Sage Journals: Discover world-class research

Abstract

The eradication of Helicobacter pylori reduces the risk of gastric cancer. A clear understanding of the factors underlying mixed infection with multiple clarithromycin-susceptible and clarithromycin-resistant H. pylori strains is necessary to design more effective therapies against H. pylori. We aimed to assess how the abundance and prevalence of H. pylori strains vary after clarithromycin-based eradication therapy. Using gastric wash samples, which represent the entire stomach, we sequentially analyzed the abundance and prevalence of H. pylori DNA by 23S ribosomal RNA pyrosequencing before and 1, 2, and 3 years after eradication therapy. Low levels of H. pylori DNA were still detectable at the first-year follow-up in all samples with negative post-treatment urea breath test results. The abundance of H. pylori DNA decreased significantly until the 2-year follow-up, but it switched to an increase at the 3-year follow-up. Importantly, the ratio of the prevalence of mutant strains to the prevalence of wild-type strains had already increased at the first-year follow-up and continued to increase, suggesting the selection and growth of clarithromycin-resistant strains during the follow-up periods. Being sensitive and representative, our assay will be useful in effectively addressing gastric cancer development by enhancing the long-term success of intervention strategies and consecutive surveillance for H. pylori eradication.

Keywords

mixed infection gastric wash 23S ribosomal RNA pyrosequencing

Introduction

The eradication of Helicobacter pylori infection reduces the risk of recurrent peptic ulcer diseases and the development of gastric adenocarcinoma.^1–6 It is now widely accepted that clarithromycin (CAM) resistance among H. pylori strains has been increasing worldwide and is the most important factor responsible for the failure of H. pylori eradication.^6–19 Various molecular methods have been developed to detect CAM resistance in H. pylori isolates in clinical specimens, such as biopsy samples and feces.^13,15,16 These assays for detecting CAM resistance in H. pylori are based on the detection of mutations in the 23S ribosomal RNA (rRNA) gene.

In an earlier study, 39% of patients with CAM-resistant H. pylori were reported to be infected with both CAM-resistant and CAM-susceptible H. pylori.²⁰ Frequent mixed infections with multiple H. pylori strains in the human stomach have also been reported.^21–25 Because mixed infections are common and one of the reasons for unsuccessful therapy, an accurate quantitative test for the presence of CAM-resistant H. pylori is necessary for successful eradication therapy. However, gastric biopsy is a topical procedure in which only a small portion of tissue is excised during endoscopy. One study showed two different H. pylori strains to have variable distribution among biopsy sites in a H. pylori-infected individual,²² which indicates that biopsy-based analysis might underestimate mixed infection detection.

In contrast, gastric wash samples are collected from the whole stomach using saline during endoscopy.²⁶ Because abundant cells are exfoliated from the gastric mucosa during washing, undamaged human and H. pylori DNA can be recovered from the wash and be assayed by sensitive quantitative techniques.^26–28 We recently developed a pyrosequencing-based method for the quantification of H. pylori genotypes using gastric washes and found that many patients carried mixed infection with CAM-susceptible (wild-type) and CAM-resistant (A2143G or A2144G mutant) H. pylori.²⁹ These data suggest that our method can detect mixed infections with higher sensitivity than biopsy-based methods, because H. pylori DNA can be extracted from gastric washes, which reflect the whole stomach.

In the case of mixed infections, CAM-resistant H. pylori may be selected for survival after CAM-based eradication therapy, which might result in the failure of therapy. However, it remains unknown how the abundance and prevalence of H. pylori strains vary after eradication therapy. A better understanding of mixed infection with multiple H. pylori strains is necessary to design more effective therapies against H. pylori. In this prospective study, using gastric wash samples from patients with early gastric cancer who had undergone endoscopic resection, we sequentially analyzed the abundance and prevalence of H. pylori DNA by 23S rRNA pyrosequencing before and 1, 2, and 3 years after CAM-based triple eradication therapy.

