Evidence for translocation of oral Parvimonas micra from the subgingival sulcus of the human oral cavity to the colorectal adenocarcinoma

doi:10.21203/rs.3.rs-2096158/v1

Download PDF

Research Article

Evidence for translocation of oral Parvimonas micra from the subgingival sulcus of the human oral cavity to the colorectal adenocarcinoma

https://doi.org/10.21203/rs.3.rs-2096158/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 08 Aug, 2023

Read the published version in Molecular Oncology →

Version 1

posted

You are reading this latest preprint version

Background: The carcinogenesis of colorectal cancer (CRC) is a multifactorial process involving both environmental and host factors, such as human genetics or the gut microbiome, which in CRC patients appears to be enriched in oral microorganisms. The aim of this work was to investigate the presence and activity of Parvimonas micrain CRC patients. To do that, samples collected from subgingival sulcus and neoplastic lesions were used for culturomics. Then, samples from different body locations (saliva, gingival crevicular fluid, feces, non-neoplastic colon mucosa, transition colon mucosa, adenocarcinoma, adenomas, metastatic and non-neoplastic liver samples) were used for 16S rRNA metabarcoding and metatranscriptomics. Whole genome sequencing was conducted for all P. micrastrains obtained.

Results: Several P. micraisolates from the oral cavity and adenocarcinoma tissue from CRC patients were obtained. The comparison of oral and tumoral P. micra genomes identified that a pair of clones (PM89KC) were 99.2% identical between locations in one CRC patient, suggesting that the same clone migrated from oral cavity to the gut. The 16S rRNA metabarcoding analysis of samples from this patient revealed that P. micra cohabits with other periodontal pathogens such as Fusobacterium, Prevotella or Dialister, both in the intestine, liver and the subgingival space, which suggests that bacterial translocation from the subgingival environment to the colon or liver could be more efficient if these microorganisms travel together forming a synergistic consortium. In this way, bacteria might be able to perform tasks that are impossible for single cells. In fact, RNA-seq of the adenocarcinoma tissue confirmed the activity of these bacteria in the neoplastic tissue samples and revealed that different oral species, including P. micra, were significantly more active in the tumor compared to non-neoplastic tissue from the same individuals.

Conclusion: P. micra appears to be able to translocate from the subgingival sulcus to the gut, where oral bacteria adapt to the new niche and could have a relevant role in carcinogenesis. According to our findings, periodontal disease, which increases the levels of these pathogens and facilitates their dissemination, could represent a risk factor for CRC development and P. micra could be used as a non-invasive CRC biomarker.

Microbiome

colorectal cancer

Parvimonas micra

oncobiome

culturomics

genomics

metatranscriptome

16S rRNA metabarcoding

oral-gut axis

periodontal disease

Cancer is a multifactorial disease linked to individual genetic predisposition and environmental factors like epigenetics, chronic tissue inflammation and personal lifestyle [1]. In the last five years the role of a patient’s own microbiome in onset, progress and prognosis of cancer disease has been established [1–8]. Previous studies have reported that gut microbiome dysbiosis could promote inflammation, tissue impairment and disruption of gastrointestinal (GI) barrier, all of which can lead to carcinogenesis [1, 2, 5, 9, 10]. Similarly, the dysbiosis of oral microbiota has also been proven to be associated not only to oral diseases but also to systemic ones such as colorectal cancer (CRC), the second most deadly type of cancer, in many cases due to liver metastasis with relapses of the disease after several months [11, 12]. In fact, oral microbiota composition differs between CRC patients and healthy individuals and it can be used for CRC prediction [3]. Understanding the correlation between oral microbiome and GI diseases is essential to develop strategies to prevent and treat large intestine cancer [3, 4]. Techniques like the 16S rRNA gene sequencing and species-specific quantitative PCR identified different periodontal pathogens, such as Parvimonas micra, Porphyromonas gingivalis, Fusobacterium nucleatum or Peptostreptococcus stomatis, which were over-represented in CRC tumor tissues, as part of the tumor microbiome (oncobiome) [13]. This consortium of oral microbes, detected in malignant tissues, has been proposed to migrate from the oral cavity to the gut and take part in the adenoma to carcinoma progression promoting polymicrobial and procarcinogenic biofilms, which protects tumor from the host immune system [14].

P. micra is an anaerobe, Gram-positive commensal cocci normally presented as low-abundance bacterium in the subgingival cavity, respiratory system, GI tract and sometimes in urogenital mucosa. Nevertheless, P. micra can also act as an opportunistic pathogen in periodontal disease [15–18], being one of the most predominant species in periodontitis lesions and infected root canals [19, 20]. This microbe has been found in high prevalence in patients suffering from periodontitis, leading to microbial oral dysbiosis and host-bacteria homeostasis breakdown by disrupting the NOD2 signaling pathway into the host cells [16]. Moreover P. micra is the main microbe present in Gram-positive anaerobic cocci (GPAC) sepsis. In the subgingival microenvironment, P. micra could carry out a “cross-talk” process with other red and orange complex periodontal pathogens, such as P. gingivalis, driving inflammation, gingival bleeding, breakdown of periodontal tissues and, if the disease progresses, tooth loss [17]. Moreover, in apical chronic periodontitis, P. micra and F. nucleatum exhibit synergistic biofilm formation [15]. However, virulence mechanisms of P. micra and its interactions with other pathogenic bacteria, which may contribute to these diseases, remain unknown [21].

In previous studies, 16S ribosomal ribonucleic acid (rRNA) gene amplicon sequencing and quantitative PCR (qPCR) showed that P. micra was enriched in feces of CRC patients, when compared to feces of healthy individuals, being this bacteria a possible non-invasive fecal biomarker for CRC early diagnosis [22]. Furthermore, a recent genomic study indicates that P. micra, enterotoxigenic Bacteroides fragilis and P. stomatis could be accurate biomarkers for the presence of laterally spreading tumors (LSTs) [23]. P. micra appears in high abundance in tumors specifically characterized by a clear immunological response or Consensus Molecular Subtype 1 (CMS1) [6]. Moreover, P. micra showed protumorigenic ability in colon cell lines, Apc^min/+ mice and germ-free mice by an altered Th17 immune response and enhanced inflammatory eukaryotic pathways [24]. In addition, another research indicated that P. micra promotes tumor development and cell proliferation by triggering alterations in the immune system, modifications in human DNA methylation status and consequently promoting colon inflammation [25].

Besides all the aforementioned studies, there is still no evidence of how P. micra migrates from the oral cavity to the colon. Recent works reported that some F. nucleatum isolates from the oral cavity and tumor can be genetically identical, suggesting that this anaerobic microbe could migrate from the oral cavity to the tumor via a transient bacteremia [26, 27]. Mapping the possible routes of these oncobacteria traced from the hypothetical original region to the tumor could be essential to target pathogens before inflammation and damage occurs in colon mucosa [7, 26, 28].

Most microbiome and GI cancer studies have focused on the microbiome analysis in fecal samples [1, 29], which is a convenient and non-invasive approximation of the patient intestinal microbial biodiversity. However, it is well known that fecal samples do not represent the microbial communities of colon tissues [30, 31]. In addition, last year’s works support the correlation between oral and tumor microbiota [3, 26, 27, 29, 32], underlining that further microbiome analyses of oral fluids and colon mucosa tissues in CRC patients are needed in order to study the potential role of anaerobe oral pathobionts in initiation and/or aggravation of CRC. To do that, bacterial RNA sequencing represents a valuable approach, since it allows studying the activity of only the viable microbes presented in different-nature tissues. Only a few metatranscriptomic analyses of the oncobiome have been developed [33, 34], which calls for an in-depth characterization of the activity and functions of the tumor microbiome in CRC in order to detect possible protumorigenic bacterial factors.

