Long-Read- and Short-Read-Based Whole-Genome Sequencing Reveals the Antibiotic Resistance Pattern of Helicobacter pylori

ABSTRACT The rates of antibiotic resistance of Helicobacter pylori are increasing, and the patterns of resistance are region and population specific. Here, we elucidated the antibiotic resistance pattern of H. pylori in a single center in China and compared short-read- and long-read-based whole-genome sequencing for identifying the genotypes. Resistance rates of 38.5%, 61.5%, 27.9%, and 13.5% against clarithromycin, metronidazole, levofloxacin, and amoxicillin were determined, respectively, while no strain was resistant to tetracycline or furazolidone. Single nucleotide variations (SNVs) in the 23S rRNA and GyrA/B genes revealed by Illumina short-read sequencing showed good diagnostic abilities for clarithromycin and levofloxacin resistance, respectively. Nanopore long-read sequencing also showed a good efficiency in elucidating SNVs in the 23S rRNA gene and, thus, a good ability to detect clarithromycin resistance. The two technologies displayed good consistency in discovering SNVs and shared 76% of SNVs detected in the rRNA gene. Taking Sanger sequencing as the gold standard, Illumina short-read sequencing showed a slightly higher accuracy for discovering SNVs than Nanopore sequencing. There are two copies of the rRNA gene in the genome of H. pylori, and we found that the two copies were not the same in at least 26% of the strains tested, indicating their heterozygous status. Especially, three strains harboring a 2143G/A heterozygous status in the 23S rRNA gene, which is the most important site for clarithromycin resistance, were found. In conclusion, our results provide evidence for an empirical first-line treatment for H. pylori eradication in clinical settings. Moreover, we show that Nanopore sequencing is a potential tool for predicting clarithromycin resistance. IMPORTANCE Helicobacter pylori resistance has been increasing in recent years. The resistance profile, which is important for empirical treatment, is region and population specific. We found high rates of resistance to metronidazole, clarithromycin, and levofloxacin in H. pylori in our center, while no resistance to tetracycline or furazolidone was found. These results provide a reference for local physicians prescribing antibiotics for H. pylori eradication. Nanopore sequencing recently appeared to be a promising technology for elucidating whole-genome sequences, which generates long sequencing reads and is time-efficient and portable. However, a relatively higher error rate of sequencing reads was also found. In this study, we compared Nanopore sequencing and Illumina sequencing for revealing single nucleotide variations in the 23S rRNA gene, which determines clarithromycin resistance, and we found that although there were a few false discoveries, Nanopore sequencing showed good consistency with Illumina sequencing, indicating that it is a potential tool for predicting clarithromycin resistance.

-Please check the information about the number of strains sequenced as they seem to be 104 according to the whole manuscript but according to Lines 114 and 115 "Four Hp strains exhibited very slow growth and thus, AST was not performed in these strains; however, whole genome sequencing was still performed." - Table 1 should include an extra row with the total number of strains tested and resistance percentages without subgroups. -Please include the legends of supplementary tables and figures -Please include a section in methods describing the statistical test used -In the supplementary table some data has a ?. Please clarify the meaning (for example Hpfe076, female, 43, ＞32？) -The strains Hpfe024 has a levofloxacin MIC very low, ＜0.002 S, but it has a mutation by WGS. Was the E test repeated? -Mutation in PBP1A should be included in supplementary table both in AMX susceptible or resistance and the genotypes obtained for metronidazole resistance should also be included. Although the results may not seem relevant they are interesting for authors studying the same subject.
Reviewer #2 (Comments for the Author): The relationship between 23SrRNA mutation(A2143G,A2142G) and clarithromycin phenotype resistance has been well clarified by many researchers, however in this paper the A2143G mutation was still found in a CLR-S strain, the author did not mention ( detected by Nanopore or Illumina sequencing) or discuss this discovery.Is the mutation A2143G in a CLR-S strain(MIC 0.25), a false among the 8.5%?
In the discussion part, this paper were not writtened logically nor more powerfully (eg,discussed on amoxicillin rather than clarithromycin).
Line79-82, the overall eradication rate of PPI+Bismuth+metronidazole+tetracyclin regime is only 72.3% (ITT) [11], which is too low to be an acceptable regime as David G.Y. suggested , in fact this low eradicaton rate is due to patient's low compliance (such as missed doses, drink or smoke)rather than antibiotic resistance in the referred paper, so the author should state more rigorously to improve the evidence.
Antibiotic resistance differs in different backgroud, the author had better display whether the strains obtained from patient accepting initial or rescued therapy.

