Editor's ChoiCE
Comparison of new and classical point mutations associated
with clarithromycin resistance in Helicobacter pylori strains
isolated from dyspeptic patients and their effects on phenotypic
clarithromycin resistance
Bekir Kocazeybek1,*, Merve Kutlu Sakli1, Pelin Yuksel1, Mehmet Demirci2, Reyhan Caliskan1, Tevhide Ziver Sarp3, Suat
Saribas1, Suleyman Demiryas4, Fatma Kalayci1, Huseyin Cakan5, Hayriye Kirkoyun Uysal6, Nesrin Gareayaghi7, Sevgi
Ergin1, Yusuf Ziya Erzin8, Kadir Bal8, İhsan Tascı4 and Hrisi Bahar Tokman1 DOI 10.1099/jmm.0.000944
Received 06 November 2018; Accepted 28 January 2019; Published 06 February 2019
Author affiliations: 1Department of Medical Microbiology, Cerrahpasa Faculty of Medicine, Istanbul University – Cerrahpasa, Istanbul, Turkey; 2Beykent University Medical Faculty, Department of Medical Microbiology, Istanbul, Turkey; 3Eastern Mediterranean University, Faculty of Health Sciences, Nutrition and Dietetic Department, Famagusta, North Cyprus; 4Department of General Surgery, Cerrahpasa Faculty of Medicine, Istanbul University – Cerrahpasa, Istanbul, Turkey; 5Department of Microbiology, Istanbul University-Cerrahpasa, Institute of Forensic Sciences, Istanbul, Turkey; 6Istanbul University, Istanbul Medical Faculty, Department of Medical Microbiology, Istanbul, Turkey; 7Istanbul Sisli Hamidiye Etfal Training and Research Hospita, Blood Center, Istanbul, Turkey; 8Department of Gastroenterology, Cerrahpasa Medical Faculty, Istanbul University – Cerrahpasa, Istanbul, Turkey. *Correspondence: Bekir Kocazeybek, bzeybek@ istanbul. edu. tr
Keywords: Helicobacter pylori; A2115G, A2144T and G2141A; point mutations; clarithromycin.
Abbreviations: B, beta regression coefficient; CI, confidence interval; d.f., degree of freedom; DNA, deoxyribonucleic acid; EUCAST, European Committee on Antimicrobial Susceptibility Testing; H. pylori, Helicobacter pylori; IDT, integrated DNA technologies; MALT, mucosa associated lymphoid tissue; OR, odds ratio; PPI, proton pump inhibitör; SE, standard error.
InTRoduCTIon
The Gram-negative microaerophilic bacterium Helicobacter
pylori can colonize the human gastric mucosa. H. pylori is
involved in a wide variety of diseases, such as asymptomatic
gastritis, peptic ulcer, gastric adenocarcinoma and mucosa-associated lymphoid tissue (MALT) lymphoma, and is accepted as a type I carcinogen by the World Health Organization [1, 2]. To eradicate H. pylori infections, the standard therapy, worldwide, is a combination of a proton
Abstract
Purpose. We aimed to investigate the presence of three recently identified point mutations (A2115G, G2141A and A2144T) of the 23 S rRNA gene and compare them with the three most frequently encountered point mutations (A2142G, A2142C and A2143G) in Helicobacter pylori strains in Turkey.
Methodology. A total of 63 patients (mean 47.08±12.27) were included. The E-test method (for clarithromycin) was used for the clarithromycin antimicrobial susceptibility test of isolated H. pylori strains. Real-time PCR was used to detect the point mutations.
Results. A total of 24 out of 63 H. pylori strains (38.1%) were detected as clarithromycin resistant (>0.5 mg l−1 ). The new A2115G
(n:6, 25%), A2144T (n:7, 29.1%) and G2141A, 8 (n:8, 33.3%) mutations and the classical A2142G (n:8, 33.3%) and A2143G (n:11, 45.8%) point mutations were detected in the 24 clarithromycin-resistant strains. The A2144T point mutation had the highest median MIC value (3 mg l−1 ) amongst the new mutations, but the classical mutations (A2142G and A2143G) had the highest
median MIC values (256 mg l−1 ) overall. The presence of the A2115G (OR:31.66), A2144T (OR:36.92) or G2141A (OR:28.16)
muta-tions increased the likelihood of clarithromycin resistance in H. pylori strains by 31.66-, 36.92- and 28.16-fold (ORs), respec-tively, according to the binary logistic regression analysis.
Conclusion. We concluded that classical mutations of the 23 S rRNA gene resulted in higher clarithromycin MIC values than new mutations. These new point mutations caused moderate elevations in the MIC values of clarithromycin-resistant H. pylori strains.
pump inhibitor (PPI), or ranitidine bismuth citrate, and two antibiotics (chosen from among amoxicillin, clarithromycin and metronidazole) for 7 days; however, treatment failure is inevitable as the result of H. pylori rapidly developing resistance to clarithromycin. The treatment success of even this triple therapy had decreased into the ‘unacceptably low’ (<80%) category in most regions of the world by the 2000s [3–5].
