Antibiotic resistance of Helicobacter pylori in Mongolia

Introduction.The resistance of Helicobacter pylori to recently available antibiotic treatment regimens has been recognized as a growing problem. Therefore, the aim of this study was to determine the prevalence of antibiotic resistance among H. pylori strains isolated from Mongolians. Methodology. All gastric biopsy specimens were obtained during upper gastrointestinal endoscopy from patients referred for the exploration of dyspepsia. The urease positive samples by rapid urease test were cultured according to standard microbiological procedures and H. pylori were grown under microaerophilic conditions on selective Pylori agar. H. pylori antibiotic sensitivity was examined using E-test. In addition, the mutations of the corresponding gene were studied by GenoType HelicoDR DNA strip testing. Results. Three hundred twenty patients, 216 female and 104 male in the ages range of 18 to 83 years were included in this study. Rapid urease test yielded positive results for 65.9% (211/320). Among them, we have successfully obtained 72% H. pylori isolates. The antibiotic resistance rates were 35.5% for clarithromycin, 68.4% metronidazole, 23.0% amoxicillin, 25.0% tetracycline, 28.2% erythromycin and 14.5% nitrofuranton. Resistance for 2 drugs was 34.5% and that of 3 drugs was observed in 14.5% of isolates. The most prevalent mutation was A2147G followed by A2146G and D91Y. The prevalence of H. pylori infection increased among Mongolian population and the prevalence of resistance of H. pylori is very high to metronidazole, and moderate to clarithromycin. Conclusion. The data on antimicrobial susceptibilities provided by the present study is may assist the clinicians on the effectiveness of treatment regimens.


Introduction
Helicobacter pylori is a Gram-negative bacterium associated with various digestive diseases, such as gastritis, peptic ulcer, mucosa-associated lymphoid tissue lymphoma, and gastric cancer [1]. Current recommendations for the management of H. pylori infection were elaborated by the European Helicobacter Study Group (EHSG) and presented in Maastricht IV/Florence Consensus Report in 2012 [2]. Treatment regimens containing a PPI and combination of 2 or more antibiotics, including amoxicillin (AMX), clarithromycin (CLR), metronidazole (MNZ) or tetracycline (TET) are considered to be most efficacious [3]. However, the cure rate of H. pylori has been decreasing progressively, primarily due to increased resistance to antimicrobial agents. Increased resistance to CLR is a growing worldwide problem [4]. Prevalence of bacterial resistance to antibiotics varies in different geographic areas and it has been correlated with the consumption of antibiotics in the general population [4,5]. For example, MNZ resistance varies from 10 to 80% among geographic regions [6][7][8][9]. In developing countries antibiotic resistance is considered to be higher than in developed countries. CLR is found to be one of the most effective antimicrobial agents used for the treatment of H. pylori infection. CLR resistance in H. pylori is due to point mutations in the rrl gene encoding 23S rRNA, with three major mutations described: A2146C, A2146G, and A2147G [4]. In Mongolia, no domestic guidelines are available for the treatment of H. pylori because of insufficient domestic data. The aim of this study was to determine the prevalence of antibiotic resistance among H. pylori strains isolated from Mongolians.

Methodology Patients
A total of 320 consecutive patients, who visited Shastin hospital and SonginoKhairkhan district hospital in Ulaanbaatar for upper endoscopy during 2011-2014 were enrolled in the study. All subjects provided informed consent, and the study protocol was approved by the Ethics Committee at Mongolian National University of Health. The information provided in the pathology reports or patients' files were recorded for each patient, which included patient's hospital ID number, age, gender, medical history, clinical diagnosis based on endoscopy according to Sydney classification 1994, and previous treatment. Patients who received bismuth compounds, antibiotics during the 4 weeks before endoscopy were not included in the study. Other exclusion criteria also included regular use of nonsteroidal anti-inflammatory drugs including acetylsalicylic acid, malignancy, and severe liver diseases.

Endoscopic study
All patients included in the study underwent upper gastrointestinal endoscopy. Two sets of gastric biopsy specimens were obtained from the antrum and body in all patients and one set was tested for H. pylori using rapid urease test and the other specimen was used for H. pylori culture.