Materials and methods

Clinical samples

In this prospective study, 19 patients with early gastric cancer who had undergone endoscopic resection were diagnosed with H. pylori infection at the time of endoscopic resection. All the patients (13 males and 6 females) were diagnosed as H. pylori infection–positive by the H. pylori IgG assay, rapid urease test, or urea breath test (UBT). The patients received CAM-based triple eradication therapy. Successful eradication was confirmed by negative results of the UBT 1 month or more after the eradication. Patients were followed-up annually by endoscopic evaluation and collection of gastric wash samples at 1 (n = 14), 2 (n = 14), and 3 (n = 13) years after endoscopic resection and H. pylori eradication therapy. Several patients could not be followed-up annually by endoscopic evaluation at all three timepoints because of non-medical reasons or medical reasons other than gastrointestinal disease. Samples were immediately frozen at −80°C and DNA was extracted as previously reported. The study was conducted in accordance with all the rules and regulations of the St. Marianna University School of Medicine Institutional Review Board (#1929), and written informed consent was obtained from each patient (Table 1).

Table 1.

Clinical characteristics of gastric cancer patients and time course of mixed Helicobacter pylori infection.

No.	Age	Sex	Pre HP-IgG	Pre RUT	Pre UBT	Post UBT	Histology^a	Smoking	2143G^b (%) Pre	1 year	2 years	3 years	2144G^b (%) Pre	1 year	2 years	3 years
1	73	M	64			0.5	M/M	+	0	24	24	27	1	26	28	30
2	79	M	36			0.3	M/M	+	0	23			1	27
3	76	F	28			0.1	M/M	N/A	0	27	32	25	1	35	45	36
4	72	M	12			2.1	M/M	+	14		30		18		41
5	80	F	18			1.3	M/M	−	0	32	31		1	34	39
6	67	M	43			0.0	M/M	−	0	35	28		1	50	38
7	69	M	15			1.6	M/M	+	0	22	31	30	2	22	46	37
8	62	M	80			0.4	M/M	+	69	34			94	60
9	72	M			17.3	1.7	M/M	+	0		48	45	12		82	73
10	75	M		+		0.2	M/M	+	10	53	40		35	89	70
11	74	F	30			1.3	M/M	−	49	61		50	85	91		79
12	51	M	18			0.7	M/M	−	0	40	55	52	7	75	89	83
13	78	M		+		0.5	M/M	−	5		57	59	30		84	83
14	70	F	>100			0.7	M/M	+	0		61	62	11		91	90
15	69	M	69			0.0	M/M	+	19	57	61	61	78	87	85	84
16	68	M	19			0.3	M/M	+	31	64	65	61	63	65	93	94
17	60	F	54			0.4	M/M	−	60	63		56	90	90		81
18	69	F	13			1	M/M	−	21		60	62	54		95	92
19	73	M	22			1.5	M/M	+	41	66		68	74	90		96

M: male; F: female; HP-IgG: H. pylori immunoglobulin G; RUT: rapid urease test; UBT: urea breath test.

Histology (Pre/Post): marked/marked inflammation.

Time course of the ratio of the prevalence of A2143G or A2144G to that of wild-type H. pylori.

Sample collection of gastric washes

Approximately 10 min prior to endoscopy, patients were asked to swallow 100 mL water containing 80 mg dimethylpolysiloxane (GASCON; Kissei Pharmaceutical Co., Ltd., Matsumoto, Japan), 1 g sodium bicarbonate, and 20,000 units pronase (Pronase MS; Kaken Pharmaceutical Co., Ltd., Tokyo, Japan). Gastric washes were collected in specimen containers (No. 111219; Forte Grow Medical, Tochigi, Japan), which were directly fitted to the endoscope modulator. Gastric washes were manually aspirated under vacuum through the suction channel of the endoscope and centrifuged immediately, and resulting pellets were frozen at −80°C. Disposable sample collection tubes, connector tubes, and other endoscopic devices were used. The endoscope was washed after each procedure according to the guidelines, using 0.55% DISOPA (Johnson and Johnson, Langhorne, PA, USA) and an automatic washing machine.