Clarithromycin is a macrolide antibiotic that binds to the 23S rRNA of bacterial ribosomes, thereby inhibiting bacterial peptide translation. The clarithromycin resistance of H. pylori is typically caused by point mutations in domain V of the 23S rRNA gene, which has a role in the peptidyl transferase reac-tion, an important step for translation [6]. The most frequent 23S rRNA mutation is adenine-to-guanine conversions at positions 2142 and 2143 [7]. Other mutations also reported in clarithromycin-resistant strains are A2515G+A2223G, T2717C, A2115G+C2244T+A2116G, T2295C+T2182C, G2224A, A2302G, C2245T, T2289C, G2141A+A2144T and C2288T [8–10].
In this study, we aimed to investigate the presence of three recently identified point mutations (A2115G, G2141A and A2144T) and the three most frequently encountered point mutations (A2142G, A2142C and A2143G), i.e. ‘classical’ mutations, in H. pylori strains in Turkey. Additionally, we also aimed to compare the effects of the new and clas-sical mutations on phenotypic clarithromycin resistance in
H. pylori strains. We hope to provide new, alternative
perspec-tives toward antimicrobial therapy regimens in the treatment of H. pylori infections.
METHodS
This study was conducted between June 2014 and June 2017 in the Department of Internal Medicine and the Division of Gastroenterology as well as the Cerrahpasa Medical Faculty, Department of Medical Microbiology. Sixty-three patients with dyspeptic symptoms who were admitted to the endos-copy unit of Istanbul University in the Cerrahpasa Faculty of Medicine in the Department of Internal Medicine and the Division of Gastroenterology were enrolled during this period. The exclusion criteria were: age less than 18 years, no consumption history of antibiotics for H. pylori eradication treatment in the previous month, or consump-tion of antisecretory drugs without previous gastric surgery history. Any patients with bleeding or coagulation disorders that might be complicated by biopsy sampling were also excluded. Two biopsies from the antrum and corpus were collected and transferred immediately, in Brucella broth, to the laboratory.
The study was approved by the Clinical Research Ethics Board of Istanbul University, Cerrahpasa Faculty of Medicine, and all patients gave their informed consent before participating in the study. Ethical approval No: A-15.
Helicobacter pylori isolation and antimicrobial
susceptibility testing
H. pylori was cultured using Helicobacter Agar (Salubris,
Turkey) supplemented with 10% (v/v) defibrinated horse blood and an H. pylori-selective antibiotic supplement (Oxoid, UK) containing vancomycin (5 mg l−1), cefsulodin
(2.5 mg l−1), trimethoprim (2.5 mg l−1) and amphotericin B
(2.5 mg l−1). The plates were incubated for 3–5 days at 37 °C in
an incubator with 5% CO2 (Sanyo, Panasonic, USA). H. pylori was identified based on the colony and cellular morphology (i.e. Gram-negative translucent colonies) with positive urease, catalase and oxidase tests.
The E-test method was used to test the clarithromycin antimi-crobial susceptibility of isolated H. pylori strains. McFarland 3 suspensions were prepared in Brucella broth from 3 to 5 day agar plate cultures and inoculated with a Steers apparatus onto Mueller–Hinton agar supplemented with 10% horse blood. Clarithromycin E-test (BioMerieux, Marci L’étoile France) strips were placed onto the inoculated plates. The plates were placed in the incubator at 35±2°C and microaerophilic conditions for 72 h. To test antimicrobial susceptibility, we used the H. pylori control strain ATCC 43504 and the results were evaluated according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria [11]. Broth microdilution was performed in Brucella broth supple-mented with 2% fetal calf serum. Twofold dilutions of each antimicrobial agent, ranging from 0.75 to 256 mg l−1, were
used. The standardized inoculum was diluted to achieve a final inoculum concentration of approximately 5×105 c.f.u.
per well. The microtitre plates were incubated at 37 °C under microaerophilic conditions. MICs were read after 72 h of incubation [12].
detection of A2115G, G2141A and veA2144T point mutations in 23s rRnA of H. pylori strains by real-time PCR
Biopsies were homogenized using a Magna Lyser Homog-enizer (Roche Diagnostic, Basel, Switzerland). Deoxyribo-nucleic acid (DNA) was extracted using the QIAamp DNA Mini Kit (QiagenGmbH, Hilden, Germany) according to the manufacturer’s instructions.