Isolation of H. pylori
Biopsy specimens from the antrum and body were used for H. pylori culture. Biopsy specimens were macerated and homogenized in meat liver dextrosa broth and a 250 μL aliquot was inoculated on selective Pylori Agar (Biomerieux, Marcy-l'Etoile, France). Incubation was performed in microaerophilic (5% O2, 15% CO2, and 80% N2) conditions (Genbox and Genbagmicroaerophilic. Biomerieux, Marcy-l'Etoile, France) at 37°C for a maximum of 5-7 days. Colonies were identified as H. pylori according to standard criteria including Gram-negative bacteria with typical cell morphology, and positive reactions to catalase, oxidase, and urease.

Antimicrobial susceptibility testing
The susceptibility of H. pylori isolates to CLR, MNZ, AMX, TET, nitrofurantoin (NIF) and erythromycin was examined by E-test strip (Biomerieux, Marcy-l'Etoile, France). The bacterial suspensions were spread onto Mueller-Hinton II agar plates (Becton Dickinson, New Jersey, USA) supplemented with 7% defibrinated sheep blood by sterile cotton swabs. After drying, each E-test strip of the corresponding antibiotic was placed on separative plate and all plates were incubated in anaerobic jar for 3 days at 37°C under microaerophilic conditions (GENbox microaer, Biomérieux, Marcy-l'Etoile,  [10].

PCR Method
Amplification of bacterial DNA was performed using hot-start DNA polymerase (Hain Lifescience, Nehren, Germany). Biotinylated primers were used for this study and were provided in the amplification kit (Hain Lifescience, Nehren, Germany). Polymerase chain reaction (PCR) for a single mixture had a final volume of 50 L containing 35 L primer/nucleotide mix (PNM), 5 L 10 ×polymerase incubation buffer, 2 L of 1.5mM MgCl2, 3 L of nuclease free water 0.2 L Thermo-Start Taq DNA polymerase (1-2 units were added to each tube), and 5 L DNA template. PCR was performed with a thermal cycler (Applied Biosystem, Foster city, USA). In protocols, the denaturation cycle was 1 cycle at 95°C for 15 min, followed by 10 cycles at 95°C for 30 s and at 58°C for 2 min. Then, 20 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 ended with 8 min at 70°C. Hybridization was performed using the TwinCubator at a temperature of 45°C. The denaturation solution was mixed with 20 L of the amplified sample and submitted to the usual protocol for hybridization.

GenoTypeHelicoDR Analysis
Confirmation of isolates as H. pylori, antimicrobial susceptibility, and mutational analysis to clarithromycin was performed using the GenoTypeHelicoDR kit (Hain Lifescience, Nehren, Germany). The kit employs the use of reverse hybridization performed using hybridization trays and Twin-Cubator (Hain Lifescience, Nehren, Germany) according to the manufacturer's instructions. Briefly, 20 L of amplified DNA was denatured and added to biotinylated probes on the strip and the hybrids formed were detected by enzyme linked immunosorbent assay (ELISA) upon addition of enzyme conjugate and substrate. Four gyr87 wild type probes (gyr87WT1-gyr87WT4) and one mutant probe (gyr87MUT), one wild type probe (gyr91WT1), and three mutant probes (gyr91MUT1-gyr91MUT3) were used for detecting fluoroquinolone resistance at position 87 and 91, respectively. For CLR, one wild type probe (23SWT) and three mutant probes (23SMUT1-23SMUT3) were used for detecting resistance. There were designated conjugate control (CC), amplification control (AC) and H. pylori (HP). The presence of a band at CC and AC meant that the conjugate control and amplification control were in the right frame while at HP implied presence of H. pylori according to the manufacturer's instruction (Hain Lifescience, Nehren, Germany).

Statistical analysis
Statistical analysis was performed using Chi-square and Fisher's exact tests. Null hypotheses of no difference were rejected if p-values were less than 0.05.