In this context, the main objective of the present work was to clarify the origin of P. micra, as well as its presence and activity within the colonic tissues. For that, different samples from the oral cavity and colon, including gingival crevicular fluid, saliva, feces, non-neoplastic colon mucosa, colorectal adenocarcinoma tissues and metastatic tissues, were analyzed using culturomics, genomics and metatranscriptomics as well as 16S rRNA metabarcoding approaches.

Sample collection

Patients from the University Hospital of A Coruña (HUAC) recently diagnosed as CRC positive after colonoscopy, were enrolled in the study. For selection of the correct CRC diagnosed patients, the following exclusion criteria were established: (1) no antibiotics intake in less than one month prior to diagnosis and/or no infectious disease, (2) no chemotherapy and/or radiotherapy treatments prior to colon laparoscopy resection, (3) no genomic predisposition to develop CRC (family history of CRC, Lynch syndrome among others) and/or other malignant lesions, (4) no presence of other gut disorders (such as inflammatory bowel disease), (5) no immunological diseases and (6) no transplants and/or any inmunosupresor treatment. An informed consent was signed by all patients. Both stool (~ 20 mL) and unstimulated saliva (~ 5 mL) samples were collected at home by the CRC patient before any treatment or diet. Samples were stored at 4ºC until delivered to the laboratory. Feces were kept in presence of 10 mL of RNAlater reagent (ThermoFisher, UK). Both samples were stored in the laboratory at -80ºC until further analysis.

The gingival crevicular fluid samples (Fig. 1) were collected using sterile paper points ISO 30 (Henry Schein, USA) that were inserted in the subgingival sulcus of different teeth for 10 seconds. Eight paper points were placed in an eppendorf tube containing 1 mL of RNAlater (ThermoFisher, UK) and frozen at -80ºC until use. Another four sterile endodontic strips were collected and preserved using eSwab™ liquid tubes (Copan Diagnostics, USA) for culturomic procedures.

Primary tumor tissue sample, non-neoplastic tissue samples from distant area (colon and liver), colon transition area (interface between non-neoplastic and adenocarcinoma region) and metastatic visceral lesion in liver (when presented) were collected by sterile carefully dissection in the same operating room after surgical resections (Fig. 1), both performed at HUAC. It should be pointed out that at the beginning of both laparoscopic surgeries 2 g of amoxicillin/clavulanic acid were administered as well as during postoperatory (a total of 3 doses every 8 h). All tissue samples were immediately stored in GutAlive collection devices (MicroViable Therapeutics, Spain). A total of 20 mg of each type of tissue sample were stored at -80ºC, in presence of 500 µL of RNAlater reagent (ThermoFisher, UK), for nucleic acids extraction and sequencing procedures. The remaining tissues were used for immediate culture under anaerobic conditions, as explained below. Also, formalin-fixed paraffin-embedded (FFPE) tissue samples (non-neoplastic, adenoma and adenocarcinoma tissues) from the surgical specimens were selected and studied in the Pathological Anatomy Service of HUAC and kept at 4ºC until 16S rRNA gene analysis. Figure 1 shows a graphic summary of all different-nature samples used in this study as well as the different approaches (culturomics, genomics and metatranscriptomics) developed for each kind of sample.

Culture media and conditions for P. micra growing

First, all tissue samples from CRC patients were cut into small fragments and immersed in Thioglycollate Fluid Medium (Becton-Dickinson, USA) using a sterile scalpel. Afterwards, small fragments were homogenized using a glass sterile mortar and pestle until solid tissue fragments were reduced to fine particles. Immediately, the final viscous fluid was vortexed at high speed (2 min). For gingival fluid samples, eSwab™ liquid tubes (Copan Diagnostics, USA) containing four paper points were vortexed directly at maximum speed for 2 min. A few drops of tissues and/or gingival samples were spread on Brucella blood agar plates supplemented with Hemin and Vitamin K₁ plates (Becton-Dickinson, USA) and incubated into an anaerobic jar with an atmosphere of 85% N₂, 5% CO₂ and 10% H₂ at 37ºC during at least 2 weeks.

MALDI-TOF MS identification

Bacteria grown in plates were identified using Matrix-Assisted Laser Desorption/Ionization Time-Of-Flight Mass Spectrometry (MALDI-TOF MS) at the Microbiology Service (HUAC). Each colony isolated was spotted into the MALDI plate (Bruker-Daltonik, USA) and treated with formic acid and α-Cyano-4-hydroxycinnamic acid matrix (Bruker-Daltonik, USA), both prepared following manufacturer’s instructions. A protein standard was used (Bruker-Daltonik, USA) to allow matrix and system calibration. Finally, the plate was introduced in a MALDI Biotyper® Smart instrument (Bruker-Daltonik, USA). A mean spectrum was constructed and comparison with the spectra contained in the Bruker Biotyper® database (2021) was generated for microbial identification. All P. micra isolated in this work are listed in Table 1.

Table 1

*Parvimonas* genomes used in this work.
Strain	GenBank sequence (accession no.)	References	Source
PM79KC-G-1	CP101407	This study	Gingival crevicular fluid (Patient 79¹)
PM79KC-AC-1	JANDZT000000000	This study	Adenocarcinoma (Patient 79¹)
PM79KC-AC-2	CP101408	This study	Adenocarcinoma (Patient 79¹)
PM79KC-AC-3	JANDZU000000000	This study	Adenocarcinoma (Patient 79¹)
PM79KC-AC-4	JANDZV000000000	This study	Adenocarcinoma (Patient 79¹)
PM89KC-G-1	CP101409	This study	Gingival crevicular fluid (Patient 89²)
PM89KC-G-2	CP101410	This study	Gingival crevicular fluid (Patient 89²)
PM89KC-AC-1	CP101411	This study	Adenocarcinoma (Patient 89²)
PM94KC-G-1	JANDZW000000000	This study	Gingival crevicular fluid (Patient 94³)
PM102KC-G-1	CP101412	This study	Gingival crevicular fluid (Patient 102⁴)
PM114KC-AC-1	CP101413	This study	Adenocarcinoma (Patient 114⁵)
KCOM 1037	CP031971.1	Korean Collection for Oral Microbiology	Postoperative maxillary cyst
NCTC11808	LR134472.1	Wellcome Sanger Inst., UK	Unknown
KCOM 1535	CP009761.1	Korean Collection for Oral Microbiology	Periapical abscess
ATCC 33270	ABEE02	Genome Sequencing Center, USA	Gastrointestinal tract
FDAARGOS 569	RKIS01	USA Food and Drug Administration	Purulent pleurisy
MGYG-HGUT-01301	CABKNC01	The European Bioinformatics Inst., UK	Human gut
13-07-26	BHYQ01	Sunstar Inc., US	Lung abscess
A293	AXUQ01	University of Malaya	Abdominal abscess
EYE 66	JAHXOL01	C. for Disease Control and Prevention, China	Ocular surface
EYE 30	JAHXOH01	C. for Disease Control and Prevention, China	Ocular surface
EYE 29	JAHXOG01	C. for Disease Control and Prevention, China	Ocular surface
EYE 25	JAHXOC01	C. for Disease Control and Prevention, China	Ocular surface
S3374*	JACVDA01	University Medicine Rostock, Germany	Type strain of P. parva
¹Patient 79 (P79): 63 years old female, ²Patient 89 (P89): 58 years old female. ³Patient 94 (P94): 66 years old male. ⁴Patient 102 (P102): 84 years old male. ⁵Patient 114 (P114): 67 years old male. Asterisk () indicates that S3374 is a P. parva* strain, used to root phylogenetic trees performed in this work.