In the clarithromycin-resistant strains, the A2115G, G2141A and A2144T point mutations were investigated using the following primers and hydrolysis probes (IDT; Integrated DNA tech., USA) in a LightCycler 480 II (Roche Diagnostics, Mannheim, Germany) (Table 1). In this method, the Light-Cycler 480 Probe Master (Roche Diagnostics Mannheim, Germany) kit was used according to the manufacturer’s instructions. The H. pylori 23S rRNA gene sequence was amplified by real-time PCR and the results were evaluated using the Endpoint genotyping analysis module in LightCy-cler 480 1.5 software. Thus, clarithromycin-resistant strains were determined based on the detection of A2115G, G2141A and A2144T mutations. The final volume was 20 µl after adding 5 µl of template DNA. The real-time PCR amplifi-cation consisted of an initial denaturation step at 95 °C for
10 min that was followed by 45 amplification cycles consisting of 95 °C for 15 s, annealing at 60 °C for 30 s, and extension at 72 °C for 15 s.
detection of A2143G, A2142C and A2142G point mutations by real-time PCR
In clarithromycin-resistant strains, the A2143G, A2142C and A2142G point mutations were investigated using the following primers and HybProbes (hybridization probes) (TıbMolbiol, Berlin, Germany) in a LightCycler 480 II (Roche Diagnostics, Mannheim, Germany) (Table 2). The LightCycler 480 Genotyping Master (Roche Diag-nostics Mannheim, Germany) kit was used for the detec-tion of the A2143G, A2142C and A2142G mutadetec-tions according to the manufacturer’s instructions. The final volume was 20 µl by adding 5 µl of template. Real-time PCR amplification consisted of an initial denaturation step at 95 °C for 10 min, followed by 45 amplification cycles consisting of 95 °C for 10 s, annealing at 60 °C for 15 s, and extension for 1 min, and a final step of cooling to 40 °C for 1 min. Results were evaluated using the Melting Curve analysis module in LightCycler 480 1.5 software.
Sequencing of the 23S rdnA
To verify the detected 23 rRNA mutations after real-time PCR, all the real-time PCR products were sequenced. Each real-time PCR product was separated on 2% agarose gels with ethidium bromide and a 100 bp ladder was used as the DNA molecular weight standard. Amplification fragments were purified from agarose gel using the QIAquick PCR Purifi-cation Kit (Qiagen GmBH, Hilden Germany) according to the manufacturer’s instructions and the pure amplification products were sequenced in two directions (with forward and reverse primers) using a BigDye Terminator Cycle Sequencing kit (Applied BioSystems, Foster City USA) on an ABI PRISM 3130 Genetic Analyzer (Applied BioSystems, Foster City USA) according to the manufacturer’s instruc-tions. The results were processed via Sequence Analysis Software (Applied Biosystems, Foster City USA). Alignment of single consensus sequences were carried out using clustal W. and FASTA files compared to the reference sequence for the 23S rRNA gene (GenBank Accession number: U27270.1).
Table 1. Primers and probes for the detection of the A2115G, G2141A and A2144T mutations
Primers/probes Oligonucleotide sequences (5′−3′) References
23SA2115G-2115F 5′-GCTGTCTCAACCAGAGATTCAGT-3′ De Francesco et al. [13],
23SA2115G-2115R 5′-CTTGCCGCGGGTAGGA-3′ De Francesco et al. [13],
VIC 5′-AAGACGGAAAGACC-3′ for wild-type DNA De Francesco et al. [13],
FAM 5′-AGGTGAGAATTCC-3′ for mutated DNA De Francesco et al. [13]
23SG2141A-2144F 5′-TCAGTGAAATTGTAGTGGAGGTGAAAA-3′ De Francesco et al. [13]
23SG2141A-2144R 5′-AGTGCTAAGTTGTAGTAAAGGTCCAC-3′ De Francesco et al. [13]
VIC 5′-CAAGACGGAAAGACC-3′ for wild-type DNA De Francesco et al. [13]
FAM 5′-CAAGACGAAAAGACC-3′ formutated DNA De Francesco et al. [13]
23SA2144T-2144F 5′-TCAGTGAAATTGTAGTGGAGGTGAAAA-3′ De Francesco et al. [13]
23SA2144T-2144R 5′-AGTGCTAAGTTGTAGTAAAGGTCCAC-3′ De Francesco et al. [13]
VIC 5′-AGACGGAAAGACCC-3′ for wild-type DNA De Francesco et al. [13]
FAM 5′-AAGACGGAATGACCC-3′ for mutated DNA De Francesco et al. [13]
Table 2. Primers and probes for the detection of the A2143G, A2142C and A2142G mutations
Primers/probes Oligonucleotides Nucleotide
positions Accession No. References HPY-S 5-AGGTTAAGAGGATGCGT CAGTC-3 1931–1952 U27270 Menard et al. [14] HPY-A 5-CGCATGATATTCCCATTAGC AGT-3 2175–2197 U27270 Menard et al. [14]
Red640 5-GGCAAGACGGAAAGACC-3 2504–2520 U27270 Oleastro et al. [15]
Statistical analysis
For the analyses of the variables, we used SPSS 25.0 (IBM Corporation, Armonk, New York, USA). The chi-squared test and Fisher’s exact test for multiple comparisons were used to compare subgroups (Table 3a–e). Results were defined as the Benjamini Hochberg-adjusted P-value. Binary logistic regression with an enter method was employed to assess the association of
point mutations and the likelihood of clarithromycin resistance in H. pylori strains (Table 4). All reported CIs were calculated at the 95% level. Significance was recognized when P<0.05.
