Characterization of Lung and Oral Microbiomes in Lung Cancer Patients Using Culturomics and 16S rRNA Gene Sequencing

ABSTRACT Recently, microbiota dysbiosis in lung cancer has attracted immense attention. Studies on lung microbes are mostly based on sequencing, which has left the potentially functional bacteria with extremely low abundance uncovered. In this study, we characterized and compared the lung and oral cavity microbiotas using culturomics and 16S rRNA gene sequencing. Of the 198 bacteria identified at the species level from bronchoalveolar lavage fluid (BALF) samples, Firmicutes was predominant (39.90%). Twenty bacterial species isolated from BALF samples were present in at least half of the patients and were also highly abundant in oral samples. Of all isolated strains, Streptococcus and Veillonella were highly dominant. The abundance of Prevotella and Veillonella decreased from the oral cavity to the lung, whereas that of Pseudomonas increased. Linear discriminant analysis effect size demonstrated that Prevotella was more abundant in the healthy samples than in the cancerous ones, which is in accordance with the isolation of Prevotella oralis only from the healthy group using culturomics. Moreover, Gemella sanguinis and Streptococcus intermedius were isolated only from the non-small-cell lung cancer (NSCLC) group, and 16S rRNA gene sequencing showed that they were higher in the NSCLC than in the small-cell lung cancer group. Furthermore, while Bacillus and Castellaniella were enriched in lung adenocarcinoma, Brucella was enriched in lung squamous cell carcinoma. Overall, alterations were observed in the microbial community of patients with lung cancer, whose diversity might be site and pathology dependent. Using culturomics and 16S rRNA gene amplicon sequencing, this study has provided insights into pulmonary and oral microbiota alterations in patients with lung cancer. IMPORTANCE The relationship between lung microbiota and cancer has been explored based on DNA sequencing; however, culture-dependent approaches are indispensable for further studies on the lung microbiota. In this study, we applied a comprehensive approach combining culturomics and 16S rRNA gene amplicon sequencing to detect members of the microbiotas in saliva and BALF samples from patients with unilateral lobar masses. We found alterations in the microbial community of patients with lung cancer, whose diversity might be site and pathology dependent. These features may be potential bacterial biomarkers and new targets for lung cancer diagnosis and treatment. In addition, a lung and oral microbial biobank from lung cancer patients was established, which represents a useful resource for studies of host-microbe interactions.

authors to consider. I believe the manuscript will be much improved if the authors incorporate my suggestions.
Major comments • The authors use the word "cultruomics" throughout the Text, but I believe the word is "culturomics". Is it a typo throughout the manuscript? I think the authors must address this point, because the word is a part of the manuscript title. • I firmly believe the manuscript requires extensive English editing. I have come across numerous grammatical errors and typos. I have tried to highlight most of them in the minor comments, but still, I have the feeling that the text requires further editing. • In the discussion section, I think the authors should add some information about the bacterial species that were identified in the cancer patients. For instance, why Pseudomonas is such a potential threat in lung cancer, is it due to certain pathogenicity factors that are encoded by its genome? Similarly, I think adding some information about the other identified species will provide more rigor to the Discussion section. Minor comments • Line 25, mention the full name of BALF not just the abbreviation • Line 26, "Twenty" not 20. • Line 27, "BALF" instead of "BLAF". • Line 31, mention the full name of "LEfSE". • Line 33, "Prevotella Oralis" not "Prevotella orails" • Line 39, "found" not "founded". • Line 41, add "this study" before "provides" • Line 56-58, should be rephrased. • Line 67, add "an" instead of "the" • Line 68 has some spacing issues. • Line 81 again "BLAF". • Line 87, add a full stop after 1 • Line 93, "Twenty five" instead of "25" • Line 93, I think "from patients" is a mistake. • Line 95, the word should be "transbronchoendoscopy" • Line 98-99, replace ", total 21 patients' oral samples were collected" with "oral samples from 21 patients were collected". • Line 113, "100 µL" not "100ul" • Line 116, mention the full name of "MALDI-TOF MS".
• Line 118 has a strange spacing kindly check. • Line 141 "raw" instead of "Raw". • Line 143 add "were" after "sequences" • Line 171, replace "were not" with "did not". • Line 176, check if the word "cultruoims" is correct? • Line 182, I think the exact values should be mentioned that are given in the Figure 2A.
Reviewer #12 (Comments for the Author): The objective of this research aimed to analyze the lung and oral microbiome of patients with lung cancer.
There is a lack of information concerning the patients (oral pathologies?). The conclusions must be accompanied by several reservations concerning the size of the sample, the natural biodiversity of the microbiome between patients. Comparing the microbiomes (oral and pulmonary) of the same patient from a cancerous lung area with another non-cancerous area of the same patient cannot be 100% reliable. In the study, a significant segregation was found at the level of the genus, but without having sought the equivalent level at the species. The taxonomic levels of the genus are not the relevant biological measurement units for some authors (BIK, ME et al. Bacterial diversity in the oral cavity of 10 healthy individuals) 1nternational society for microbial ecology. (2010). However in the mouth, the level of ecological interest chosen to label is the genus rather than species. There also reservations concerning the 2. Notch box plot should be used instead of box whisker plot. 3. The no of clinical samples may be increased. 4. The authors should also comment on the keystone species and species diversity of each mirobiome and their changes.