Bacterial nucleic acids extraction

DNA extraction from stool and saliva samples

Defrosted stool and saliva samples (Fig. 1) at room temperature were well-vortexed and then centrifuged 2 min at 5000 rpm and 4ºC. Afterwards, 2 mL of supernatants were centrifuged again 10 min at 13000 rpm and 4ºC. Remaining supernatants were stored at -80ºC until use. Final pellets were resuspended in 100 µL of nuclease-free water. To ensure cellular wall lysis of all different type of bacteria, samples were incubated at 37ºC and 400 rpm during 1 h in the presence of 5 µL of an enzymatic cocktail (EC), containing 20 mg/mL of lysozyme (Sigma-Aldrich, USA), 1.25 KU/mL of lysostaphin (Sigma-Aldrich, USA) and 0.625 KU/mL of mutanolysin (Sigma-Aldrich, USA). DNA from stool and saliva samples was extracted using the MasterPure™ Complete DNA & RNA Purification Kit (Epicentre, USA).

DNA and RNA extraction from tissue samples and gingival crevicular fluids

DNA and RNA were extracted in parallel from 20 mg of each tissue sample (primary tumor, non-neoplastic intestinal and liver tissues, transition zone between tumor and normal tissue and metastatic lesions, Fig. 1) using the AllPrep® DNA/RNA Mini kit (Qiagen, Germany). Tissue homogenization was performed using Lysing Matrix E tubes (MP Biomedicals, China) and a 1600 MiniG system (SPEX SamplePrep, USA) at 1500 rpm during 10 min. A total of 30 µL of EC was added to samples after homogenization.

The same procedure was done for DNA and RNA isolation from the gingival fluid sample (Fig. 1) but without the homogenization step. In these samples, 2 min-vortexing at high speed was used to remove bacteria from endodontic absorbent strips. After that, papers were discarded carefully and 500 µL of sterile phosphate buffered saline were added to the sample. A long centrifugation (13000 rpm during 30 min at 4ºC) was conducted discarding supernatant. The bacteria pellet was used to follow the AllPrep® DNA/RNA Mini kit manufacturer’s instructions with the additional enzymatic lysis step (30 µL of EC).

DNA extraction from FFPE samples

A total of five histological cuts of 10 µm per type of FFPE sample (Fig. 1) were extracted using GeneRead™ DNA FFPE kit (Qiagen, Germany) following manufacturer’s instructions. After deparaffinization and tissue lysis, the enzymatic treatment was performed adding 17 µL of EC to each sample.

Negative controls (without any type of sample) were done for all nucleic acid extraction procedures explained above. In the case of DNA extraction from FFPE samples, negative controls were performed using a slice of paraffin selected from the margins of FFPE blocks free of any remaining tissue.

Genomic DNA purification from P. micra

The Wizard® Genomic DNA Purification kit (Promega, USA) was used for genomic DNA (gDNA) extraction from P. micra. Isolated colonies were transferred to an eppendorf tube containing 500 µL of nuclease-free water. After a brief centrifugation (14000 rpm, 2 min), the pellet was used to start the genomic purification protocol following manufacturer’s instructions for Gram positive bacteria. Vortex was not used after cellular lysis to avoid DNA fragmentation.

In all cases DNA extracted was eluted in EB buffer (Qiagen, Germany). RNA from freeze tissues was eluted in RNase free water (ThermoFisher, UK) and stored at -20ºC and RNA at − 80ºC until analysis.

16S rRNA sequencing

16S rRNA sequencing for saliva, stool, tissues and gingival crevicular fluid samples

Two hypervariable regions of the 16S rRNA gene (V3-V4) [35] were amplified using 5'TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCC TACGGGNGGCWGCAG as primer forward and 5'GTCTCGTGGGCTCGGAGATGT GTATAAGAGACAGGACTACHVGGGTATCTAATCC as primer reverse with a DNA input of 5ng/µL for each sample. Libraries were prepared following the Illumina 16S Metagenomic Sequencing Library Preparation protocol (Illumina, USA). Pooled final libraries were diluted to a final concentration of 10 pM and, after, 20% of 10 pM PhiX (Illumina, USA) was added for sequencing using an Illumina MiSeq v3 reagent kit 2×300 paired end (Illumina, USA). Quantification of DNA was performed using the Qubit dsDNA HS Assay Kit (Invitrogen, USA) and library size was checked using a 2100 Bioanalyzer Instrument (Agilent Technologies, USA). Negative controls were done in all cases to discard possible contaminations.

16S rRNA sequencing for FFPE samples

For FFPE samples, microbiome analysis was performed by amplifying five hypervariable regions (V2, V3, V5, V6 and V8) following the method previously described by Nejman et al. [8] using 100 ng of DNA as template. Pooled final libraries were diluted to a final concentration of 10 pM and, after, 15% of 10 pM PhiX (Illumina,USA) was added for sequencing using a MiSeq v2 reagent Illumina kit, 2×150 paired end (Illumina, USA). In all cases, clean-up steps were done using AMPure XP beads (Beckman Coulter, USA). Quantification of DNA and library size analyses were checked likewise in the case of the 16S rRNA V3-V4 gene sequencing. Negative controls of PCRs and extractions were sequenced to delete contaminant bacteria during bioinformatic analysis.

Whole genome sequencing of P. micra isolates

Complete sequencing of gDNA from the different isolates of P. micra previously identified by MALDI–TOF MS was conducted using two different sequencing platforms; MiSeq (Illumina, USA) and MinION (Oxford Nanopore, UK).

An input of 100 ng of gDNA was used to prepare libraries using the Illumina DNA Prep kit (Illumina, USA) and the Nextera® XT Library Preparation kit (Illumina, USA), following in all cases the manufacturer’s instructions. Final tagged libraries (12 pM) supplemented with 10% of 12 pM PhiX (Illumina, USA) were 2×150 base paired end-sequenced.

MinION library preparation was performed using the Rapid Barcoding Sequencing kit (Oxford Nanopore, UK) and the Flow Cell Priming kit (Oxford Nanopore, UK) using 60 ng/µL of gDNA.

In both cases, DNA quantification and size and quality determination were conducted as described above for 16S rRNA gene sequencing.

RNAseq of colon tissues

RNA isolated from tissue samples (30 µL) was treated twice with Invitrogen™ Kit TURBO DNA-free™ DNAse (Thermo Fisher Scientific, USA) for 30 min at 37ºC to eliminate the possible remaining DNA. RNA concentration was measured by the Qubit RNA HS Assay Kit (Thermo Fisher Scientific, USA). Samples were treated with Illumina Ribo-Zero Plus rRNA Depletion Kit (Illumina, USA) and libraries were obtained and sequenced using NextSeq Illumina Technology (Illumina, USA) (single ends, mid-output 1×150 bp) at the FISABIO sequencing platform (Valencia, Spain).

Bioinformatic analysis

16S rRNA sequencing data

The quality of all FASTQ files generated from 16S rRNA gene sequencing was checked using FastQC [36]. High quality sequences were analyzed using Qiime2 (version 2021.11) [37]. First, paired-end reads were trimmed, removing the primers and, in the case of FFPE samples, reads were also demultiplexed into regions by Cutadapt tool [38]. In order to correct the Illumina reads errors, remove chimeras and output the Amplicon Sequence Variants (ASVs), DADA2 was used [39]. Taxonomy was generated using SILVA 138 99% reference database [40]. In the case of sequences obtained from FFPE samples, reconstruction of the five small fragments resulting from the sequencing method described by Nejman et al. [8] was conducted using Short MUltiple Regions Framework (SMURF) implementation [41] in Qiime2 (version 2021.11) [37] called Sidle [42]. The Sidle pipeline was used, adapting the steps to the characteristics of the dataset obtained in this study by trimming the sequences to a 100 nt length.