Table 3. (a) Comparison of H. pylori strains with and without the A2115G point mutation due to clarithromycin resistance. (b) Comparison of H. pylori strains with and without the A2144T point mutation due to the clarithromycin resistance. (c) Comparison of H. pylori strains with and without the G2141A point mutation due to the clarithromycin resistance. (d) Comparison of H. pylori strains with and without the AG2142G point mutation due to the clarithromycin resistance. (e) Comparison of H. pylori strains with and without the AG2143G point mutation due to the clarithromycin resistance
Clarit. S Clarit. R Total P OR 95% C.I.
Lower Upper (a) A2115G mutation
A2115G A2115G (−) strains 37 (67.3%) 18 (32.7%) 55 0.045 6.167 1.130 33.639
A2115G (+) strains 2 (25%) 6 (75%) 8
(b) A2144T mutation
A2144T A2144T (−) strains 37 (68.5%) 17 (31.5 %) 54 0.012 7.618 1.430 40.592
A2144T (+) strains 2 (22.2%) 7 (77.8 %) 9
(c) G2141A mutation
G2141A G2141A (−) strains 36 (69.2%) 16 (30.8%) 52 0.015 6000 1405 25 623
G2141A (+) strains 3 (27.3%) 8 (72.7%) 11
(d) A2142G mutation
A2142G A2142G (−) strains 39 (70.9%) 16 (29.1%) 55 0.0001 19 2.07 439.54
A2142G (+) strains 0 (0%) 8 (100%) 8
(e) A2143G mutation
A2143G A2143G (−) strains 36 (73.5%) 13 (26.5%) 49 0.0004 10.15 2.11 55.36
A2143G (+) strains 3 (21.4%) 11 (78.6%) 14
C.I., confidence interval;Clarit. R, chlarithromycin resistant;Clarit. S, chlarithromycin susceptible.
Table 4. Clarithromycin resistance risk analysis in H. pylori strains according to the A2115G, A2144T and G2141A point mutations
B S.E. d.f P OR 95% C.I. for EXP(B)
Lower Upper A2142G −4.025 1.163 1 0001 56.00 5.726 547.646 A2143G −3,245 0.808 1 0 0001 25.667 5.268 125.059 Constant −1.946 0.478 1 0 0001 A2115G(1) −3,466 1016 1 0001 31,66 19 673 41 862 A2144T(1) −3,620 1004 1 0001 36,92 14 956 43 120 G2141A(1) −3,348 0907 1 0001 28,16 16 377 48 333 Constant 8066 1796 1 0001
RESuLTS
A total of 63 patients [33 (52.3 %) females and 30 (47.7 %) males] with a mean of age of 47.08±12.27 years (range: 20–70 years) were included in this study. The A2142C mutation was not detected in any of the H. pylori strains.
Clarithromycin MIC values in H. pylori strains
The clarithromycin MIC values of H. pylori strains were evaluated according to the EUCAST criteria. A total of 39
H. pylori strains were identified as clarithromycin susceptible
(61.9%) (<0.25 mg l−1) and 24 H. pylori strains were identified
as clarithromycin resistant (38.1%) (>0.5 mg l−1). A2142G, A2143G, A2115G, A2144T and G2141A point mutations detected in H. pylori strains
The distribution of point mutations in phenotypically resistant and susceptible H. pylori strains is shown in Table 5.
In the phenotypically clarithromycin-resistant strains (n:24), the new A2115G, A2144T and G2141A point mutations were detected in six (25%), seven (29.1%) and eight (33.3 %) patients, respectively. On the other hand, in the same strains (n:24), the classical A2142G and A2143G point mutations were detected in eight (33.3%) and eleven (45.8%) patients, respectively (Table 5). No A2115G/A2144T/G2141A point mutation or other combinations of new mutations were detected in any of the H. pylori strains. Hence, in the pheno-typically clarithromycin-resistant strains (n:24), the classical + new mutations were as indicated at the bottom of Table 5.
Relationship between clarithromycin resistance and the A2115G, A2144T and G2141A point mutations of H. pylori
The median MIC values of clarithromycin in phenotypi-cally clarithromycin-resistant H. pylori strains are shown in Table 6. The A2144T point mutation had the highest median MIC value (3 mg l−1) amongst the new mutations; the classical
mutations (A2142G and A2143G) had the highest median MIC values (256 mg l−1) overall.
