Subspecies Niche Specialization in the Oral Microbiome Is Associated with Nasopharyngeal Carcinoma Risk

The relationship between oral health and the risk of nasopharyngeal carcinoma (NPC) was previously established. However, the role of oral microbiome has not been evaluated in the disease in a large epidemiological study. This paper clearly establishes a difference in the oral microbiomes between NPC patients and healthy controls which cannot be explained by other confounding factors. It furthermore identifies a pair of closely related coexcluding organisms associated with the disease, highlighting the importance of modern methods for single-nucleotide resolution in 16S rRNA sequence characterization. To the best of our knowledge, this is one of the first examples of cancer-associated niche specialization of the oral microbiome.

Below you will find the comments of the reviewers.
To submit your modified manuscript, log onto the eJP submission site at https://msystems.msubmit.net/cgi-bin/main.plex. If you cannot remember your password, click the "Can't remember your password?" link and follow the instructions on the screen. Go to Author Tasks and click the appropriate manuscript title to begin the resubmission process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Provide (1) point-by-point responses to the issues raised by the reviewers as file type "Response to Reviewers," not in your cover letter, and (2) a PDF file that indicates the changes from the original submission (by highlighting or underlining the changes) as file type "Marked Up Manuscript -For Review Only." 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to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by mSystems.
To avoid unnecessary delay in publication should your modified manuscript be accepted, it is important that all elements you upload meet the technical requirements for production. I strongly recommend that you check your digital images using the Rapid Inspector tool at http://rapidinspector.cadmus.com/RapidInspector/zmw/. This study utilized 16S rRNA sequencing of DNA from the saliva of nasopharyngeal carcinoma (NPC) cases and sex and aged matched controls to look for significant differences in the oral microbiome. Extensive in silico analysis of the data revealed a reduction in community richness in the cases compared to controls after accounting for potential confounders. Feature-based analysis identified specific amplicon variants in Granulicatella adiacens that were strongly associated with NPC status. This is a wellwritten manuscript. Suggested modifications are listed below.
We thank the reviewer. Amplicon sequence variants (ASVs) represent a methodological improvement over traditional operational taxonomic units (OTUs) used in microbiome sequencing. Both act as a sort of molecular barcode for organisms and are used in 16s rRNA sequencing as a proxy for organisms which often cannot be resolved with sufficient specificity for genus or species level identification. However, unlike OTUs which are clusters of sequences, ASVs provide single nucleotide resolution. Previous work has suggested single nucleotide resolution is critical to fully understanding the oral microbiome (1), and in the case of our study, this single nucleotide resolution is critical to detecting novel differences in the microbiome.
We have updated the text on page three (lines 111-113) to define amplicon sequence variants and explain their importance: The oral microbiome was characterized using the V3-4 hypervariable region of the 16s rRNA gene; paired end sequences were denoised to amplicon sequence variants (ASVs). ASVs provide a molecular barcode with single nucleotide resolution to distinguish different organisms. We thank the reviewer for their careful reading. This reflects the use of different metrics between the two figures to provide a more robust understanding of the community. All our distance metrics are calculated based on ASVs, and the difference is that unweighted UniFrac (Figure 1c) considers only the presence/absence while both weighted UniFrac and Bray Curtis distance (figure S3) account for abundance. Most larger studies identify a difference in the oral microbiome between smokers and non-smokers measured with beta diversity; however, the studies are heterogenous and multiple metrics are rarely used (2)(3)(4)(5).
Lines 150-154 mention tobacco use as a factor in the beta diversity analysis, as did lines 163-165.

Page 11, line 214. What is Granulicatella? Overall, is it associated with a health or dysbiosis? How does it fit with the networks identified? What is the relevance to NPC?
Granulicatella is a ubiquitous oral genus. It was originally described as a nutritionally variant Streptococcus, but molecular phylogeny indicates it is a unique genus. Two species live within the human oral cavity: G. elegans and G. adicans. G. elegans is localized to the buccal mucosa, while G. adicans is often found on the dorsal tongue but has a less strict habitat (6). We have presented a similar summary in lines 224-228.
Much of the medical literature related to the genus describes it as an opportunistic pathogen (7). Previous studies have reported a decrease in Granulicatella in periodontal disease (although due to the age of the studies, it is somewhat unclear if this is the consequence of failing to account for compositionality) (8,9); in a very small number of asthmatics (10). It is also reported to be associated with metabolic syndrome in a Korean study (11). However, these cases differ from our results in that we see no association between the abundance or prevalence of the genus as a whole and disease status (line 231-232), instead it is the specific ASV-level differences that are associated with outcome.
We cannot find literature which associates Granulicatella with head and neck cancers nor in previous descriptions of the oral microbiome in NPC, suggesting ours is a novel finding. There is also little literature around its interaction with other co-occuring species in the network. Previous papers have described biofilms containing Veillonella, Streptococcus and Granulicatella (12,13), however, since this is a novel finding, we are unsure of its implications in this context.