Patients
Three hundred twenty patients including 216 female and 104 male, with median age of 43.7 years ranged from 18 to 83 years who underwent upper gastroendoscopy were recruited in this study. Relationship between H. pylori infection, age and gender of the patients are presented in Table 1 The prevalence of CLR resistance in male and female was 32.1% vs. 37.5%. In general, there was a higher prevalence of resistant isolates in female compared to male patients. However, this did not reach statistical significance (p = 0.301). The frequency of CLR resistance was found in patients with gastritis (n = 23, 42.5%), stomach erosion (n = 13, 24%), gastric ulcer and atrophic gastritis (n = 7, 12.9%) and nodularity (n = 4, 7%). CLR resistance rate within the age group of 18-39 years was higher than other age groups of male patients. However, female was similar in all age groups in terms of CLR resistance rate.
Overall, 68.4% (n =117) H. pylori culture-positive patients had isolates that were fully resistant to MNZ. Male subjects were more likely to carry MNZ resistant H. pylori isolates than female subjects. However, this did not reach statistical significance (p = 0.543) MNZ resistance rate within age group of 30-39 years was higher than other age group of male. Female was 40-59 age group was higher than other age group. However, this did not reach statistical significance (p = 0.605).

Analysis of GenoType HelicoDR test
The MUT and WT probes were designed from the mutations observed in the resistant strains such as mutations in the rrl gene encoding the 23S rRNA for the CLR-resistant strains. Probes hybridized on the DNA strip of the GenoType HelicoDR test for detection of mutations in the rrl genes and gyrA listed in Table 3 [11]. The method of GenoType HelicoDR for 23S rNA (rrl) genotyping had been used to analyze the qualifying full growth of 95 samples.
Out of 95, 64 (67.3%) H. pylori isolates hybridized with wild type probe of 23S rRNA gene of CLR. All CLR resistance isolates had at least one of the three common point mutation in 23SrRNA gene, while none of the CLR susceptible isolates had this mutation. Table 4 shows the distribution of MUT3 in 51 CLA-R strains isolated from consecutive patients based on age and gender. Nineteen out of 32 (59.3%) MUT3 positive strains and fourteen out of 19 MUT3 negative strains isolated from the female population had this point mutation. Eleven out of 32 of the MUT3 positive strains were isolated from the most frequent age group i.e. 50-59 (Table 4). There was no significant relation between gender (p = 0.301) and age with this mutation.
Overall, the most frequent mutation was A2147G (MUT3 profile), observed in 32 strains (62.7%) of the mutated alleles. The frequency and rate of mutations encoding clarithromycin resistance in H. pylori isolates is shown Table 5. Eight percent of CLR resistance isolates (4 out of 51 isolates) had A2146G point mutation. We did not detect A2146C (MUT2 profile) in any of clarithromycin resistant strains. MUT3 was the most frequently detected mutation in clarithromycin resistant strains (p = 0.001).

Rate of multidrug resistance
Of the 152 strains, 16 (10.5%) showed no resistance to any antibiotics. Table 7 shows that rate of multidrug resistance in H. pylori in Mongolia. Resistance for one drug was observed in 28.2% (95% CI 25.7-43.2) isolates, for two drugs 34.5% (95% CI 25.7-43.2) and for 3 drugs was 14.4% (95% CI 8.6-19.7). Within the strains tested for susceptibility to all 6 antibacterial agents, H. pylori resistance for 4 or 5 drugs was detected in 5-6% whilst all 6 agents resistance was observed in one strain. The incidence of multiple drug resistance of H. pylori isolates is listed in Table 7. The first choice of treatment for H. pylori infection was CLR + MNZ for 9.4% and CLR + AMX for 5.6% cases. The second choice and alternatives to these medications were MNZ + AMX 20.7% and MNZ + TET had 9.4% resistance rate. It has been shown that a popular use of MNZ + AMX combination to eradicate H. pylori affects the result of treatment negatively.