Clarithromycin resistance was more common in H. pylori strains with the A2115G, A2144T, G2141A, A2142G and A2143G point mutations than in the strains without these point mutations (P=0.045, P=0.012, P=0.015, P=0.0001 and
P=0.0004, respectively) (Table 3a–e). The point mutations
were confirmed by sequencing analyses. In the binary logistic regression analysis of the H. pylori A2115G, A2144T and G2141A point mutations, the estimation of the prediction accuracy of the analysis was 84.1% and the prediction ratio of clarithromycin resistance for any of the three point mutations was 87.5% (P<0.001). The presence of the A2115G (OR:31.66, 95 % CI: 19.673–41.862, P=0.001), A2144T (OR:36.92, 95% CI: 14.956–43.120, P=0.001), and G2141A (OR:28.16, 95 % CI: 16.377–48.333, P=0.001) point mutations increased the likelihood of clarithromycin resistance in H. pylori strains by 31.66-, 36.92- and 28.16-fold (ORs) according to the binary logistic regression analysis (Table 4). In the binary logistic regression analysis of the H. pylori A2142G and A2143G point mutations, the presence of the A2142G (OR:56.00, 95 % CI:5,726–547,646, P=0.001) and A2143G (OR:25.667, 95 % CI:5.268–25.059, P=0.0001) point mutations increased the likelihood of clarithromycin resistance in H. pylori strains by 56.00- and 25.66-fold (ORs), respectively, according to the binary logistic regression analysis (Table 4).
dISCuSSIon
Amongst the major antibiotics, clarithromycin has the greatest in vitro bactericidal effect for the treatment of
H. pylori infections. However, primary clarithromycin
resist-ance is a common and serious public health issue and varies across geographical regions [16]. The eradication rates of
H. pylori infections are decreasing while the rates of antibiotic
resistance are increasing. There is a consensus that this inverse
Table 5. Distribution of point mutations in phenotyically resistant and susceptible Helicobacter pylori
Clarithromycin
resistance A2142G A2143G A2115G A2144T G2141A
Phenotypic resistance (n:24) 8 (33.3 %) 11 (45.8 %) 6 (25 %) 7 (29.1 %) 8 (33.3 %) Phenotypic susceptibility (n:39) 0 (0 %) 3 (7.6 %) 2 (5.1 %) 2 (5.1 %) 3 (7.6 %) Total (n:63) 8 (12.7 %) 14 (22.2 %) 8 (12.7 %) 9 (14.2 %) 11 (17.4 %)
In the phenotypically clarithromycin-resistant strains (n:24), the classical + new mutations were detected as follows; 2115+2143 point mutation; 2/24 (8,3%), 2141+2143 point mutation; 3/24 (12.5%), 2144+2143 point mutation; 3/24 (12.5%), 2142+2144 point mutation; 3/24 (12.5%), 2115+2142 point mutation; 2/24 (8,3%), 2141+2142 point mutation; 3/24 (12.5%)
Table 6. Median MIC values of 24 phenotypically chlarithromycin-resistant H. pylori strains according to the point mutations (n:40)
Point mutations (n) MIC values (mg l−1) median
(range) A2115G [6] 1.4 (0.75–256) G2141A [8] 1.25 (0.75–256) A2144T [7] 3 (0.75–256) A2142G [8] 256 (0.75–256) A2143G [11] 256 (8–256)
relationship is a result of clarithromycin resistance. Clarithro-mycin resistance rates vary worldwide and are reported as ∼30 % in Italy and Japan, ∼40 % in Turkey, ∼50 % in China and ∼15 % in Sweden and Taiwan. A threshold of 15% was advised to separate high and low clarithromycin resistance due to regional differences in clarithromycin resistance rates. To overcome clarithromycin resistance, susceptibility-based antimicrobial therapy provides optimal treatment for
H. pylori infections [17, 18].
In Greece and Bulgaria, clarithromycin resistance rates were reported as 20.8 and 31 %, respectively [19, 20]; however, in a study covering 18 European countries, the primary clarithromycin resistance rate for H. pylori was 17.5%, which was higher in Western/Central and Southern Europe (>20%) than in Northern European countries (<10%). The same European study also reported that clarithromycin resistance almost doubled over the past 10 years, from 9.8 to 17.5 % [21]. This clarithromycin resistance is mostly connected with the use of macrolides, and in Norway, where macrolides are rarely prescribed, the prevalence of clarithromycin resistance was lower than in other European countries (5.9%) [21, 22]. In Turkey, clarithromycin resistance rates vary between 41 and 28.5 % in northwestern and central Turkey, respectively, and 18.2% in southern Turkey [23].
In this study, the clarithromycin resistance rate was 38.1% (24/63), as assessed by the E-test method, which was higher than the proposed threshold of 15% [18]. Our results were higher than the results of European countries, but mostly in parallel with the results of previous Turkish studies. As stated earlier, the A2143G, A2142G and A2142C mutations in the region of domain V of the 23S rRNA are the most common mutations and are responsible for the majority of clarithro-mycin resistance cases [24]. Among these point mutations, the most common mutation is an A-to-G transition at posi-tion 2143 (A2143G) [25]; however, other mutaposi-tions, such as A2058G, A2059G, A2116G, A2115G, G2141A, G1939A, C2147G, T2182C, T2190C, C2195T, A2223G and C2694A, might also be responsible for clarithromycin resistance [16]. In this study, the A2142G mutation was detected in 12.6% of all H. pylori strains (8/63) and 33.3% of the clarithromycin-resistant H. pylori strains (8/24). On the other hand, the A2143G mutation was detected in 16.6% of all H. pylori strains (11/63) and 45.8% of the clarithromycin-resistant H.