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.
The manuscript titled, "Characterization of lung and oral microbiome in lung cancer patients by cultruomics and 16S rDNA sequencing" by Sun et al provides valuable insight about the bacterial populations in the lung and oral bacteriome in lung cancer patients. Overall, I firmly believe the study provides important information which can be translated to future animal models, and also pave the way for studies that may be use the identified bacterial species as biomarkers. The methods are also well explained and the bioinformatic tools were utilized well in the study. However, I have some major and minor comments for the authors to consider. I believe the manuscript will be much improved if the authors incorporate my suggestions.

Major comments
• The authors use the word "cultruomics" throughout the Text, but I believe the word is "culturomics". Is it a typo throughout the manuscript? I think the authors must address this point, because the word is a part of the manuscript title. • I firmly believe the manuscript requires extensive English editing. I have come across numerous grammatical errors and typos. I have tried to highlight most of them in the minor comments, but still, I have the feeling that the text requires further editing. • In the discussion section, I think the authors should add some information about the bacterial species that were identified in the cancer patients. For instance, why Pseudomonas is such a potential threat in lung cancer, is it due to certain pathogenicity factors that are encoded by its genome? Similarly, I think adding some information about the other identified species will provide more rigor to the Discussion section.

Minor comments
• Line 25, mention the full name of BALF not just the abbreviation • Line 26, "Twenty" not 20.
• Line 67, add "an" instead of "the" • Line 68 has some spacing issues.
• Line 87, add a full stop after 1 • Line 93, "Twenty five" instead of "25" • Line 93, I think "from patients" is a mistake.
• Line 118 has a strange spacing kindly check.
• Line 176, check if the word "cultruoims" is correct?
• Line 182, I think the exact values should be mentioned that are given in the Figure 2A.
• Line 188, you mention that 156 bacterial species were identified, but there is not mention of this in the Supplementary information. • Line 197-198, rewrite the sentence. It has grammatical errors.
• Line 331, there is a double space before "Germ-free".

191
In addition to previously known bacteria, 15 potentially new species were isolated 192 from this cultruomics study. (Table S1) 193 3、Comparison of microbiota in lung and oral cavity 194 We compared the microbiota composition between cancerous-side lung (C),

196
There was obvious difference between lung and oral cavity, but C and H seemed

213
We also compared the bacteria isolated by cultruomics between different groups.

214
Different with sequence data, there was no obvious difference for bacterial proportion 215 at phylum level between cancerous-side (C), "healthy"-side (H), and oral cavity (O).

216
Firmicutes predominated in all three sites ( Figure 3E). More than half of the species 217 (89) were isolated in all three sites, and there was 34, 32 and 39 special species 218 isolated in C, H and O group respectively ( Figure 3F). Further, we analyzed the 219 prevalence bacteria in lung and oral cavity. For cultruomics results, we defined the 220 strains isolated from more than 50% of patients as prevalent strains. For BALF 221 samples, 20 species were identified as prevalent strains, which belonged to 12 genus 222 ( in C group ( Figure 4A).

