Helicobacter pylori susceptibility to clarithromycin and fluoroquinolone can be determined through Etest or molecular assays. We examined the correlation between phenotypic susceptibility (MIC results) and genotypic susceptibility in H. pylori strains isolated from gastric biopsies.
Results
Out of 85 duplicate biopsies obtained from patients from northern Israel who underwent endoscopy, 70 were positive for H. pylori using the CUTest (urease test). These biopsies were cultured for H. pylori and Etest breakpoints were determined for levofloxacin and clarithromycin. H. pylori detection, characterization of wild type (WT) or mutant alleles, and antibiotic resistance were determined molecularly using GenoType HelicoDR kit. Phenotypic and genotype results were compared. The molecular method had higher sensitivity for detection of H. pylori than culture (94.3% vs. 77.1%). Resistance to clarithromycin was higher than to levofloxacin by both methods. Levofloxacin resistance was found in 21.2 and 0% using genotypic and phenotypic methods, respectively. Clarithromycin resistance was found in 66.7 and 62.9% by genotypic and phenotypic methods, respectively. Some samples were found susceptible by Etest and resistant by GenoType HelicoDR (13/53 samples) and some vice versa (10/53 samples).
Conclusion
GenoType HelicoDR kit has high sensitivity rate for H. pylori detection. It has the ability to detect many mutations and can help to determine initial antibiotic treatment at the beginning of therapy. However, there were a few cases where there was a phenotype-genotype mismatch result that could derive from possible causes. For example, a resistance mechanism that is not tested in the kit.
Hinweise
Dana Binyamin and Nina Pastukh contributed equally to this work
Abkürzungen
WT
wild type
MUT
mutation
MIC
minimal inhibitor concentration
Background
Helicobacter pylori is a gram-negative bacterium known as a common pathogen associated with atrophic and metaplastic changes in the stomach, and peptic ulcer disease [1‐3]. Antibiotic treatment for H. pylori commonly includes a combination of proton pump inhibitors (PPIs) and one or several antibacterial agents such as amoxicillin, clarithromycin, and metronidazole [4]. Antibiotic resistance in H. pylori has been increasing worldwide, leading to failure of H. pylori eradication [5].
Clarithromycin remains the most powerful antibiotic currently available against H. pylori with minimal inhibitor concentrations (MICs) being the lowest compared to the other molecules. Clarithromycin inhibits protein synthesis by binding to the 50S bacterial ribosomal subunit [6]. Common resistance mechanisms to clarithromycin include point mutations in bacterial domain V of 23S rRNA, which prevents antibiotic binding [7]. There are 3 point mutations in the 23S rRNA gene: A2143G, A2142G, and A2142C; these account for 90% of cases for primary resistance in Western countries [8].
Anzeige
Fluoroquinolone agents (mainly levofloxacin, which is a broad-spectrum quinolone) are an alternative therapy for infections caused by H. pylori and serve as second-line treatment [9]. In recent years, resistance to fluoroquinolone has increased but seems to have remained low [10]. Fluoroquinolone targets the DNA gyrase, an enzyme responsible for negative supercoiling during the DNA replication process. This enzyme contains two A subunits and two B subunits, encoded by gyrA and gyrB, respectively [11]. Resistance to fluoroquinolones is mainly mediated by a mutation in the Quinolone Resistance Determining Region (QRDR) in the gyrA gene [12].
Antibiotic resistance is usually identified by phenotypic methods that determine the MIC breakpoints, and thus might be efficient for treating infection [13].
Resistance to clarithromycin and fluoroquinolones, which is mostly acquired through point mutations, can be detected by molecular techniques. GenoType HelicoDR (Hain Life Science, Germany) test is a molecular diagnostic method for easy and simultaneous detection of frequent point mutations responsible for clarithromycin and fluoroquinolones resistance [14].
The aim of this study was to evaluate the correlation between in vitro phenotypic susceptibility versus genotypic susceptibility to clarithromycin and fluoroquinolone, as expressed in H. pylori strains from gastric biopsies of patients in northern Israel.