Page 13, Survey metadata and sample collection. Does incident mean newly diagnosed, untreated patients? Indicate both inclusion and exclusion criteria, antibiotic usage. Indicate that unstimulated saliva was obtained. How was it stored?
Many of these details for the full cohort were described in the original publication describing the full study cohort (14). We have expanded lines 330 -343 to describe inclusion criteria and definitions: Participant recruitment for the full study has been previously described (20). Briefly, incident cases of NPC in Guangdong Provenance and Guangxi Autonomous Region in southern China between March 2010 and December 2013 were invited to participate in the full study. Cases were primarily identified through a rapid ascertainment network at hospitals within the study area and in some cases through the Chinese public health network. Age and sex frequency-matched controls were selected from the total population and recruited between November 2010 and November 2014. All study participants had to be between 20 and 74 years of age and had to live in the study region. Individuals with previous malignant disease and acquired or congenital immune deficiency were excluded. Additionally, all individuals had to be fluent in Cantonese and deemed mentally and physically competent to participate. Additionally, controls could not have a diagnosis or history of NPC based on self-reported health history with medical confirmation, nor could then have lived outside the study area for more than 10 years. The study was approved by the Institutional Review Board or Ethical Review Board at all participating centers. All study participants provided written or oral informed consent.
The original study was designed starting in 2008 and sample collection began in 2010. Information about antibiotic use was not collected (lines 169-176). However, there is some evidence that the oral microbiome is resilient to antibiotic use and so although we recognize this as a limitation we do not believe that it alters our conclusions.

Reviewer #2 (Comments for the Author):
An interesting and well-presented manuscript describing the oral microbiota present in healthy vs NPC patients. The authors also put a case for two putative strains of Granulicatella adiacens as centers of niche specialization in NPC patients. However, the rest of the microbiota is not discussed much -how is the NPC microbiome different from the non-NPC state? This manuscript could be improved with more discussion of the oral microbiology aspect.
We thank the reviewer for their comments and careful reading.
General comments: 1. Why wasn't dental plaque (subgingival and supragingival) collected and analyzed from all participants? Saliva is more convenient but does not provide data on the microbiota inhabiting specific ecological niches, e.g. the prevalence of Granulicatella is usually higher in buccal swabs.
We agree with the reviewer that this would indeed be an interesting approach, especially in light of our novel finding with G. adicans. However, these samples come from a larger epidemiological study (15) and were originally collected to sequence the human and Epstein Barr Virus genomes (i.e. Xu et al (16)). Perhaps in light of our results, a future study will focus on this area.

What were the relative abundances of Gran-7770 and Gran-5a37 -looks like they were >0.02% but what was the range?
For all the abundant ASVs in our study, we defined "present" as at least 1/5000 sequences in a sample. We picked this threshold based on rarefaction depth after filtering and it was selected without consideration for specific organisms.
We added a line describing the average abundance of Granulicatella in the samples (line 230-232). We have also expanded figure 3 and added an additional set of panels (3a) to provide a more comprehensive descriptive view of the total Granulatella abundance, the relative abundance of each ASV variant, and the ratio between Gran-7770 and Gran-5a37.

Which species (or genus) were most abundant in NPC patients?
We introduced a descriptive paragraph (lines 120-126) describing the overall observed salivary microbiome. We also expanded the description of the phylofactor result (lines 201-206) and introduced ANCOM to test abundance-based differences at the ASV-level (lines 210-221). We chose to use ASVs here rather than collapse the data to genus since ASVs are more likely to be externally valid if other researchers choose to use different databases. We also believe that the single nucleotide resolution granted by ASVs is superior to the collapsed genus-level information and better recapitulates the underlying community. We also applied ANCOM collapsed to genus level, but nothing achieved the a priori threshold of a normalized W of 0.8. We have chosen not to present these results in the text other than specifically in relation to Granulicatella (line 231-232).