DNA extraction

DNA was extracted from gastric wash samples by phenol chloroform. Briefly, frozen pellets from gastric washes were resuspended in 10% sodium dodecyl sulfate (SDS) and proteinase K, mixed vigorously, and incubated at 37°C for 6 h. Samples were then mixed vigorously for approximately 30 s with one volume of 25:24:1 phenol:chloroform:isoamyl alcohol and centrifuged at room temperature for 5 min at 16,000×g. The aqueous layer was then transferred to a fresh tube and extracted in the same manner another two times. Final extracts were then mixed with 100% ethanol, centrifuged at 4°C for 5 min at 16,000×g, and stored at −20°C overnight to precipitate DNA. Precipitates were collected by centrifugation, washed with 70% ethanol, and centrifuged at 4°C for 2 min at 16,000×g. As much of the supernatant was removed as possible, and pellets were resuspended in 500 μL Tris-EDTA buffer. Final DNA concentration and quality were measured on a ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).

Abundance of H. pylori 23S rRNA in gastric wash samples

We analyzed gastric wash DNA samples by SYBR Green real-time polymerase chain reaction (PCR), with H. pylori 23S rRNA as target and human GAPDH as reference. It was necessary to normalize abundance of H. pylori 23S rRNA gene using total volume of gastric washes. However, it was difficult to accurately quantify total volume of gastric washes obtained in each endoscopy. Alternatively, the amounts of DNA of human cells exfoliated from the gastric mucosa during washing reflect total volume of gastric washes.^26,29 Therefore, we normalized abundance of H. pylori 23S rRNA gene using human GAPDH gene instead of other bacterial gene. H. pylori 23S rRNA was amplified with sense primer “acgagatgggagctgtctcaacc” and antisense primer “agcattgtcctgcctgtggataac,” while human GAPDH was amplified with sense primer “cgagatccctccaaaatcaa” and antisense primer “ctgcaaatgagcctacagca.” Reactions were performed on an ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. Data were analyzed in SDS2.1 by comparative cycle threshold (ΔCt).

Quantitative pyrosequencing analysis of multiple H. pylori strains

Pyrosequencing enables quantitative analysis of the abundance of polymorphisms at a given site. 23S rRNA was amplified and biotinylated by nested PCR from gastric wash samples. In the first reaction, a 255 bp fragment was amplified with sense primer “acgagatgggagctgtctcaacc” and antisense primer “agcattgtcctgcctgtggataac.” Amplification products were then used as a template in the second reaction to amplify a 90 bp fragment with sense primer “gaggtgaaaattcctcctacccgcg” and antisense primer “gcgcatgatattcccattagcagtgc.” Reactions consisted of touchdown PCR with denaturation at 95°C for 30 s, annealing at appropriate temperatures for 30 s, and extension at 72°C for 30 s. Finally, amplification products were analyzed by pyrosequencing on a Pyromark Q24 (QIAGEN, Valencia, CA, USA) using primer “acccgcggcaagacg.” Sequence to analyze A2143V is “GVAAGACCCCGTGGACCTTTACTACAA” and A2144R “RAGACCCCGTGGACCTT TACTACAAC.”

Results

H. pylori abundance in gastric wash samples

We clinically diagnosed 19 patients in whom H. pylori had been “successfully eradicated” on the basis of the UBT, because all results showed values below the cutoff line (<2.5‰; Table 1). Interestingly, the values showed some variance among patients, although they were below the cutoff line (0.0‰–2.1‰, mean 0.77‰ ± 0.64‰). We used real-time PCR to estimate the abundance of H. pylori DNA in gastric wash samples, as measured by the abundance of 23S rRNA gene relative to the abundance of human GAPDH gene. Abundance of H. pylori 23S rRNA gene significantly decreased 1 year after H. pylori eradication, but none of the samples showed a value of 0 (1242.0 ± 1313.0 to 367.6 ± 467.2, p = 0.02; Figure 1). The abundance of H. pylori DNA was the lowest at the 2-year follow-up, but it was found to have increased at the 3-year follow-up.

Figure 1.

Time course of the abundance of Helicobacter pylori DNA at four timepoints. The H. pylori 23S rRNA gene and human GAPDH gene were analyzed as the target and housekeeping reference genes, respectively.