In both cases of 16S rRNA sequencing data, each of the ASVs hits of negative controls were subtracted to each corresponding sample. Only bacteria kingdom ASVs were included. After that, typical genus, involved in reagent contamination, were removed from analysis according to the literature [43] as well as two extra contaminant elements that were also detected and excluded (Dermacoccaceae and Acidocella). In addition, for FFPE samples bioanalysis, a list of common environmental and paraffin contaminants were ousted [8]. Furthermore, ASVs were filtered by an abundance of < 0.01% in 16S V3-V4 rRNA gene sequencing samples or by a relative abundance of < 0.0001% in 16S V2, V3, V5, V6 and V8 rRNA sequencing samples (due to the greater microbial diversity obtained in this kind of samples). Then ASVs were filtered by intra-group prevalence (species or genus at least in 50% of samples in V3-V4 in contrast with a filter of 20% in FFPE samples). Elements not classified at the selected taxon are named with the last known taxon for that ASV and annotated with an extra “NA”. The resultant ASVs were grouped and relative abundances for each sample type was calculated. Barplots were constructed with packages Phyloseq (version 1.36.0) [44] and ggplot2 [45] in R (version 4.1) [46]. Moreover, the Venn diagram was constructed using the VennDiagram package [47] in R (version 4.1) [46].

Whole genome analysis of P. micra

Illumina reads were first processed using BBDuk (version 38.94) [48] to remove PhiX contamination, compressed losslessly with Clumpify (version 38.94) [48] to minify space on disk and trimmed with Trimmomatic (version 0.39) [49] for adapter removal and quality control. Read quality was assessed before and after this cleaning process with FastQC (version 0.11.9) [36] and MultiQC (version 1.11) [50].

Oxford Nanopore reads were basecalled and demultiplexed using Guppy (version 6.0.1) with the high accuracy model, barcoding kit SQK-RBK004, configuration file dna_r9.4.1_450bps_hac.cfg, doing read splitting and trimming adapters and barcodes. The resulting reads were further cleaned with Porechop (version 0.2.4) [51], processed by Filtlong (v. 0.2.1) [52] and quality assessed with Nanoplot (version 1.38.1) [53] before and after.

The two sets of reads (Illumina and Oxford Nanopore) were then combined using Unicycler (version 0.4.9) [54] in order to obtain contiguous and precise hybrid assemblies. Those that were closed had their start fixed with Circlator (version 1.5.5) [55] to facilitate posterior comparisons. Once these hybrid assemblies were finished, their completeness, contamination and other quality statistics were assessed with CheckM (version 1.1.3) [56]. They were then identified using KmerFinder (version 3.0.2) [57] and annotated with PGAP (version 2022-02-10.build5872) [58]. Searches against various databases were also performed, such as the Comprehensive Antibiotic Resistance Database (CARD, version 3.1.4) using Resistance Gene Identifier (RGI, version 5.2.1) [59], Resfinder (version 4.1) [60], PlasmidFinder (version 2.1) [61], Virulence Factor Database (VFDB, version 2021-10-04) [62], Oral Virus Database (OVD, version 2022-06-08) [63], Gut Phage Database (GPD, version 2020-10-29) [64] and the NCBI database using DIAMOND [65] and BLAST [66]. The different CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) systems types and its components were identified using CRISPRMiner [67], CRISPRidentify (version 1.1.0) [68] and CRISPRcasIdentifier (version 1.1.0) [69].

Comparative genome analysis

Several analyses were performed in order to find the phylogenomic relationship among strains. Additionally, twelve non-metagenomic sequences belonging to P. micra and one to Parvimonas parva deposited in NCBI database were also included in these series of tests, having their start fixed with Circlator (if circular and complete) and annotated with PGAP. P. parva was only used to root the phylogenetic tree.

Average Nucleotide Identity by Orthology (OrthoANI) was calculated using OrthoAniU (version 1.2) [70], whereas the number of variations between the samples and the reference genome was calculated using Snippy (version 4.6.0) [71]. Additionally, the pairwise number of variants between each sample and the rest was also calculated with Snippy, creating a Single Nucleotide Polymorphisms (SNPs) matrix where each assembly was used as reference to the others.

Pangenome gene clustering was performed with PIRATE (version 1.0.4) [72]. The produced whole genome alignments were converted to PHYLIP format and then used by RAxML (version 8.2.12) [73] to create a phylogenetic tree. The tree was plotted using the Interactive Tree of Life (iTOL) online tool [74].

Finally, pairwise synteny analyses of closed genomes were performed using Synima [75] and Sibelia (version 3.0.7) [76] in order to locate large scale genome movements and loss or gain of genes.

Metatranscriptome analysis of gut tissues

Raw reads were trimmed to remove adapters with Cutadapt 1.18 [38] and then filtered by quality and length using PRINSEQ [77]. After that, sequences which aligned to the reference human genome (GRCh38.p9) or to the ribosomal database SILVA 132 [40] were detected with Bowtie2 [78] and discarded. Remaining reads were mapped to the Unified Human Gastrointestinal Genome (UHGG) database [79], using Bowtie2 (version 2.4.2) [78] (with parameter: --very-sensitive) and Samtools (version 1.12) [80] to convert from SAM to sorted BAM. According to the alignments and genes coordinates, we counted (R package GenomicAlignments 1.16.0) [81] the number of hits of each gene in order to build an abundances matrix of genes. An abundances matrix of genomes was also built counting the number of hits in all the contigs of the same genome.

Kyoto Encyclopedia of Genes and Genomes (KEGG, version 2016) [82] annotation was added for each gene. First, the genes were translated into amino acids by means of EMBOSS Transeq (version 6.6.0) [83]⁠, with translation table 11 (bacterial) and frame 1. Next, each peptide was mapped against the KEGG database with HMMERsearch (version 3.3.2) [84], using a maximum e-value of 1-e06. Finally, for each gene, the KEGG annotation corresponding to the best (highest domain score) alignment was selected, with the aid of a custom R script (version 3.6.0) [46].

After the quality filter and the removal of host and ribosomal sequences, a mean of 5.4x10⁵ (SD 1x10⁵) was obtained for tumor and non-neoplastic tissue. Among them, 83.25% were annotated to the UHGG database.

The minimum number of annotated reads in a sample, 3.95 x10⁵, was used to calculate rarefaction curves and diversity indexes with a custom R script. In order to compare the gene expression both within and between samples the number of hits to a gene was normalized by its length (Kbp) and the size (in Megabp) of the annotated dataset (Reads per Kilobase per Megabasepair, or RPKM). To evaluate the fold-change in gene expression between adenocarcinoma and non-neoplastic colon tissue ratios on a logarithmic scale to base 2 were calculated (log2FC).

Genes of interest were reannotated with Phyre2 using normal mode on 25 of august 2022 [85].

Figures design

Biorender software was used to create the graphical abstract and Figs. 1 and S2 (Additional file) under license [86].

Comparative genomics of isolated P. micra strains

Gingival crevicular fluid and adenocarcinoma tissues samples (Fig. 1) from different CRC patients belonging to a larger study were grown into agar plates. A total of eleven P. micra isolates were identified from five patients: 79, 89, 94, 102 and 114 (Table 1), all of them diagnosed with colon adenocarcinoma stage IIA (T3N0). P. micra strains isolated from gingival fluid or adenocarcinoma were differentiated using the suffix “-G” or “-AC”, respectively, as listed in Table 1. An ending number was added to code the different isolates. For example, PM79KC-G-1 corresponded to isolate number 1 from gingival fluid. Patients were named as P79, P89, P84, P102 and P114.