pylori strains (11/24). We did not detect the A2142C
muta-tion in any of our cases. Our main aim was to evaluate the recently discovered A2115G, A2144T and G2141A point mutations in H. pylori strains. We detected these A2115G, A2144T and G2141A point mutations in 8, 9 and 11 of the 63 H. pylori strains. Later, we focused on clarithromycin-resistant H. pylori strains, and in 24 clarithromycin-clarithromycin-resistant strains, the prevalences of these mutations were 25% (6/24), 29.16% (7/24) and 29.16% (7/24) for the A2115G, A2144T and G2141A point mutations, respectively. In other words, 75% (6/8), 77.8% (7/9) and 77.8% (7/9) of H. pylori strains with the A2115G, A2144T and G2141A point mutations, respectively, were detected to be clarithromycin resistant.
Briefly, the clarithromycin resistance rates in the H. pylori strains with the A2115G, A2144T and G2141A point muta-tions were significantly higher than those of the H. pylori strains without these point mutations (P=0.045, 0.012 and 0.015, respectively). The presence of the A2115G, A2144T and G2141A point mutations increased the clarithromycin resist-ance rates by 31.66-, 36.92- and 28.16-fold (ORs), respectively, according to the binary logistic regression analysis. Out of the 24 clarithromycin-resistant strains, the prevalences of the A2142G and A2143G mutations were 8 (33.3%) and 11 (45.8 %), respectively. The clarithromycin resistance rates in the H. pylori strains with the A2142G and A2143G point mutations were significantly higher than those of the H. pylori strains without these point mutations (P=0.0001 and 0.0004, respectively). The presence of the A2142G and A2143G point mutations increased the clarithromycin resistance rates by 56.00- and 25.667-fold (ORs), respectively, according to the binary logistic regression analysis. The detected point muta-tions in the phenotypically clarithromycin-sensitive strains (n:39) showed that these point mutations did not lead to statistically significant differences in resistance (P>0.05). Since these mutations were also observed in phenotypically sensitive strains, at least two explanations could make this situation reasonable. One is that specimens demonstrating a point mutation with a susceptible phenotype may present a mixed bacterial population in the gastric specimens, i.e. the specimens may have both susceptible and resistant strains. While an E-test result may identify susceptible strains, point mutations in the resistant strains are indicated by molecular methods. A possible limitation of this study may be that it was not possible to discriminate between resistant and susceptible colonies in H. pylori cultures with mixed colonies (suscep-tible, resistant or both). Another explanation may be due to the structure of bacterial DNA. H. pylori has two copies of the 23S rRNA gene in its chromosomal DNA; therefore, a muta-tion in one gene copy may not be enough to enable resistance to clarithromycin. Therefore, the detected mutations in our strains with a genotypic resistance could not be detected by the E-test, but the point mutations in another copy may be enough for detection by molecular methods [26].
In an extensive literature search, we identified only one study that reported these new mutations, which was performed by de Francesco et al. [13] in Italy. These authors investigated the influence of six point mutations (A2143G, A2142G, A2142C A2115G, G2141A and A2144T) on phenotypic antimicrobial resistance in 82 H. pylori strains, and they detected clarithromycin resistance in 51.2% (42) of the 82
H. pylori strains. In their genotypic analysis, at least one of
the three most commonly detected point mutations (A2143G, A2142G and A2142C) was detected only in 23 (54.8%) strains. These authors detected 19 H. pylori strains with phenotypic clarithromycin resistance but without any of these classical three mutations, and in 14 of these 19 H. pylori strains, they detected three strains that carried A2115G, six with G2141A and four with A2144T, as well as one with the A2115G and A2144T mutations.
Van Doorn et al. [27] investigated the A2115G and G2141A point mutations, but they could not detect any of these point mutations in their samples. Therefore, we suggest that these new point mutations were not developed due to the effective usage of macrolides against H. pylori infections; however, the differences in the detection methods for the point muta-tions could have been responsible for their absence in their