Anzeige
Results
Helicobacter pylori presence in gastric specimens
A total of 70 positive specimens detected by urease test were examined for H. pylori presence by PCR and culture methods (Table 1). Thirty-four samples were from children (mean age 10.8 ± 4) and thirty-six from adults (mean age 36 ± 13.1); 45 females and 25 males.
Table 1
H. pylori presence in gastric specimens by PCR and culture methods
CUTest positive
CUTest negative
Total
Children
Adults
Total
Children
Adults
No. of samples
70
34
36
15
8
7
Age (mean ± SD)
23.76 ± 15.8
10.8 ± 4
36 ± 13.1
21.96 ± 11
10.98 ± 4
34.5 ± 14
Male/female
25/45
3/15
22/30
8/7
3/4
4/4
PCR+ culture+
53
28
25
0
0
0
PCR+ culture−
13
5
8
2
1
1
PCR− culture+
1
1
0
0
0
0
PCR− culture−
3
0
3
13
1
1
Sixty-six specimens were positive for H. pylori by PCR (94.3%), 54 (77.1%) were positive by culture or both. Thirteen samples were found positive only by PCR versus one positive sample determined only by culture method Out of 15 biopsies that were negative for H. pylori by urease test, 2 were positive by PCR.
Determination of H. pylori resistance to levofloxacin and clarithromycin by PCR or culture
Sixty-six specimens were examined for levofloxacin and clarithromycin resistance using the GenoType HelicoDR kit and 54 specimens (grown by culture) by E test (Table 2). Both methods detected a higher resistance rate to clarithromycin than to levofloxacin. Similarity in resistance rates to clarithromycin were demonstrated by GenoType HelicoDR kit and Etest (66.7 and 62.9%, respectively). Resistance to levofloxacin was found only by means of GenoType HelicoDR kit in 21.2% of specimens.
Table 2
Rates of resistance and susceptibility to levofloxacin and clarithromycin in accordance with detection methods (GenoType HelicoDR kit or Etest)
Method/antibiotic
Levofloxacin and clarithromycin susceptibility
Levofloxacin resistance
Clarithromycin resistance
GenoType HelicoDR (npositive = 66)
20 (30.3%)
14 (21.2%)a
44 (66.7%)a
E test (npositive = 54)
20 (37.1%)
0 (0%)
34 (62.9%)
aResistant to both antibiotics
Only 30.3 and 37.1% of samples were susceptible to both levofloxacin and clarithromycin by GenoType HelicoDR kit and Etest, respectively).
Determination of MIC breakpoints (phenotype) in comparison to gene expression (genotype)
Fifty-three samples that were positive for H. pylori by both methods were tested by GenoType HelicoDR for distribution of WT and mutant alleles of the gyrA and 23s genes that are known to be associated with fluoroquinolone and clarithromycin resistance, respectively. Strains with mutations in gyrA (N87K, D91N, D91G, and D91Y) were determined as resistant to levofloxacin; strains with mutations in 23S rRNA (A2142G, A2142C, and A2143G) were determined as resistant to clarithromycin. For each sample the MIC was determined by Etest, and resistance to clarithromycin was determined by MIC higher than 0.5 μg/mL and to levofloxacin by MIC higher than 1 μg/mL (Table 3).
Table 3
Distribution of genes (gyrA, 23S) in accordance with MIC breakpoints for fluoroquinolone and clarithromycin antibiotics
Gene
No. of samples
MIC range (μg/mL)
<0.125
0.19–0.25
0.38–0.5
0.75–1
1.5–2
3
>4
gyrA87 and gyrA91 genes
Children
WT
23
3
2
18
N87K
5
1
1
2
1
D91N
4
1
1
1
1
D91G
4
1
1
1
1
D91Y
0
Adults
WT
17
5
2
10
N87K
4
3
1
D91N
4
2
2
D91G
4
2
1
1
D91Y
2
1
1
23S rRNA (rrl) genes
Children
WT
9
2
2
2
1
2
A2142G
0
A2142C
2
1
1
A2143G
19
2
1
5
1
5
4
1
Adults
WT
7
1
1
1
2
2
A2142G
0
A2142C
2
1
1
A2143G
18
2
3
6
3
4
More than one gene was present in several samples
The gyrA WT genes were mainly identified in 40 samples (23 in children and 17 in adults) and all were in levofloxacin-susceptible MIC zones (0.38–0.75 μg/mL), in agreement with the presence of the WT detected genes. Mutants to gyrA were found in 14 samples, with more than one mutation in several. A mutation in D91Y was found in only 2 adult patients.