Complete genomes were compared using other nine Parvimonas genomes from the NCBI genome database (Table 1). OrthoANI comparisons revealed the highest homology between genomes obtained from three strains in patient 89 (P89): PM89KC-G-1 and − 2 and PM89KC-AC-1 (Additional file; Figure S1), showing an identity of 99.23% between PM89KC-G-1 and PM89KC-AC-1 and 99.25% between PM89KC-G-2 and PM89KC-AC-1. In patient 79 (P79), identities between PM79KC-G isolate 1 and PM79-ACisolates 1–4 were 96.84%, 96.89%, 96.99% and 97.02%, respectively. In both cases (P89 and P79), isolates from each patient shared higher genome identity between gingival and adenocarcinoma pairs than in comparison with the reference genomes from the NCBI database (Additional file; Figure S1). Genomic guanine and cytosine (G + C) content percentage obtained for genomes of P. micra isolates was ~ 29%. Considering that the three isolates from P89 showed the highest identity, strongly suggesting a potential common origin, samples from P89 were studied in depth.

Clinical background of patient 89

The selected CRC patient is a 58-year-old woman who does not consume alcohol or tobacco, diagnosed with CRC in November 2020 in HUAC (Spain). This woman underwent a colon resection by laparoscopy in December 2020 (Fig. 1A). Adjuvant chemotherapy treatment was not administered before resection. Primary tumor, an intestinal ulcerated adenocarcinoma, was located on right colon (cecum) and classified as stage IIA (T3N0). No vascular and/or perineural invasion was detected. Furthermore, three tubular low grade dysplasia adenomas were located throughout the right colon section. Histopathological study revealed an intestinal adenocarcinoma with low grade tumor budding. Primary tumor presented a driver mutation in KRAS gene (G12X). No alterations in expression of DNA mismatch repair proteins (coded by MLH1, MSH2, MSH6 and PMS2 genes) or mutations in NRAS, BRAF genes were detected. No microsatellite instability was found. Moreover, in February 2022 this patient was diagnosed with liver metastasis (VII segment) and surgical resection was performed in the same hospital. Molecular analyses in metastasis biopsy did not detect any mutations in KRAS, NRAS or BRAF genes or microsatellite instability. The expression of mismatch repair proteins was also preserved. No lymph node affectation was detected in this patient even before any neoplastic lesion in the liver was noticed. All these clinical data were obtained from the Pathological Anatomy and General and Digestive Surgery Services of CHUAC.

An intraoral and extraoral examination of the patient was performed at Pardiñas Dental Clinic (A Coruña, Spain). Clinical measurements of probing depth, clinical attachment level, and bleeding on probing were recorded at six points around each tooth. Miller's mobility index was recorded for each tooth. A Silness Loe score of 0.3 for bleeding on probing and a CPO index for dental caries (tooth cavities) of 7 were obtained, indicating that the patient had a high degree of tooth decay and tooth loss. The Cone Beam Computed Tomography (CBCT) from the patient’s upper and lower jaw revealed an advanced degree of bone loss, the furcation involvement of teeth numbers 16, 26, 27 and 38, and the absence of teeth numbers 18, 17, 28, 35, 36, 37, 46, 47 and 48. Figure 2 shows the generalized horizontal bone loss and the high-grade furcation involvement detected in the oral cavity of P89 after oral exploration. Based on these data, a diagnosis of periodontitis in stage IVB was made, according to the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions [87].

In-depth analysis of P. micra genomes from strains isolated from patient 89

Based on a whole genome alignment phylogenetic tree (Fig. 3), which includes eleven P. micra genomes described in the present study and twelve P. micra genomes obtained from NCBI database (Table 1), it can be concluded that isolates from P89 belong to a very well differentiated group, distinguishing themselves from other P. micra isolates obtained in this study and from the rest of P. micra genomes (Fig. 3).

Furthermore, pangenome clustering revealed differences in the genes shared between strains, including differences between gingival and adenocarcinoma isolates of P89 (Fig. 4).

Due to these findings, an in-depth genome comparison of Parvimonas genomes from P89 was conducted. A total of 2120 non-synonymous mutations were found (Additional file; Table S1), including 1298 SNPs, 745 complex mutations, 40 deletions and 37 insertions, between oral PM89KC-G isolates 1 and 2, which were virtually identical, and the tumor PM89KC-AC-1 isolate. Inside these mutations, a total of 1603 genes were affected.

When studying presence and absence of genes between oral and tumor isolates from P89, some differences were detected (Tables 2 and 3). The biggest difference corresponded to a fragment of 25728 bp containing a group of 23 genes found in both oral isolates and absent in the tumor strain (Table 2). Additionally, an identical transposase element repeated at several different genome locations was present in PM89KC-AC-1 and absent in the oral isolates PM89KC-G 1 and 2 (Table 3).

Table 2

Missing genes in *P. micra* PM89KC-AC-1 isolate when compared to gingival PM89KC-G 1/ 2 isolates.
Product	Average length (bp)	Cluster	Locus Tag
Product	Average length (bp)	Cluster	PM89KC_G_1	PM89KC_G_2
TetR/AcrR family transcriptional regulator*	573	24	NM219_06315	NM220_06315
ABC transporter ATP-binding protein/permease*	1740	24	NM219_06310	NM220_06310
ABC transporter ATP-binding protein/permease*	1710	24	NM219_06305	NM220_06305
ATP-binding cassette domain-containing protein*	1518	24	NM219_06300	NM220_06300
Energy-coupling factor transporter transmembrane protein EcfT*	705	24	NM219_06295	NM220_06295
MptD family putative ECF transporter S component*	585	24	NM219_06290	NM220_06290
Sigma-70 family RNA polymerase sigma factor*	411	24	NM219_06285	NM220_06285
Hypothetical protein*	162	24	NM219_06280	NM220_06280
Hypothetical protein*	567	24	NM219_06270	NM220_06270
Exonuclease domain-containing protein*	1872	24	NM219_06265	NM220_06265
Hypothetical protein*	1935	24	NM219_06260	NM220_06260
Hypothetical protein*	885	24	NM219_06255	NM220_06255
Hypothetical protein*	618	24	NM219_06250	NM220_06250
Hypothetical protein*	1113	24	NM219_06245	NM220_06245
Hypothetical protein*	387	24	NM219_06240	NM220_06240
InlB B-repeat-containing protein*	1809	24	NM219_06235	NM220_06235
Putative ABC transporter permease*	717	24	NM219_06230	NM220_06230
Branched-chain amino acid transport system II carrier protein*	1287	24	NM219_06225	NM220_06225
O-acetylhomoserine aminocarboxypropyltransferase/cysteine synthase*	1290	24	NM219_06220	NM220_06220
DKNYY domain-containing protein*	1479	24	NM219_06215	NM220_06215
LPXTG cell wall anchor domain-containing protein*	2193	24	NM219_06210	NM220_06210
MBL fold metallo-hydrolase*	810	24	NM219_06205	NM220_06205
Aldehyde dehydrogenase*	1362	24	NM219_06200	NM220_06200
Hypothetical protein	324	25	NM219_00070	NM220_00070
IS200/IS605 family transposase	459	25	NM219_00075	NM220_00075
Hypothetical protein	1104	25	NM219_00080	NM220_00080
DUF1307 domain-containing protein	486	26	NM219_00915	NM220_00915
DUF1307 domain-containing protein	477	26	NM219_00920	NM220_00920
DUF2087 domain-containing protein	270	27	NM219_01675	NM220_01675
GNAT family N-acetyltransferase	480	27	NM219_01680	NM220_01680
ABC transporter permease	1263	28	NM219_06715	NM220_06715
Hypothetical protein	636	29	NM219_03355	NM220_03355
Hypothetical protein	372	30	NM219_00160	NM220_00160
Hypothetical protein	183	31	NM219_03780	NM220_03780
ABC transporter ATP-binding protein/permease	1611	32	NM219_01745	NM220_01745
Hypothetical protein	1638	33	NM219_06615	NM220_06615
Hypothetical protein	198	34	NM219_00320	NM220_00320
Hypothetical protein	432	35	NM219_00645	NM220_00645
Hypothetical protein	147	36	NM219_03720	NM220_03720
CPBP family intramembrane metalloprotease	678	37	NM219_06015	NM220_06015
GntR family transcriptional regulator	150	38	NM219_02485	NM220_02485
Hypothetical protein	180	39	NM219_07000	NM220_07000
Hypothetical protein	384	40	NM219_06370	NM220_06370
Asterisks (*) indicates a group of 23 genes (25728 bp) belonging to cluster 24 present in the oral isolates (PM89KC-G 1 and 2) and absent in the tumor isolate (PM89KC-AC-1). All listed genes were 100% identical between the two gingival isolates.