Discussion
This is the first study exploring the antibiotic resistance pattern of H. pylori strains isolated from Mongolian population. In Mongolia, no domestic guidelines are available for the treatment of H. pylori because of insufficient data.
Although the prevalence of H. pylori in developed countries is decreasing, gastric colonization by H. pylori remains widespread in developing countries. Infection with H. pylori can be diagnosed by a variety of tests and can often be successfully treated with antibiotics [12]. The prevalence of H. pylori strongly varies between developing and developed countries, where the prevalence among adult is typically around 80-90% and <40% respectively [13]. In this study, we performed 320 endoscopies in patients with upper gastrointestinal symptoms and confirmed that the 65.9% had gastric H. Pylori infection using rapid urease test. Antimicrobial resistance varies by geographical region and is highly influenced by patterns of antimicrobial use within a population [4,5,14]. The key determinants of the outcome of eradication therapy for H. pylori infection are compliance and the presence of pretreatment antibiotic resistance of the isolate [15]. In this study we reported the susceptibility of H. pylori strains from Mongolia, a country with a high prevalence of infection, to the most commonly used antibiotics. Our results revealed that rate of antimicrobial susceptibility were 35.5% for CLR, 68.4% for MNZ, 23% for AMX, 25.0% for TET, 28.2% for ERY and 14.5% for nitrofurantoin. CLR is one of the core antibiotics used in PPI triple régimen [3,16]. Many researchers found that resistance to CLR is critical to the effectiveness of H. pylori eradication with triple therapy [17]. Prevalence of cLR resistance has been studied widely. It ranges from close to nil to 25% [10]. In Asian countries, high prevalence of CLR resistance was detected in Japan (40.7%) whereas the lowest (2.1%) prevalence rate was seen in Malaysia [18]. In the present study, resistance rate to CLR was 35.5%. In fact, the Maastricht III guidelines on H. pylori infection management recommend that CLR should not be used when resistance to the antibiotic exceeds 15-20% [19]. Therefore, CLR-based triple therapy might not be helpful to eradicate H. pylori in Mongolia.
The assay used in this study was designed to target the presence of A2147G, A2146G, and A2146C mutations associated with CLR resistance [11]. We detected that most frequent mutation was A2147G (MUT3 profile) in our studied isolates. Also, the strips in the GenoTypeHelicoDR assay are designed to generally target fluoroquinolones. The codons N87 and D91 are recognized as the most important target sites for ciprofloxacin binding. The N87H, N87I, N87K, and N87Y as well as D91G, D91N, and D91Y mutations in gyrA have been reported in fluoroquinolone-resistant H. pylori strains [20]. The assay used in this study was designed to depict N87K, D91N, D91G, and D91Y, which have been frequently reported. N87K was the most prevalent mutation (10/95; 10.5%) associated with fluoroquinolone resistance.
In this study, the prevalence of MNZ resistance was high (66.4%). Whereas MNZ resistance has been reported in 10-50% of all adult patients infected with H. pylori in developed countries, virtually all strains from developing countries have been found to be MNZ resistant [21], which is in agreement with our data that reports prevalence of MNZ resistance in 66.4% strains. In general, the high prevalence of MNZ resistance in developing countries is probably because of the frequent use of MNZ derivatives for the treatment of protozoal infections and gynecological problems [22], which is also common in Mongolia.
In the present study, AMX resistance was 23%. China, India, Mexico, and Italy have also reported variable prevalence rates for AMX resistance, i.e., 41.2%, 65%, 19.4%, and 45% respectively [23,24]. Similarly, it is difficult to explain variation of AMXresistance in different parts of the country as well as throughout the world. The prevalence of AMX resistance was probably a result of indiscriminate use of this antimicrobial agent, because of the lack of clearly defined guidelines for the management of H. pyloriassociated dyspepsia and other infections.
Multidrug resistance of H. pylori is occasional and found in individual countries or regions, for example in Taiwan [25]. A Bulgarian study reported that triple resistance to the evaluated agents was uncommon and detected in 3.5% of the untreated adults and 13.6% of the treated adults. Five H. pylori strains were resistant to AMX, MNZ and CLR, two of them exhibiting quadruple resistance. Resistance to four of the five antibacterials tested was found in 0.7% of the untreated and 1.8% of the treated adults [26]. Quadruple resistance of H. pylori has not been reported in Europe and the USA in last 5 years, although a study from India reported this type of resistance in 2.6% of isolates [27]. In the present study, resistance for 1 drug was observed in 28.2%, for 2 drugs was 34.5% and that for 3 drugs was observed in 14.4% of isolates. While H. pylori resistance for 4 or 5 drugs was detected in 5-6% strains.

Conclusions
This study shows that the prevalence of H. pylori infection was high among Mongolian population. We also report prevalence of resistance was high to MNZ and moderate to CLR. In addition, multidrug resistant strains were frequently found. CLR and fluoroquinolones resistance was mainly associated with A2147G and N87K mutations, respectively. We also demonstrates the need for continuous monitoring of the antimicrobial susceptibility in H. pylori to determine optimal treatment regimen.