All of the strains with mutant alleles had an MIC ≤ 1, and thus were susceptible by the phenotypic method, but resistant in the genotypic method (Fig. 1).
Fig. 1
Samples distribution by MIC ranges and alleles appearance for fluoroquinolone antibiotic
×
The most frequent allele detected in the 23 s rRNA gene was the A2143G mutation (in 69.8% of samples, 19 in children and 18 in adults), with most of these (62.2%) demonstrating an MIC to clarithromycin >1 (resistant by the phenotypic method as well). Of the 15 specimens with WT alleles, 66.7% were found to be resistant by the Etest method (Table 3; Fig. 2). The A2142G mutation was not found in the study group.
Fig. 2
Samples distribution by MIC ranges and alleles appearance for clarithromycin antibiotic
×
Anzeige
Correlation between susceptibility determination methods—GenoType HelicoDR and Etest
Comparison between the results of susceptibility tests obtained by GenoType HelicoDR and Etest was done on 53 positive samples (Table 4). In regard to levofloxacin, 40 gastric samples were found to be susceptible by GenoType HelicoDR kit and Etest. However, 13 samples demonstrated resistance by GenoType HelicoDR but were susceptible by Etest. These findings indicate a possible intermediate correlation between two methods for levofloxacin resistance determination.
Table 4
A correlation of antibiotic susceptibility in accordance with GenoType HelicoDR kit and Etest
Antibiotic
Levofloxacin
Clarithromycin
Method
Etest
Etest
Resistance
Susceptible
Resistance
Susceptible
GenoType HelicoDR kit
Resistance
0
13/53 (24.53%)
24/53 (45.28%)
13/53 (24.53%)
Susceptible
0
40/53 (75.47%)
10/53 (18.87%)
6/53 (11.32%)
In regard to clarithromycin, 24 samples were found to be resistant by Etest and GenoType HelicoDR. Six samples were susceptible by both methods. A few samples exhibited mismatched results for the two methods; some were found susceptible by Etest and resistant by GenoType HelicoDR (13 samples) and some were found susceptible by GenoType HelicoDR and resistant by Etest (10 samples). Consequently, intermediate correlation between both methods was found.
The relative risk for clarithromycin resistance was higher than 1 (1.07 and 1.04 for MUT genotype in correlation to resistance in Etest and WT genotype in correlation to susceptibility in Etest, respectively). This means that the molecular method could serve as an alternative for the Etest for prediction of susceptibility of strains to Clarithromycin.
Discussion
In the current study a higher H. pylori prevalence was found among women, but no significant difference was found between children and adults.
Anzeige
Helicobacter pylori presence in gastric specimens can be assessed by various methods including molecular PCR assay (GenoType HelicoDR kit) and culture. Our findings demonstrate that PCR is a more sensitive method (80.3% sensitivity, 75.0% specificity), with a higher H. pylori detection rate compared to culture [14‐17].
The study group consisted of patients who underwent a rapid urease test and received a positive result. It is important to mention that the urease test is characterized by a poor sensitivity, and indeed 2 out of 15 CUTest negative specimens were positive by PCR and 3 out of 70 CUTest positive specimens were found negative by PCR and culture. Thus, it could be concluded that rapid urease test is a non-accurate assay with a probability of a false positive and negative results; therefore methods such as PCR and culture are recommended for H. pylori detection. Furthermore, performance of culture or PCR enables determination of antibiotic susceptibility profile for adjustment of effective antibiotic treatment and purposes of epidemiological research.