Table 3

Gained genes in the adenocarcinoma PM89KC-AC-1 isolate when compared to the PM89KC-G 1 and 2 gingival isolates.
Consensus product	Average length (bp)	Cluster	Locus tag
Recombinase family protein	621	41	NM221_06205
SpaA isopeptide-forming pilin-related protein	2271	41	NM221_06200
Pseudouridine-5'-phosphate glycosidase	546	42	NM221_06515
ABC transporter ATP-binding protein/permease	1560	42	NM221_06510
EXLDI protein	369	43	NM221_01675
Transposon-encoded TnpW family protein	231	43	NM221_01680
Restriction endonuclease subunit S	1242	44	NM221_00825
Hypothetical protein	435	45	NM221_00630
Rhodanese-like domain-containing protein	1083	46	NM221_01690
Hypothetical protein	1017	47	NM221_05350
IS630 family transposase*	1194	48	NM221_00050; NM221_00875; NM221_01125; NM221_01455; NM221_02510; NM221_06115; NM221_07165; NM221_07365
ABC transporter ATP-binding protein/permease	1617	49	NM221_06020
Hypothetical protein	360	50	NM221_06500
Hypothetical protein	2496	51	NM221_01205
ParB/RepB/Spo0J family partition protein	606	52	NM221_07790
Relaxase/mobilization nuclease domain-containing protein	1332	53	NM221_06215
Asterisks (*) tag an identical transposase element repeated at several different genomic locations present in the tumor isolate (PM89KC-AC-1) and absent in the gingival isolates (PM89KC-G 1 and 2).

Synteny analysis also revealed loss and gain of genes, but more interestingly, a specific cross-shaped structure in PM89KC isolates (Fig. 5). This “genomic cross” was composed of a repeat region of 30 genes (pairwise identity of ~ 80%) flanking the ~ 600 Kbp and ~ 800 Kbp positions (Additional file; Table S2), present in all P89 isolates.

These repeats correspond to a shared region in two very similar prophages, where manual inspection revealed multiple genes involved in genomic mobility and recombination (recombinase, replication initiator protein, DNA binding protein, conjugal transfer protein, topoisomerase, helicase, relaxases, relaxosome proteins, helix-turn-helix transcriptional regulator and sigma 70 family RNA polymerase sigma factor), but none involving capsid or tail virus formation. These prophages have been detected in other P. micra isolates in different positions, although usually not duplicated. When the whole left prophage of PM89KC-AC-1 (~ 42 Kbp, position 564183...607520) was compared against OVD and the GPD (minimum identity ≥ 50%), good hits (e-value of 0 and bitscore over 3000) were found with a median identity of 89.226 ± 4.65% in OVD. The top hits belonged to a 42456 bp unclassified prophage (median identity of 89.77 ± 2.35%), which appears under different hosts such as Oribacterium, Fusobacterium, Streptococcus, Prevotella, Anaerosphaera and Peptoanaerobacter. The search did not return hits in the GPD. Additionally, putative virulence factors were detected in the prophages, including: an ABC transporter ATP-binding protein and a Type IV secretion system (including associated proteins PcfB and PrgI). Other relevant proteins found were a NlpC/P60 family protein in the left prophage and a CHAP domain containing protein in the right prophage. These prophages shared a large proportion of genes, but the left one had several extra proteins (Additional file; Table S2).

If KCOM 1037 strain is taken as a reference, the left prophage has been inserted into a CRISPR array of a type III-B CRISPR-Cas system in all P89 isolates (Additional file; Figure S2), separating the Cas proteins (CRISPR associated proteins) from the CRISPR array. However, the PM89KC-G and PM89KC-AC-1 isolates differ in the number and length of CRISPR arrays on both sides of the prophage. Isolate PM89KC-G-1 presents two CRISPR arrays after the prophage. The first one (G-A), of 298 bp, is located at position 598786, and the second one (G-B), of 423 bp, at position 601696. Isolate PM89KC-AC-1 presents these two arrays after the prophage: the first one (AC-B), of 2213 bp, located at position 607555, and second one (AC-C), of 423 bp, at position 612379 and an extra array (AC-A, position 563270, 749 bp) before the prophage (Additional file; Figure S2). Arrays G-A and G-B from PM89KC-G-1 correspond to arrays AC-B and AC-C in PM89KC-AC-1. The repeats in all these arrays share the same core structure (Additional file; Table S3). All spacers inside array G-B from PM89KC-G-1 are exactly identical to the ones in array AC-C in PM89KC-AC-1 and the last 3 spacers from AC-B are the ones in G-A. Furthermore, when searching for these spacers in other P. micra isolates, none contained the exact same ones, and were only shared in P89. If the subgingival isolates are taken as reference origin, the adenocarcinoma isolate could have increased the size of array G-A from 3 spacers to 32 and gained the extra array AC-A with 11 spacers (total net gain of 40 spacers). The first two spacers of AC-A on the left of the prophage for PM89KC-AC-1 were also detected before the prophage in PM89KC-G (G-Res), further supporting the common origin of these isolates. However, when comparing spacers from EYE_30 and PM79KC-G-1, many were identical, which indicates that sharing spacers across isolates is not uncommon and cannot be used as definitive proof. When the new spacers of PM89KC-AC-1 were searched against OVD and GPD, a couple in array AC-B matched unclassified phages in genus Coprococcus and Tyzzerella (both from family Lachnospiraceae) in the GPD. The rest matched oral phages in Parvimonas, Streptococcus and Fusobacterium in OVD or intestinal phages in Parvimonas in GPD. Finally, a similar insertion and duplication of this prophage has been seen in other P. micra such as PM102KC-G-1, but in a different CRISPR-Cas system (Type CAS-II-A/CAS-III-A) with different spacers and repeated sequences (Additional file; Table S3).

Potential virulence factors were analyzed in gingival and adenocarcinoma isolates of P89 using VFDB (Additional file; Tables S4 and S5). Most virulence factors were shared between PM89KC strains including multiple iron scavenging and transport proteins, type III, IV and VII secretion systems, colibactin toxins, neutrophil activating proteins, tissue adhesins, peptidases and biofilm regulators.

Interestingly, even though these isolates were > 99% identical, the phenotype of the gingival isolates extracted from P89 was considerably different to the adenocarcinoma isolate. The gingival P. micra colonies were completely white and compact with well-defined borders whereas the adenocarcinoma P. micra isolate colonies were translucent, brighter and had undefined borders. No hemolytic activity was observed for any strain (Additional file; Figure S3).

It is important to note that after the liver metastasis diagnosis for P89, our team collected, during the laparoscopy surgery, metastatic and non-metastatic liver samples in order to culture both specimens and to perform a 16S rRNA metabarcoding analysis. However, efforts to isolate P. micra in the liver were unsuccessful.