234
At the same time, we also compared the differences between H and C groups of We then sought to disclose differences in lung microbiota of NSCLC and SCLC.

244
Basic information included age, sex, smoking status and distant metastasis were 245 comparable between 2 groups (Table S2). Bacteria with abundance greater than 4% cultruomics study, we found that 77 species were only isolated from NSCLC lung 253 samples and 37 species were only isolated from SCLC lung samples ( Figure 5B).

254
Only the bacterial species isolated at least half of the patients in the group were 255 selected to the next step of comparison. is no significant difference between the two groups, the abundance showed a higher 264 tendency in the only isolated site ( Figure 5C).

Pathological diagnosis
Adenocarcinoma, n (%) 8(50%) ----  believe the study provides important information which can be translated to future animal models, and also pave the way for studies that may be use the identified bacterial species as biomarkers. The methods are also well explained and the bioinformatic tools were utilized well in the study. However, I have some major and minor comments for the authors to consider. I believe the manuscript will be much improved if the authors incorporate my suggestions.
Major comments 1. The authors use the word "cultruomics" throughout the Text, but I believe the word is "culturomics". Is it a typo throughout the manuscript? I think the authors must address this point, because the word is a part of the manuscript title.
Response: We apologize for the mistake. We agree with the reviewer's comment, and we have thoroughly checked and revised the manuscript.
2. I firmly believe the manuscript requires extensive English editing. I have come across numerous grammatical errors and typos. I have tried to highlight most of them in the minor comments, but still, I have the feeling that the text requires further editing.

Response: We apologize for the poor language of our manuscript. We have now worked on the language and have also involved native English speakers for language corrections.
3. In the discussion section, I think the authors should add some information about the bacterial species that were identified in the cancer patients. For instance, why Pseudomonas is such a potential threat in lung cancer, is it due to certain pathogenicity factors that are encoded by its genome? Similarly, I think adding some information about the other identified species will provide more rigor to the Discussion section.
Response: Thanks for the suggestion. We have added the information about the differential bacteria at species level in the discussion section, including Pseudomonas aeruginosa and Parvimonas micra, and Brucella melitensis. And we think combined the discussion of these species and our findings will provide a deeper insight for understanding the links between the species and the lung cancer.

Minor comments
• Line 25, mention the full name of BALF not just the abbreviation • Line 26, "Twenty" not 20.
• Line 67, add "an" instead of "the" • Line 68 has some spacing issues.
• Line 87, add a full stop after 1 • Line 93, "Twenty five" instead of "25" • Line 93, I think "from patients" is a mistake.
• Line 118 has a strange spacing kindly check.
• Line 176, check if the word "cultruoims" is correct?
• Line 182, I think the exact values should be mentioned that are given in the Figure 2A.
• Line 188, you mention that 156 bacterial species were identified, but there is not mention of this in the Supplementary information.
• Line 204-205, its hard to understand the meaning, kindly rephrase.
• Line 331, there is a double space before "Germ-free".
Response: We are grateful to the reviewer for his remarks, and we have corrected all these errors in the revised manuscript.