Measurement of the ratio of H. pylori strains before and after eradication therapy

Representative pyrograms were shown in Figure 2. The frequencies of the A2143G, A2143C, and A2144G mutant H. pylori strains were examined using gastric wash samples at four timepoints (Table 1, Figure 3(a) and (b)). None of the samples showed the A2143C mutant, whereas both A2143G and A2144G mutant strains were detected before eradication therapy. The prevalence of the A2143G and A2144G mutants increased after eradication therapy, and the prevalence at 3 years was significantly higher than that before treatment (16.8% ± 22.6% to 50.6% ± 14.6% for A2143G, p = 0.0001; 34.6% ± 35.4% to 73.7% ± 23.4% for A2144G, p = 0.002). Moreover, there was a significant correlation between the frequencies of A2143G and A2144G mutants at the timepoint of 3 years (r = 0.96, p = 0.0001; Figure 3(c)).

Figure 2.

Representative pyrograms of positions 2143 and 2144 of the 23S rRNA gene (a) before treatment and (b) at the timepoint of 3 years in case no. 14.

Figure 3.

(a) Time course of the ratio of the prevalence of A2143G to that of wild-type Helicobacter pylori at four timepoints. (b) Time course of the prevalence of A2144G to that of wild-type H. pylori at four timepoints. (c) Significant correlation of the ratio of A2143G prevalence to that of A2144G at the timepoint of 3 years.

Mutant H. pylori strains were inversely correlated with age

Both mutant strains A2143G and A2144G were inversely correlated with age at the timepoint of 3 years (A2143G, r = −0.74, p = 0.005; A2144G, r = −0.68, p = 0.013; Figure 4(a) and (b)).

Figure 4.

Significant inverse correlation of the frequency of (a) A2143G and (b) A2144G with age at the timepoint of 3 years.

Discussion

Using gastric wash DNA, we obtained a quantitative time-course analysis of the abundance and prevalence of H. pylori genotypes before and after CAM-based eradication therapy (Figure 5). Low levels of H. pylori DNA were still detectable at the first-year follow-up in all samples with negative post-treatment UBT results. The abundance of H. pylori DNA decreased significantly until the 2-year follow-up, but it had increased at the 3-year follow-up. These results suggest that our PCR assay is more sensitive than conventional methods and that it is necessary to monitor the impact of treatments in the follow-up period even if H. pylori is found to be successfully eradicated in patients on the basis of conventional tests after eradication therapy.

Figure 5.

The illustration of time course of mixed Helicobacter pylori infection analyzed by 23S rRNA pyrosequencing at four timepoints. The abundance of H. pylori DNA decreased significantly until the 2-year follow-up, but it had increased at the 3-year follow-up. Although the abundance of CAM-resistant strains was low after eradication therapy, these strains increased and were found to be enriched in the follow-up period, leading to persistent H. pylori infection and eventually to the recrudescence (clinically detectable recurrence) of H. pylori infection after successful eradication therapy.

Both A2143G and A2144G mutant strains were detected before CAM-based eradication therapy. Moreover, the ratio of the prevalence of mutant strains to that of the wild-type strain was found to have already increased significantly 1 year after eradication therapy and continued to increase, suggesting that CAM-resistant strains survived more easily after treatment. Although the abundance of CAM-resistant strains was low after eradication therapy, these strains increased and were found to be enriched in the follow-up period, leading to persistent H. pylori infection and eventually to the recrudescence (clinically detectable recurrence) of H. pylori infection after successful eradication therapy. In an earlier qualitative study, Matsuoka et al.²¹ suggested that a small number of CAM-resistant strains could emerge spontaneously in the stomach and that resistant strains are not formed but selected by CAM administration.

Both the mutant H. pylori strains A2143G and A2144G were inversely correlated with age at the timepoint of 3 years, indicating that the prevalence of CAM-resistant strains is on the rise in patients of younger age. This might be attributed to the overuse of macrolides in Japan. In fact, in Europe, the use of macrolides for various infections over the past 2 decades, especially long-standing macrolides, has contributed to the selection of resistant H. pylori strains.³⁰

Our assay enables the detection of H. pylori DNA at a higher sensitivity than that of conventional tests, and it can quantify multiple genotypes. Failure of H. pylori eradication is one of the independent risk factors for recurrent neoplasia after endoscopic resection of early gastric cancer.³¹ Hence, our assay is useful to (1) diagnose H. pylori infection, including mixed infections; (2) choose personalized treatment strategies; and (3) monitor the impact of the treatment in the follow-up period, especially for high-risk patients.

