritis Gastritis DU Gastritis

Number of heteroresistant pair

MIC (ug/mL)

Strains

LVX

MET

CLA

AC

0.125 0.25 0.125 0.094 0.094 0.094 0.125 0.125 0.023 0.023 0.047 >32 0.125 0.125 0.125 0.094 0.125 0.125 0.125 0.125

0.75 128 0.5 64 0.19 >256 0.75 >256 4 64 >256 0.25 0.25 >256 0.125 >256 1 64 0.38 >256

0.094 0.75