H. pylori study strains.
In the study of de Francesco et al. [13], 66.6% (6/9), 100% (6/6) and 100% (6/6) of the H. pylori strains that carried the A2115G, G2141A and A2144T point mutations, respectively, were found to be clarithromycin resistant. Our results were a little smaller than their ratios for the G2141A and A2144T point mutations but higher than their reported rate for the A2115G point mutation. Our binary logistic regression results show that the presence of the A2115G, A2144T and G2141A point mutations increased the likelihood of clarithromycin resistance in
H. pylori strains by 31.99-, 37.33- and 28.44-fold,
respec-tively. However, the presence of the A2142G and A2143G point mutations increased the likelihood of clarithromycin resistance in H. pylori strains by 56- and 25.66-fold, respec-tively, in a separate binary logistic regression analysis. Therefore, we suggest that there is a high risk for failure with standard H. pylori treatment regimens, including clarithromycin, due to the results of the binary logistic regression analysis, especially for the A2144T point muta-tion with its 37.33-fold (OR) increased clarithromycin resistance risk. However, we also detected the A2115G, G2141A and A2144T point mutations in 2/8, 2/9 and 2/8 clarithromycin-susceptible H. pylori strains, respectively. There are multiple ways to account for their role for their ineffectiveness in generating clarithromycin resistance, such as being silenced. On the other hand, classical muta-tions, such as A2142G and A2143G, resulted in greater rates of clarithromycin resistance [33% (8/24) and 45.8% (11/24)] than the new mutations [A2115G, 25% (6/24); A2144T, 29.16% (7/24); and G2141A 29.16% (7/24)]. de Franceska et al. [13] compared the impact of new point mutations (A2115G, G2141A and A2144T) with that of the classical point mutations (A2143G and A2142G) on the MICs of clarithromycin in phenotypically clarithromycin-resistant H. pylori strains. These authors concluded that no differences were present when comparing the ‘classical’ and ‘novel’ mutations and proposed that the impact on the MICs of clarithromycin and phenotypic resistance might result from multiple synergic genetic mechanisms, rather than being the simple consequence of a specific point mutation [16]. In this study, the median MIC values of the classical A2142G and A2143G mutations were higher than those of the novel A2115G, A2144T and G2141A point mutations. In contrast, in our study, the median MIC values of the new A2115G, A2144T and G2141A point mutations were lower than those of the classical point mutations (1.4, 1.25 and 3 mg l−1, respectively). It was not
possible to make a binary logistic regression analysis that included all the point mutations in one analysis due to
the small number of samples with A2142G point muta-tions. Therefore, it is not possible to make a conclusion due to the binary logistic regression analysis results, but only to make a suggestion by considering these median MIC values that the classical mutations resulted in higher clarithromycin MIC values than the new mutations. Our results do not agree with those of de Francisca et al. [13] but are in line with other studies that have suggested that high MIC values are associated with either the A2143G or A2142G point mutation [25, 28, 29]. On the other hand, if we consider the combination of the classical and new point mutations together, it is also difficult to explain why there are at least ten strains with these mutations but with no clarithromycin resistance. There might also be other defined or undefined point mutations (A2142G, A2142C, A2143G, A2058G, A2059G, A2116G, G1939A, C2147G and T2182C) or other resistance mechanisms that contribute to clarithromycin resistance in H. pylori strains. In conclusion, we suggest that for the detection of clarithromycin resistance, genotypic methods should search for point mutations other than the classical muta-tions. We concluded that classical mutations resulted in higher clarithromycin MIC values compared to the new mutations. These new point mutations caused a moderate elevation in the MIC values of clarithromycin-resistant
H. pylori strains. Therefore, in order to clarify the exact
mechanisms behind clarithromycin resistance in H. pylori strains, we now need more comprehensive and large-scale studies that include analyses of point mutations.
Funding information
This work was supported by the Istanbul University Research Fund under project number 20 900. We are grateful for this support. Conflicts of interest
The authors declare that there are no conflicts of interest. references
1. hooi JKY, Lai WY, Ng WK, suen MMY, Underwood FE et al. Global prevalence of Helicobacter pylori infection: systematic review and meta-analysis. Gastroenterology 2017;153:420–429.
2. Parsonnet J, Friedman Gd, Vandersteen dP, Chang Y, Vogelman Jh et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325:1127–1131. PMID: 1891020.
3. shinzaki s, Fujii t, Bamba s, ogawa M, Kobayashi t et al. Seven days triple therapy for eradication of Helicobacter pylori does not alter the disease activity of patients with inflammatory bowel disease. Intest Res 2018;16:609–618.
4. rimbara E, Fischbach LA, Graham dY. Optimal therapy for Helico-bacter pylori infections. Nat Rev Gastroenterol Hepatol 2011;8:79–88. 5. Graham dY, Fischbach L. Helicobacter pylori treatment in the era of
increasing antibiotic resistance. Gut 2010;59:1143–1153.
6. hultén K, Gibreel A, sköld o, Engstrand L. Macrolide resistance in Helicobacter pylori: mechanism and stability in strains from clarithromycin-treated patients. Antimicrob Agents Chemother 1997;41:2550–2553. PMCID: PMC164161.
7. Ahmad N, Zakaria Wr, Abdullah sA, Mohamed r. Characterization of clarithromycin resistance in Malaysian isolates of Helicobacter pylori. World J Gastroenterol 2009;15:3161.
8. toracchio s, Aceto GM, Mariani-Costantini r, Battista P, Marzio L. Identification of a novel mutation affecting domain V of the 23S rRNA gene in Helicobacter pylori. Helicobacter 2004;9:396–399.
9. de Francesco V, Zullo A, hassan C, Giorgio F, rosania r et al. Mech-anisms of Helicobacter pylori antibiotic resistance: an updated appraisal. World J Gastrointest Pathophysiol 2011;2:35–41. 10. Fontana C, Favaro M, Minelli s, Criscuolo AA, Pietroiusti A et al.