Thus, given that the above results suggest a possible same origin of the adenocarcinoma and the gingival P. micra isolates of P89, we decided to focus on a complete 16S metabarcoding and metatranscriptomic analysis of the different samples from this patient.

Microbiome of patient 89

To characterize the microbiome of the CRC P89 we performed a deep analysis of the bacterial 16S rDNA in feces, saliva, subgingival fluid and non-neoplastic, transition and adenocarcinoma tissues as well as in metastatic and non-neoplastic liver regions (Fig. 1).

The microbiome analysis of feces (M89-F) revealed that Faecalibacterium (26.16%), Enterococcaceae bacterium RF39 (20.73%), Eubacterium coprostanoligenes group (9.02%), Collinsella (7.71%) Ruminococcus gnavus group (7.61%), Lachnospiraceae (2.86%) and Bacteroides (2.47%) were the most abundant bacteria (Fig. 6).

Additionally, the taxonomic assignment of the saliva sample (M89-S) sequences showed that the most common genera were Streptococcus (38.04%), Neisseria (12.36%), Capnocytophaga (6.83%) and Granulicatella (4.97%) followed by Gemella (4.40%), Fusobacterium (4.33%) and Porphyromonas (4.24%). When gingival crevicular fluid sample (M89-G) was analyzed, the genus Fusobacterium (26.30%), Porphyromonas (24.73%), Prevotella (9.15%), Dialister (9.05%), Tannerella (4.94%), Parvimonas (2.83%) and a member of the Peptostreptococcaceae family (Peptostreptococcaceae bacterium oral taxon 113 str. W5053 at 3.93%) were found as the most abundant bacteria (Fig. 6). In agreement with the bad periodontal health of the patient, two bacteria species pertaining to the red complex group, P. gingivalis and Tannerella forsythia, were detected in both M89-S and M89-G samples, as well as other important periodontal pathogen species (some of them belonging to orange and green complex groups) such as Porphyromonas endodontalis, Prevotella intermedia, Dialister pneumosintes, Eubacterium nodatum or Mogibacterium timidum (Fig. 6).

Fresh tissue adenocarcinoma sample (M89-FT-Ac) collected during the laparoscopy surgery was also analyzed. The tumor microbiome, mostly shaped by anaerobic bacteria, was composed of five major genera: Bacteroides (31.69%), Fusobacterium (25.62%), Peptostreptococcus (6.49%), Prevotella (5.17%) and Hungatella (4.16%) (Fig. 6). In addition, transition tissue (M89-FT-Tr) and non-neoplastic colon mucosa tissue (M89-FT-NC) were collected in order to understand the microbe evolution in the colon mucosa. The analysis of the PM89-FT-Tr sample revealed that its microbiome was mainly composed by Bacteroides (45.80%), Faecalibacterium (13.92%), Lachnospiraceae (9.03%), Ruminococcus gnavus group (4.51%), Ruminococcus torques group (2.02%), Barnesiella (1.91%), Faecalitalea (1.64%), Peptostreptococcus (1.49%) and Prevotella (1.45%). Besides, the M89-FT-NC microbiome was composed by similar bacteria but with a significantly different abundance when compared to transition and adenocarcinoma tissues (Fig. 6). For example, the bacteria B. fragilis, over-represented in Ac sample (25.17%), showed low abundance in normal colon tissue sample (1.66%) and also in M89-FT-Tr tissue (1.72%). This tendency was also observed for other microorganisms such as Peptostreptococcus anaerobious, which was over-represented in adenocarcinoma (5.58%) and low abundant in transition (0.70%) and normal (0.82%) colon tissues. P. intermedia was the third most abundant species identified in adenocarcinoma (5.17%) but it was less frequent in transition (1.45%) and in normal (1.19%) mucosa tissues. This happened also for C. showae and Streptococcus agalactiae, which showed higher abundance in M89-FT-Ac sample (2.96% and 0.99%, respectively), while their presence in M89-FT-Tr tissue (0.08% and 0.11%, respectively) or in M89-FT-NC tissue (0.05% and 0.09%, respectively) was very low. In contrast, the microorganism B. dorei appeared with higher relative abundance in M89-FT-Tr tissue (21.73%) or in M89-FT-NC tissue (14.00%) compared to M89-FT-Ac (3.06%) (Fig. 6).

The microbiome analysis performed in the liver sample revealed that Streptococcus (20.91%), Pseudomonas (15.46%), Haemophilus (10.61%), Staphylococcus (6.75%) and Veillonella (5.54%) were the main genera found in the non-neoplastic adjacent tissue (M89-FT-NL) while in the metastatic region (M89-FT-MetL), Pseudomonas (33.74%), Streptococcus (10.51%), Bacteroides (6.63%) and Rothia (4.34%) were the most represented genera.

It is important to note that periodontal pathogens genera, such as Capnocytophaga, Parvimonas, Prevotella and Eubacterium, were detected, in some cases with low abundance, in liver samples of P89.

Parvimonas was found in colon in adenocarcinoma, non-neoplastic and transition tissues (1.24%, 0.65% and 0.27%, respectively) being enriched in cancerous tissue. Parvimonas was also present in feces (1.00%), saliva (1.34%) and subgingival crevicular sample (2.83%). In the liver sample Parvimonas was detected with a relative abundance of 3.18%. P. micra was grown after culturing gingival and colorectal adenocarcinoma but no P. micra colonies were obtained after culturing the liver tissue.

As commented before, other typical oral pathogens were detected in all types of samples, being the most abundant Fusobacterium, Prevotella, Campylobacter and Dialister. Figure 6B shows bacteria shared between gingival and adenocarcinoma samples.

FFPE samples revealed a similar bacterial composition (Fig. 7). Typical oral microbes were detected in those FFPE samples, as well as in non-paraffin embedded samples, such as Prevotella (PT-NC: 0.87%; PT-A: 1.60%; PT-Ac: 1.08%), Fusobacterium (PT-NC: 0.37%; PT-A: 0.70%; PT-Ac: 26.50%), Dialister (PT-NC: 0.01%; PT-Ac: 0.04%), Actinomyces (PT-NC: 0.16%; PT-A: 0.31%; PT-Ac: 0.20%), Gemella (PT-NC: 0.15%; PT-A: 1.03%; PT-Ac: 0.22%) and Parvimonas (PT-NC: 0.02%; PT-A: 0.13%; PT-Ac: 0.01%).

Metatranscriptomics of tumor tissues

In order to evaluate bacterial activity in the tumor tissue of P89, where Parvimonas was isolated, a metatranscriptomic analysis was performed from colon tissues (Fig. 1). In particular, we analyzed the adenocarcinoma tissue (T89-FT-Ac), as well as non-neoplastic colon mucosa tissue (T89-FT-NC) and transition tissue (interface between non-neoplastic and adenocarcinoma region, T89-FT-Tr).