A phenotype-genotype correlation for levofloxacin and clarithromycin resistance profiles has been examined. MIC measurements by Etest demonstrated results expressed as a phenotype. Determination of antibiotic resistance profile by bacterial culture and Etest technique is common after a treatment failure. Although this procedure could prevent antibiotic failure, this process is not routinely carried out in all labs due to the difficulty to culture the bacterium and the excessive time required. As a result, many alternative methods for antibiotic resistance detection are available, such as GenoType HelicoDR kit.
This kit was designed through validation on strains with known MIC values for clarithromycin and fluoroquinolones and known genotypes encoded by 23S rRNA and gyrA [14]. The kit enables detection of strains’ profile genotypes, expressed as mutant or WT alleles.
Anzeige
In the current study resistance to fluoroquinolones was detected only by the molecular method GenoType HelicoDR kit, with a higher resistance rate compared to data published in Israel in 2015 [15, 16]. This finding could be explained by the high sensitivity of the molecular kit, which is not routinely used, enabling detection of genotypic resistance, in accordance with a worldwide trend of a rise in fluoroquinolones resistance [14, 17, 18].
In addition, resistance rate to clarithromycin has increased, and is 3 times higher (similar rates in both methods) in comparison to data published earlier in Israel, in accordance with the global increasing trend [19]. A possible explanation for a lower resistance rate to fluoroquinolones in comparison to clarithromycin could be due to its later addition to treatment protocols of H. pylori infection.
The increase in resistance rate emphasizes the need for accurate detection of antibiotic resistance for an optimal adjustment of personal treatment therapy. Furthermore, success of eradication treatment is enabled by a prior, early determination of antibiotic resistance [17].
The MIC values for fluoroquinolones were low; this may indicate the presence of mainly the susceptible phenotype profile, in accordance with samples that contained WT alleles indicating a susceptible genotype profile. Therefore, a concordance between phenotype and genotype expression was observed at the susceptibility area. However, in some specimens a difference between phenotype and genotype expression was observed, expressed as susceptible MIC values and mutant alleles, in accordance with other published research [20]. Previous studies had shown that a profile of levofloxacin genotype resistance detected by GenoType HelicoDR kit is consistent with therapeutic outcomes. In contrast, a high correlation between phenotype and genotype resistance profiles to levofloxacin was shown in other studies [21].
For clarithromycin, concordance between genotype analysis and phenotype results was observed in more than 60% of the specimens. However, some differences between phenotype and genotype profiles were observed; these were characterized by phenotypic susceptibility and genotypic resistance or vice versa, in contrast to other studies [22‐24].
Clarithromycin resistance has been associated with a higher risk of treatment failure. A2143G mutation has a higher risk of eradication failure [15, 16, 21, 25]. This mutation demonstrated the highest rate among examined mutations, in accordance with other studies [14, 22, 24].
Although many samples demonstrated a phenotype-genotype correlation, no statistically significant relation was found. This could be an outcome of a small sample size group or bias due to demographic data such as age, gender, ethnicity, or a case of H. pylori recurrence.
In cases of specimens demonstrating a genotypic mutant a susceptible phenotypic profile, a possible explanation could be the presence of a mixed bacterial population residing in the gastric specimens that may consist of susceptible and resistant strains [26]. In fact, this correlates with appearance of w.t. and mutant gene in the same sample. The significant advantage of the molecular method is a direct examination of gastric specimens with a complete diversity of strains (susceptible, resistant, or both) in contrast to the culture method, which is based on a random selection of colonies to perform the Etest. Another explanation for the difference in phenotype-genotype result could be that strains of H. pylori harbor two copies of the 23S rRNA gene in the chromosomal DNA; a mutation in one gene copy is shown to enable resistance to clarithromycin [27]. Therefore, GenoType HelicoDR method detects strains with a genotypic resistance profile that may not be detected by a phenotype assay [25, 28].
In cases of phenotypic resistance expression accompanied by genotypic susceptibility, a reasonable explanation could be the presence of additional resistance mechanisms that were not examined in this study. For instance, a well-known mechanism for clarithromycin resistance is the efflux pump system or other point mutations such as T2182C and C2611A that have been associated with low resistance levels [29]. Even so, the examined alleles are present in higher rates than other resistance mechanisms [8].