Our assay has several advantages, such as being simple and safe because it circumvents the need for biopsy, as well as being quantitative, rapid, economical, and representative of the whole stomach. Moreover, it is possible to analyze gastric wash DNA for both H. pylori genotypes and human cancer–related genes.^26–29 Therefore, our assay is feasible for clinical application and should thus be validated in large-scale clinical trials. We believe that the long-term success of preventing gastric cancer development lies in effective H. pylori eradication interventions and consecutive surveillance, which can be facilitated by the application of sensitive and accurate detection techniques.

Footnotes

Acknowledgements

R.O., Y.W., and H.Y. conceived the study, designed and executed experiments, analyzed data, prepared figures and tables, and wrote the manuscript. S.M., S. O., and Y.S. obtained clinical samples and analyzed data. K.M., M.K., and F.I. provided intellectual support. H.Y. supervised all aspects of the study. R.O. and Y.W. share co-first authorship.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

The study was conducted in accordance with all rules and regulations of the St. Marianna University School of Medicine Institutional Review Board (#1929), and written informed consent was obtained from each patient.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by JSPS KAKENHI Grant (JP25460937 and JP16K09295) to Y.W.

Guarantor

Hiroyuki Yamamoto.

References

Rugge

Genta

Di Mario

et al . Gastric cancer as preventable disease. Clin Gastroenterol Hepatol. Epub ahead of print 19 May 2017. DOI: 10.1016/j.cgh.2017.05.023.

Miftahussurur

Yamaoka

Graham

. Helicobacter pylori as an oncogenic pathogen, revisited. Expert Rev Mol Med 2017; 19: e4.

Lee

Chiang

Chou

et al . Association between Helicobacter pylori eradication and gastric cancer incidence: a systematic review and meta-analysis. Gastroenterology 2016; 150: 1113–1124.

Asaka

Mabe

Matsushima

et al . Helicobacter pylori eradication to eliminate gastric cancer: the Japanese strategy. Gastroenterol Clin North Am 2015; 44: 639–648.

Fukase

Kato

Kikuchi

et al .; Japan Gast Study Group. Effect of eradication of Helicobacter pylori on incidence of metachronous gastric carcinoma after endoscopic resection of early gastric cancer: an open-label, randomised controlled trial. Lancet 2008; 372: 392–397.

Hooi

JKY

Lai

et al . Global prevalence of Helicobacter pylori infection: systematic review and meta-analysis. Gastroenterology 2017; 153: 420–429.

Sugano

Tack

Kuipers

et al . Kyoto global consensus report on Helicobacter pylori gastritis. Gut 2015; 64: 1353–1367.

Graham

Dore

. Helicobacter pylori therapy: a paradigm shift. Expert Rev Anti Infect Ther 2016; 14: 577–585.

Shiotani

Dore

et al . Treating Helicobacter pylori effectively while minimizing misuse of antibiotics. Cleve Clin J Med 2017; 84: 310–318.

10.

Dang

Graham

. Helicobacter pylori infection and antibiotic resistance: a WHO high priority? Nat Rev Gastroenterol Hepatol 2017; 14: 383–384.

11.

O’Connor

Fischbach

Gisbert

et al . Treatment of Helicobacter pylori infection 2016. Helicobacter 2016; 21(suppl. 1): 55–61.

12.

Gisbert

McNicholl

. Optimization strategies aimed to increase the efficacy of H. pylori eradication therapies. Helicobacter 2017; 22: e12392.

13.

Thung

Aramin

Vavinskaya

et al . Review article: the global emergence of Helicobacter pylori antibiotic resistance. Aliment Pharmacol Ther 2016; 43: 514–533.

14.

Kawai

Takahashi

Suzuki

et al . Changes in the first line Helicobacter pylori eradication rates using the triple therapy—a multicenter study in the Tokyo metropolitan area (Tokyo Helicobacter pylori study group). J Gastroenterol Hepatol 2014; 29(suppl. 4): 29–32.

15.

Nishizawa

Suzuki

. Mechanisms of Helicobacter pylori antibiotic resistance and molecular testing. Front Mol Biosci 2014; 1: 19.