Reviewer #12 (Comments for the Author):
The objective of this research aimed to analyze the lung and oral microbiome of patients with lung cancer.
1. There is a lack of information concerning the patients (oral pathologies?).
Response: Thanks for the suggestion. As our study focused on the microbiomes (both oral and pulmonary) of the lung cancer patients, we took into account the oral condition when setting the criteria for patient recruitment. The inclusion criteria had been added in the revised manuscript.
2. The conclusions must be accompanied by several reservations concerning the size of the sample, the natural biodiversity of the microbiome between patients.
Response: We thank the reviewer for his suggestion. We agree that the presence of natural biodiversity of the microbiome across individuals. We have added the reservations concerning the size of the sample in the discussion section as suggested .
And we will call for larger and dynamic longitudinal studies in the future to verify the association between microbiome and lung cancer.
3. Comparing the microbiomes (oral and pulmonary) of the same patient from a cancerous lung area with another non-cancerous area of the same patient cannot be 100% reliable.
Response: We thank the reviewer for his relevant comment. We admit that spatial variation of microbiota within an individual is significantly less than variation across individuals in healthy lungs. However, the local tumor microenvironment of cancerous lung may be changed. We assume that the microbial differences between the cancerous lung and "healthy lung" may be associated with the tumor initiation and development.
To date, two studies used high-throughput sequencing for characterizing the difference in microbiota compositions between cancerous and "healthy" lung. Zhuo et al found that the relative abundance of family Spiroplasmataceae, and its genus Spiroplasma was significantly increased in cancerous lung (Zhuo, M., et al.,  Response: Thanks for the suggestion. We have carefully read the bibliographies and added them in the manuscript.

Reviewer #20 (Comments for the Author):
This is an interesting study aimed at characterizing and associating the oral and pulmonary microbiome of lung cancer patients. However, the study has not considered that the oral microbiome depends mainly on the oral condition. It has been widely that high bacterial richness in the salivary microbiota is significantly associated with poor oral health, as indicated by decayed teeth, periodontitis, and poor oral hygiene. Therefore, it becomes crucial to understand oral microbial diversity and how it fluctuates under conditions of disease/disturbance. Advances in metagenomics and next-generation sequencing techniques generate rapid sequences and provide extensive information on the microorganisms inhabiting a niche. Therefore, the information retrieved can be used to develop microbiome-based biomarkers for use in the early diagnosis of oral and associated diseases. However, a homogenization of the oral clinical conditions of the samples must be considered for the results to be robust. the culture and identification of BALF bacteria has rarely been reported.

89
In this study, we applied a comprehensive approach combining culturomics and species isolated in this study were previously isolated from these four sites (47/198, 198 ( Figure S2A). We did not observe any significant difference in the richness or 217 diversity of the microbial community (α diversity) between the BALF samples from C 218 and H sites, as measured using the Shannon index (p = 0.527) and Chao1 index (p = 219 0.428). Moreover, there was no difference in the overall microbiota (β diversity) 220 between the C and H groups, as measured using the Bray-Curtis distances (p = 0.39)

221
( Figure S3). Nevertheless, the oral sample was significantly different from both the C 222 and H lung samples in α and β diversity ( Figures 3B, 3C). Metastats analysis at the 223 genus level further revealed unique anatomy-related microbial features, such as higher 224 abundance of Prevotella in oral samples ( Figure 3D). Pseudomonas was the only 225 genus concentrated in BALF samples ( Figure 3D, S2B). isolated in all three sites, and 34, 32, and 39 unique species were isolated from the C,

231
H, and O groups, respectively ( Figure 3F). In addition, we analyzed the prevalence of 232 bacteria in the lungs and oral cavity. Based on the culturomics results, we defined the 233 strains isolated from over 50% of the patients as prevalent strains. Twenty prevalent 234 strains were identified from the BALF samples, which belonged to 12 genera (Table   235 2). Streptococcus was the major genus, and Streptococcus oralis, Veillonella atypica, 236 Parvimonas micra, and Actinomyces odontolyticus were found in almost all BALF 237 samples. These 20 species were also cultured at a high frequency from oral samples, 238 which indicated that the pulmonary microbiota might originate from the oral cavity. whereas those with SCC had a higher abundance of Brucella ( Figure 5E). enabled us to expand the human respiratory and oral repertoire. We also found 20 319 prevalent strains in both BALF and oral samples of patients with lung cancer, which 320 indicates that the pulmonary microbiota might originate from the oral cavity.

321
Culturomics might reduce the number of these unclassed or no-rank OTUs by 322 increasing the number of pure-cultured microbial species. In our study,

323
Pseudoramibacter alactolyticus was recovered with a relatively high frequency only 324 from the NSCLC group but was not detected in 16S rDNA amplicon sequencing.

325
Parvimonas micra is an opportunistic pathogen that is frequently associated with