Another limitation of the kit is lack of necessary susceptibility detection of amoxicillin and metronidazole, two antibiotic agents used against H. pylori. In addition, in GenoType HelicoDR test, as in other molecular assays, a possible risk of contamination of the amplified DNA during PCR performance is present, which may yield false positive outcomes.
Conclusions
GenoType HelicoDR kit has a high sensitivity rate for H. pylori detection in accordance with others studies [14‐17]. This molecular assay can detect a mixture of genotypes and could successfully analyze biopsies without transport or storage limitations at sort time. Consequently, it might serve as an effective option for rapid H. pylori determination and for initial antibiotic selection at the beginning of therapy Due to the intermediate correlation between phenotype-genotype results, culture performance and susceptibility tests should be done for specific antibiotic treatment adjustment.
Experimental procedures
Patient selection
The specimens used for this study were gastric biopsy samples from patients residing in northern Israel who had undergone endoscopy due to abdominal pain or discomfort, from March 2015 to April 2016.
Eighty-five patients underwent gastroscopy procedures; 42 were children (mean age 10.8 years) and 43 were adults (mean age 35.7 years). In case of suspected H. pylori infection a rapid urease test (CUTest; Temmler Pharma GmbH & Co) was performed on the antrum biopsies.
A total of 70 patients were determined as positive by rapid urease test and 15 were found negative; subsequently all 85 biopsies were referred to a microbiology lab.
None of the patients included in the study received antibiotic treatment during the month preceding the gastroscopy procedure. The study was approved by the institutional review board and informed consent was obtained from all study participants. Each specimen was split into two duplicates for purposes of bacterial culture and molecular assay.
Bacterial culture and determination of in vitro susceptibility
All biopsy specimens were placed in sterile Eppendorf tubes containing 1 mL sterile physiological solution (0.9% NaCl) and were sent to the Clinical Microbiology Lab within 30 min after the gastroscopy procedure under cold conditions. First, specimens were minced manually with a sterile scalpel and seeded on Modified BD Helicobacter Agar (BD Diagnostics, Sparks, MD, USA) in accordance with the relevant protocol. Then, the plates were incubated for 7 days at 37 °C in a micro-aerobic atmosphere (5% O2 and 10% CO2) produced by a gas generating system adapted for Campylobacter (CampyGen™, Gamidor Diagnostics). If no bacterial growth was seen after 7 days, an additional 3 days of incubation were added. H. pylori was identified based on a Gram-staining procedure followed by positive oxidase, catalase, and urease tests.
Antibiotic susceptibility tests were performed using the Etest for determination of each antibiotic’s MIC (minimum inhibitory concentration) against H. pylori.
Determination of MIC
Samples from primary plates were suspended in 0.85% NaCl solution to a 3.0 McFarland standard to be subjected to Etest (bioMérieux, Durham, USA) on Mueller–Hinton agar with 10% horse blood (hy-lab, Israel) for levofloxacin. The suspensions were incubated for 72 h at 35 °C in microaerobic atmospheric condition (CampyGen™, Gamidor Diagnostics, Israel). MIC breakpoints were defined as the lowest concentration of an antimicrobial to inhibit visible growth of a microorganism.
Interpretation of susceptibility tests results was performed in accordance with European Committee on Antimicrobial Susceptibility Testing (EUCAST) recommendations. Based on these recommendations, H. pylori isolates were considered resistant to levofloxacin when MIC level was greater than 1 μg/mL and considered resistant to clarithromycin when MIC level was greater than 0.5 μg/mL; intermediate resistance was defined by MIC = 0.5 and ≤0.25 μg/mL for susceptible strains.
The MIC was read at the point of complete inhibition of all growth, including hazes and isolated colonies. For MIC QC the H. pylori 43504 strain was used.