16.

Xuan

Zhou

et al . Detection of clarithromycin-resistant Helicobacter pylori in clinical specimens by molecular methods: a review. J Glob Antimicrob Resist 2016; 4: 35–41.

17.

Chey

Leontiadis

Howden

et al . ACG clinical guideline: treatment of Helicobacter pylori infection. Am J Gastroenterol 2017; 112: 212–239.

18.

Malfertheiner

Megraud

O’Morain

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.

19.

Yeo

Shiu

et al . Taiwan Gastrointestinal Disease and Helicobacter Consortium. First-line Helicobacter pylori eradication therapies in countries with high and low clarithromycin resistance: a systematic review and network meta-analysis. Gut. Epub ahead of print 26 September 2016. DOI: 10.1136/gutjnl-2016-311868.

20.

Rimbara

Noguchi

Tanabe

et al . Susceptibilities to clarithromycin, amoxycillin and metronidazole of Helicobacter pylori isolates from the antrum and corpus in Tokyo, Japan, 1995–2001. Clin Microbiol Infect 2005; 11: 307–311.

21.

Matsuoka

Yoshida

Hayakawa

et al . Simultaneous colonisation of Helicobacter pylori with and without mutations in the 23S rRNA gene in patients with no history of clarithromycin exposure. Gut 1999; 45: 503–507.

22.

Yakoob

Fan

et al . Polycolonization of Helicobacter pylori among Chinese subjects. Clin Microbiol Infect 2000; 7: 187–192.

23.

Noguchi

Rimbara

Kato

et al . Detection of mixed clarithromycin-resistant and -susceptible Helicobacter pylori using nested PCR and direct sequencing of DNA extracted from faeces. J Med Microbiol 2007; 56: 1174–1180.

24.

Selgrad

Tammer

Langner

et al . Different antibiotic susceptibility between antrum and corpus of the stomach, a possible reason for treatment failure of Helicobacter pylori infection. World J Gastroenterol 2014; 20: 16245–16251.

25.

Cao

Didelot

et al . Progressive genomic convergence of two Helicobacter pylori strains during mixed infection of a patient with chronic gastritis. Gut 2015; 64: 554–561.

26.

Watanabe

Kim

Castoro

et al . Sensitive and specific detection of early gastric cancer with DNA methylation analysis of gastric washes. Gastroenterology 2009; 136: 2149–2158.

27.

Baba

Oishi

Watanabe

et al . Gastric wash-based molecular testing for antibiotic resistance in Helicobacter pylori. Digestion 2011; 84: 299–305.

28.

Yamamoto

Watanabe

Oikawa

et al . BARHL2 methylation using gastric wash DNA or gastric juice exosomal DNA is a useful marker for early detection of gastric cancer in an H. pylori-independent manner. Clin Transl Gastroenterol 2016; 7: e184.

29.

Miyamoto

Watanabe

Oikawa

et al . Analysis of Helicobacter pylori genotypes in clinical gastric wash samples. Tumour Biol 2016; 37: 10123–10132.

30.

Megraud

Coenen

Versporten

et al .; Study Group participants. Helicobacter pylori resistance to antibiotics in Europe and its relationship to antibiotic consumption. Gut 2013; 62: 34–42.

31.

Kwon

Heo

Lee

et al . Failure of Helicobacter pylori eradication and age are independent risk factors for recurrent neoplasia after endoscopic resection of early gastric cancer in 283 patients. Aliment Pharmacol Ther 2014; 39: 609–618.

Enrichment of Helicobacter pylori mutant strains after eradication therapy analyzed by gastric wash–based quantitative pyrosequencing

Abstract

Keywords

Introduction

Materials and methods

Clinical samples

Sample collection of gastric washes

DNA extraction

Abundance of H. pylori 23S rRNA in gastric wash samples

Quantitative pyrosequencing analysis of multiple H. pylori strains

Results

H. pylori abundance in gastric wash samples

Measurement of the ratio of H. pylori strains before and after eradication therapy

Mutant H. pylori strains were inversely correlated with age

Discussion

Footnotes

Acknowledgements

Declaration of conflicting interests

Ethical approval

Funding

Guarantor

References