New site of modification of 23S rRNA associated with clarithro-mycin resistance of Helicobacter pylori clinical isolates. Antimicrob Agents Chemother 2002;46:3765–3769.
11. the European Committee on Antimicrobial susceptibility testing (EUCAst). Breakpoint tables for interpretation of MICs and zone diameters (version 6.0) 2013. Available at: http://www. eucast. org/ clinical_ breakpoints/ (Accessed 6 August, 2018).
12. tharmalingam N, Port J, Castillo d, Mylonakis E. Repurposing the anthelmintic drug niclosamide to combat Helicobacter pylori. Sci Rep 2018;8:3701. PMID: 29487357.
13. de Francesco V, Zullo A, Giorgio F, saracino i, Zaccaro C et al. Change of point mutations in Helicobacter pylori rRNA associ-ated with clarithromycin resistance in Italy. J Med Microbiol 2014;63:453–457.
14. Ménard A, santos A, Mégraud F, oleastro M. PCR-restriction fragment length polymorphism can also detect point muta-tion A2142C in the 23S rRNA gene, associated with Helicobacter pylori resistance to clarithromycin. Antimicrob Agents Chemother 2002;46:1156–1157.
15. oleastro M, Ménard A, santos A, Lamouliatte h, Monteiro L et al. Real-time PCR assay for rapid and accurate detection of point mutations conferring resistance to clarithromycin in Helicobacter pylori. J Clin Microbiol 2003;41:397–402.
16. Agudo s, Pérez-Pérez G, Alarcón t, López-Brea M. High preva-lence of clarithromycin-resistant Helicobacter pylori strains and risk factors associated with resistance in Madrid, Spain. J Clin Microbiol 2010;48:3703–3707.
17. Malfertheiner P, Megraud F, o'Morain CA, Gisbert JP, Kuipers EJ et al. European Helicobacter and microbiota study group and consensus panel. Management of Helicobacter pylori infection-the Maastricht V/Florence consensus report. Gut 2017;66:6–30.
18. thung i, Aramin h, Vavinskaya V, Gupta s, Park JY et al. Review article: the global emergence of Helicobacter pylori antibiotic resistance. Aliment Pharmacol Ther 2016;43:514–533.
19. Georgopoulos sd, Xirouchakis E, Martinez-Gonzalez B, sgouras dN, spiliadi C et al. Clinical evaluation of a ten-day regimen with esomeprazole, metronidazole, amoxicillin, and clarithromycin for the eradication of Helicobacter pylori in a high clarithromycin resistance area. Helicobacter 2013;18:459–467.
20. Boyanova L, Mitov i. Geographic map and evolution of primary Heli-cobacter pylori resistance to antibacterial agents. Expert Rev Anti Infect Ther 2010;8:59–70.
21. Megraud F, Coenen s, Versporten A, Kist M, Lopez-Brea M et al. Helicobacter pylori resistance to antibiotics in Europe and its rela-tionship to antibiotic consumption. Gut 2013;62:34–42.
22. Larsen AL, ragnhildstveit E, Moayeri B, Eliassen L, Melby KK. Resistance rates of metronidazole and other antibacterials in Helicobacter pylori from previously untreated patients in Norway. APMIS 2013;121:353–358.
23. Caliskan r, tokman hB, Erzin Y, saribas s, Yuksel P et al. Antimi-crobial resistance of Helicobacter pylori strains to five antibiotics, including levofloxacin, in Northwestern Turkey. Rev Soc Bras Med Trop 2015;48:278–284.
24. Monno r, Giorgio F, Carmine P, soleo L, Cinquepalmi V et al. Heli-cobacter pylori clarithromycin resistance detected by Etest and TaqMan real-time polymerase chain reaction: a comparative study. APMIS 2012;120:712–717.
25. de Francesco V, Zullo A, ierardi E, Vaira d. Minimal inhibitory concentration (MIC) values and different point mutations in the 23S rRNA gene for clarithromycin resistance in Helicobacter pylori. Dig Liver Dis 2009;41:610–611.
26. Binyamin d, Pastukh N, on A, Paritsky M, Peretz A. Phenotypic and genotypic correlation as expressed in Helicobacter pylori resist-ance to clarithromycin and fluoroquinolones. Gut Pathog 2017;9:1. 27. van doorn LJ, Glupczynski Y, Kusters JG, Megraud F, Midolo P
et al. Accurate prediction of macrolide resistance in Helicobacter pylori by a PCR line probe assay for detection of mutations in the 23S rRNA gene: multicenter validation study. Antimicrob Agents Chemother 2001;45:1500–1504.
28. owen rJ. Molecular testing for antibiotic resistance in Helicobacter pylori. Gut 2002;50:285–289.
29. Mégraud F. H pylori antibiotic resistance: prevalence, importance, and advances in testing. Gut 2004;53:1374–1384.