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publication Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees. Need to upgrade your membership level? Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.

Reviewer comments:
Reviewer #1 (Comments for the Author): In this study two whole genome sequencing technologies were compared to study mutations related to antimicrobial resistance in H. pylori.
Some comments: -Please check the information about the number of strains sequenced as they seem to be 104 according to the whole manuscript but according to Lines 114 and 115 "Four Hp strains exhibited very slow growth and thus, AST was not performed in these strains; however, whole genome sequencing was still performed." Answer: Thanks for carefully reviewing the manuscript and pointing out the issue. Because the four strains were not included in the following genotype and phenotype analysis and no information of the 4 stains were used, we decide to delete the sentence to avoid confusion, and change "108" to "104" in the revised version of manuscript (Line 112).
- Table 1 should include an extra row with the total number of strains tested and resistance percentages without subgroups.
Answer: An extra row was added in Table 1 Answer: Thanks for the reminder. We routinely repeat two times for the E-test. At the reminder of the reviewer, we repeated the E-test again and performed sanger sequencing of the strain which verified the N87K mutation and sensitivity to levofloxacin. The QRDR region mutation of gryA was not always consistent with antibiotic resistant phenotype, as indicated in this research that the diagnostic Youden's index of 0.671, and consistent with previous reports (for example, PMID: 33374988, in which three strains with MIC values of 0.008, 0.023, 0.004 were reported with N87I, N87K and N87T mutations respectively. and other reports, PMIDs: 27454429, 32545318, 35124867). It worth noting and interesting that the MIC < 0.002 here is even lower than the reports, which we will keep in mind.
-Mutation in PBP1A should be included in supplementary table both in AMX susceptible or resistance and the genotypes obtained for metronidazole resistance should also be included. Although the results may not seem relevant they are interesting for authors studying the same subject.
Answer: We have added information of the genotype related to AMX and MTZ to supplementary Table 1.

Reviewer #2 (Comments for the Author):
The relationship between 23SrRNA mutation(A2143G,A2142G) and clarithromycin phenotype resistance has been well clarified by many researchers, however in this paper the A2143G mutation was still found in a CLR-S strain, the author did not mention ( detected by Nanopore or Illumina sequencing) or discuss this discovery. Is the mutation A2143G in a CLR-S strain(MIC 0.25), a false among the 8.5%?
Answer: Thanks for carefully reviewing the manuscript and providing the thoughtful suggestion. In the original manuscript, we described that possible contamination reads might exist for Hpfe091, which was sensitive to CLR (MIC 0.25), but detected with A2143G mutation by the Illumina sequencing data. We have re-sequenced the stain with Illumina platform and found no A2142G or A2143G mutation. Accordingly, we have updated the data in Tables 2 of the revised manuscript.
In the discussion part, this paper were not writtened logically nor more powerfully (eg,discussed on amoxicillin rather than clarithromycin).
Answer: Thanks for your helpful comment on the Discussion section. Accordingly, we have modified the section to pay more attention to CLR in the section of the revised manuscript (paragraph two of Discussion section. Lines 351-362).
Firstly, we discussed the epidemiology of Hp resistance that we found in the present study, with reference to previous data. Secondly, we discussed the relationship of A2142G, A2143G mutations of 23S rRNA to CLR resistance. Thirdly, we discussed the error rate of Nanopore sequencing and its potential in elucidating point mutations mediated antibiotic resistance. Fourthly, we discussed the heterozygous status and its possible impact of 23S rRNA.
We hope that the revised Discussion section is now acceptable.
Line79-82, the overall eradication rate of PPI+Bismuth+metronidazole+tetracyclin regime is only 72.3% (ITT) [11], which is too low to be an acceptable regime as David G.Y. suggested , in fact this low eradicaton rate is due to patient's low compliance (such as missed doses, drink or smoke)rather than antibiotic resistance in the referred paper, so the author should state more rigorously to improve the evidence.