Tissue obtained from gastroscopic biopsy was minced using a sterile scalpel, lysed by tissue lysis buffer and Proteinase-K enzyme (Bioneer, Daejeon, Korea), and incubated for 10 min in 60 °C. Then, total DNA was extracted with an AccuPrep Genomic DNA Extraction Kit (Bioneer, Daejeon, Korea). This kit contains a glass filter fixed in a column tube that can bind efficiently to DNA in the presence of chaotropic salts. The spin-column method was used to eliminate contaminants and enzyme inhibitors. Additional washing steps were performed for proteins and salt removal, then high-purity DNA was eluted using a Tris–EDTA (pH 8.0) low-concentration elution buffer. DNA samples were stored at −20 °C until required. Aliquots of 50 µL were used for PCR amplification.
PCR amplification
DNA regions involved in acquired clarithromycin (rrl gene) or fluoroquinolone (gyrA gene) resistance were amplified by multiplex PCR as previously described [14‐16]. Briefly, reaction mixtures (volume of 45 μL) contained 1 × 5 μL reaction buffer, 2 μL MgCl2, 3 μL DDW, 35 μL biotinylated primers and nucleotide mixture, and 0.2 μL Hot Star Taq polymerase (hy-labs, Rehovot, Israel); 5 μL of DNA were added to each reaction mixture. PCR reactions were carried out for 35 cycles for each biopsy specimen. The denaturation cycle consisted of 1 cycle at 95 °C for 5 min, followed by 10 cycles at 95 °C for 30 s, and 58 °C for 2 min. Then, 25 cycles were composed of a first step at 95 °C for 25 s, a second step at 53 °C for 40 s, and a third step at 70 °C for 40 s. The PCR reaction terminated with 8 min at 70 °C.
Hybridization
Hybridization was performed at a temperature of 45 °C. The denaturation solution was mixed with 20 μL of the amplified sample and was hybridized using a standard hybridization protocol. The hybridization forms were detected by enzyme immunoassay upon addition of conjugate and substrate solution.
This strip was coated with different specific oligonucleotides (DNA probes) designed to hybridize with the sequences of the amplified wild type (WT probes) or the mutant alleles (MUT probes). GenoType HelicoDR test was designed for identification of mutations in A2142C, A2142G, and A2143G for 23S rRNA gene (encoding for clarithromycin resistance) and in N87K, D91N, D91G, and D91Y for gyrA gene (encoding for fluoroquinolone resistance).
Validation control bands are designed on the strip, such as: a conjugate control (CC), amplification control (AC), and H. pylori (HP), which must appear in addition to gyrA and 23S control bands.
Interpretation
Strips were attached to the evaluation sheet after hybridization. A determination of positive band was done through comparison of each band’s stain intensity with the amplification control band. More intense stain as compared with the amplification band was interpreted as positive for the presence of the specific allele.
In case of outcome different than strains resistant to clarithromycin or fluoroquinolone, or strains resistant to both antibiotics, a wild type (WT) strain was determined by the presence of WT bands. Mutants were determined by presence of MUT band. Each amplified sample could exhibit more than one band for each type of gene (WTs or MUTs).
Statistical analysis
All measured variables were tabulated by descriptive statistics. Categorical variables were summarized in tables providing sample size, and absolute and relative frequency by Culture and PCR.
Sensitivity and specificity analysis and 95% confidence intervals (CIs) were computed using a 2 × 2 table, in order to evaluate conditional probability and to predict positive or normal condition in concordance with the test result. The data was analyzed using SAS® version 9.3 (SAS Institute, Cary, North Carolina, USA).
Categorical data were compared using Chi square test. The level of statistical significance was specified as 0.05. Statistical analyses were performed using SPSS 23 statistical software. Relative risk was calculated for determination of the association strength between MUT genotype determined by molecular assay and phenotype resistance determined by Etest for Clarithromycin.
Authors’ contributions
DB and NP performed the experiments, analysed and interpreted the data. AP conceived and designed the study and participated in the interpretation. AO and MP collected the data. All authors have provided comments on the drafts. All authors read and approved the final manuscript.
Acknowledgements
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Not applicable.
Ethics approval and consent to participate
The study is approved by the Helsinki Committee of the Medical Center, No. committee approval is 23-13-POR.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