Were nitrate and nitrite reduction genes present in the genomic DNAs purified from participants? Perhaps real-time PCR could be used for this?
We agree this is an interesting experiment and perhaps we will be able to conduct it in the future. However, the effect of nitrate/nitrite reducing genes is one of multiple hypotheses for the mechanism behind the relationship with the oral microbiome and NPC presented in lines 311-316.
Minor/Specific comments: 1. Lines 188 and 235 -Blasted/Blasting is not a proper verb in this context -"homology searches with BLAST" could be used. We have corrected these lines.

Line 268 -2013 or 2014?
Cases were between 2010 and 2013; controls between 2010 and 2014. We have expanded the methods section to address this.

Line 276 -composition of Tris-EDTA buffer -concentrations of Tris and EDTA?
50mM Tris and 50mM EDTA were used. This has been added in lines 348-350.

Line 285 -what PCR enzyme/kit was used to amplify the rRNA genes? What hypervariable region was amplified?
We have added detailed information in the "DNA extraction, PCR, and sequencing" section (lines 370-379).

Line 289 -Agencourt 6. Line 382-383 -why was the sample that didn't contain both Gran ASVs excluded?
There was a single sample which did not contain either variant. Our analyses focused on comparing the effect of G. adicans carriage specifically, and this single sample reflected a unique carriage state (neither ASV). It is not possible to characterize a distribution from a single sample and not possible to perform statistical tests without a distribution. The sample was therefore excluded as an outlier specifically from the G. adicans related analyses, although it is present in all other sections, including the cooccurance networks. We have expanded the description in figure 3 to describe this exclusion and explained it in the text. 7. Line 432 -accession number not given. May 8, 2020 Dr. Weimin Ye Karolinska Institutet Stockholm Sweden Re: mSystems00065-20R1 (Sub-species niche specialization in the oral microbiome is associated with nasopharyngeal carcinoma risk) Dear Dr. Weimin Ye: I am satisfied that the authors have addressed all remaining reviewer concerns, and I am now happy to recommend this manuscript for publication at mSystems. However, before final acceptance please ensure that the underlying data in the "Data Availability" section is made public. The ENA accession number (PRJEB37445) is not yet publicly available -please ensure these data are published before final acceptance .
Below you will find the comments of the reviewers.
To submit your modified manuscript, log onto the eJP submission site at https://msystems.msubmit.net/cgi-bin/main.plex. If you cannot remember your password, click the "Can't remember your password?" link and follow the instructions on the screen. Go to Author Tasks and click the appropriate manuscript title to begin the resubmission process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Provide (1) point-by-point responses to the issues raised by the reviewers as file type "Response to Reviewers," not in your cover letter, and (2) a PDF file that indicates the changes from the original submission (by highlighting or underlining the changes) as file type "Marked Up Manuscript -For Review Only." Due to the SARS-CoV-2 pandemic, our typical 60 day deadline for revisions will not be applied. I hope that you will be able to submit a revised manuscript soon, but want to reassure you that the journal will be flexible in terms of timing, particularly if experimental revisions are needed. When you are ready to resubmit, please know that our staff and Editors are working remotely and handling submissions without delay. 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 mSystems.
To avoid unnecessary delay in publication should your modified manuscript be accepted, it is important that all elements you upload meet the technical requirements for production. I strongly recommend that you check your digital images using the Rapid Inspector tool at http://rapidinspector.cadmus.com/RapidInspector/zmw/.
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 Publicat ion Fees, including supplemental material costs, please visit our website. Your manuscript has been accepted, and I am forwarding it to the ASM Journals Department for publication. For your reference, ASM Journals' address is given below. Before it can be scheduled for publication, your manuscript will be checked by the mSystems senior production editor, Ellie Ghatineh, to make sure that all elements meet the technical requirements for publication. She will contact you if anything needs to be revised before copyediting and production can begin. Otherwise, you will be notified when your proofs are ready to be viewed.
As an open-access publication, mSystems receives no financial support from paid subscriptions and depends on authors' prompt payment of publication fees as soon as their articles are accepted. 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 Publicat ion 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.