At the species taxonomic level, a higher richness and diversity were observed in the non-neoplastic tissue compared to the adenocarcinoma sample (Table S6). Microbial activity profile was different at the species level when these samples were compared (Fig. 8A). The active microbiota in the non-neoplastic colon tissue (T89-FT-NC) and colon transition tissue (T89-FT-Tr) from P89 was dominated by B. dorei (FT-NC: 25.55%, FT-Tr: 39.53%, FT-Ac: 5.68%), Faecalibacterium prausnitzii (FT-NC: 6.06%, FT-Tr: 6.94%, FT-Ac: 0.87%), Faecalicatena gnavus (FT-NC: 8.35%, FT-Tr: 3.29%, FT-Ac: 1.27%) and Bacteroides thetaiotaomicron (FT-NC: 2.83%, FT-Tr: 6.72%, FT-Ac: 0.59%). In contrast, B. fragilis (FT-N: 1.65%, FT-Tr: 2.74%, FT-Ac: 12.94%), Peptostreptococcus anaerobius (FT-NC: 0.73%, FT-Tr: 0.78%, FT-Ac: 10.29%) and Fusobacterium polymorphum (FT-N: 0.3%, FT-Tr: 0.23%, FT-Ac: 8.96%) dominated the adenocarcinoma tissue. In addition, other oral associated species like Fusobacterium species, P. intermedia and P. micra also accounted for a higher percentage of activity in adenocarcinoma than in the non-neoplastic tissue. Parvimonas was detected by 16S rRNA gene sequencing and also metatranscriptomic analysis in the three different tissues. Moreover, Parvimonas, represented a higher percentage of transcripts (MTT) and a higher ratio of percentage of transcripts vs the relative abundance in 16S rRNA metabarcoding (16S) (MTT/16S: 1.58 in T89-FT-NC, 2.5 in T89-FT-Tr and 3.33 in T89-FT-Ac) in the adenocarcinoma tissue than in the other tissues (Fig. 9A).

Since no replicates were obtained, no statistical tests could be performed. Therefore, we considered a gene to be overexpressed when it had a mean abundance (RPKM) higher than 5 and a difference in abundance (log2[foldchange]; log2FC) > 1). According to these criteria, we found 16 genes that were overexpressed in non-neoplastic tissue and 14 genes in the adenocarcinoma (Fig. 8B).

Interestingly, among those overexpressed in the adenocarcinoma we found some stress indicators such as dps; starvation-inducible DNA-binding protein or arsR transcriptional regulator (implicated in ion homeostasis, biofilm formation, primary and secondary metabolism, response to adverse condition, and virulence). In addition, increased expression of ompW, an outer membrane protein which acts as a receptor for colicin S4 (colicins are plasmid-encoded toxic proteins produced by Escherichia coli strains), or the tcdAB, toxin A/B (pro-inflammatory and cytotoxic, causing disruption of the actin cytoskeleton and impairment of tight junctions in human intestine) were also found.

Other genes related with metabolism like different metal transporters (copA, Cu+-exporting ATPase; cbiN; cobalt/nickel transport protein), enzymes involved in carbon metabolism (ACADS; butyryl-CoA dehydrogenase, pycB; pyruvate carboxylase subunit B, gcvH; glycine cleavage system H protein), and a glutamate dehydrogenase (gudB) that allows the use of glutamate as a carbon source were also more expressed in adenocarcinoma. Meanwhile, in non-neoplastic tissue there are other several genes overexpressed that code for proteins involved in carbon metabolism (mdh, malate dehydrogenase; PGD,6-phosphogluconate dehydrogenase; G6PD glucose-6-phosphate 1-dehydrogenase), in amino-sugar metabolism (nagB), as well as two subunits of ribose transporter (rbsB, rbsC), atpE a F-type H⁺-transporting ATPase (used by aerobic organisms for synthesizing ATP) and SOD2 superoxide dismutase (which neutralizes toxic levels of reactive oxygen species).

Focusing on the transcriptional profile of Parvimonas in P89, a total of 808 different KEGG genes were assigned. The ones with higher expression were genes that codifies for several ribosome proteins, DNA replication proteins (hupA, ssb) transcription machinery (rpoA, nusG) and translation factors (tuf, fusA, infA) confirming that Parvimonas was transcriptionally active (Fig. 9B). Genes related to the metabolism of carbohydrates (gcvH, pflD, galE), amino acids (prdA, kbl, trxB) and proteins (nlpC, mltA, plsX) were also found. In fact, the most expressed gene by P. micra in adenocarcinoma was a probable lipoprotein (nlpC) that was also among the more overexpressed in this tissue globally. Further annotation of this gene with Pyre2 showed that only the last part of the protein (119 residues) was similar to the putative cell wall hydrolase (autolysin acd24020 catalytic domain, that belongs to the NlpC/P60 family) from Clostridium difficile (Table S7).

Regarding the potential virulence factors detected in the genomic analysis of isolates we found transcripts of 13 out of 25 genes identified (considering those with a nucleotide identity > 40%). All of them were more expressed in adenocarcinoma with the exception of clpP (more expressed in transition tissue) and LpxC-fabZ (more expressed in non-neoplastic tissue) (Fig. 9C).

In this work, culturomic approaches allowed us to isolate P. micra from gingival crevicular fluid and adenocarcinoma samples from CRC patients, and the corresponding bacterial genomes were fully sequenced using two different technologies (Illumina and Oxford Nanopore). Even though culturomic procedures were done for a big set of CRC diagnosed patients, P. micra isolates were only obtained from five CRC patients and not in all cases for all types of samples, probably due to the fastidious nature of this GPAC.

Whole genome analyses on gingival and adenocarcinoma P89 isolates evidenced that gingival-adenocarcinoma pairs were > 99% identical at nucleotide level and had a high degree of synteny. Interestingly, two prophages were detected in all PM89KC isolates. When comparing these prophages in PM89KC-Ac-1 they had a pairwise identity of 84% in matching regions. These prophages have been found in other typically oral genera such as Oribacterium, Fusobacterium, Streptococcus, Prevotella, Anaerosphaera, Lachnoanaerobaculum, Peptoanaerobacter and Catonella. They have also been found in other Parvimonas isolates, located in different places across the genome, although usually not duplicated. One of these prophages is inserted in a CRISPR array in P89. Furthermore, the acquisition of prophages can also prevent invasion by other phages [88] or give the microbe some advantages (higher virulence, major adhesion or biofilm formation ability) [89]. Additionally, the adenocarcinoma isolate has an extra CRISPR array and more spacers when compared to the subgingival isolates. Spacers contain pieces of phages and are used by the CRISPR-Cas system as a defense mechanism against these organisms. Two of the new spacers in PM89KC-AC-1 have high similarity to unidentified phages in genus Coprococcus and Tyzzerella (both from family Lachnospiraceae) in the GPD. There are 8 spacers with 100% identity between all P89 isolates, which would be congruent with a putative common origin.

In summary, the high OrthoANI values obtained, the high synteny observed and the presence of a same type III-B CRISPR-Cas system with some shared spacers, together with the same duplicated prophages in paired PM89KC isolates strongly suggest that both gingival and adenocarcinoma isolates have a recent common ancestor, and that P. micra seems to be able to travel from periodontal pockets to other locations such as the gut. In the colon, a great abundance of different phages [90] could have induced the increase of CRISPR-Cas defenses, enlarging the number of spacers observed in the PM89KC-AC-1 isolate. Additionally, it is possible that the duplication of the prophage could provide new advantages to P. micra in the gut, as several studies have reported with other microbes in biofilm formation or defense against other viruses [88, 89, 91]. For example, several virulence factors were detected in the PM89KC prophages including a Type IV secretion system and some of its associated proteins like PcfB and PrgI [92]. Other relevant proteins are the NlpC/P60 family protein in the left prophage which contributes to cell wall remodeling, and a CHAP domain containing protein in the right phage which is related to virulence [93, 94].

Therefore, the results support that P. micra, a strictly anaerobic bacterium commonly found in the oral cavity, and overabundant in patients with periodontal diseases [15, 17], travels from the subgingival pocket to the colon. Similar results were also observed by using an animal model approach for the periodontal pathogen F. nucleatum [26]. Regarding P79 where P. micra were isolated also in gingival and tumor samples, paired isolates obtained showed a low percentage of identity (OrhoANI values: 96.84%-97.02%). This fact indicates that PM79KC-AC and PM79KC-G isolates obtained from P79 were different strains, although an old event of bacterial translocation leading to higher sequence divergence cannot be discarded.

When fastidious anaerobic bacteria, such as Parvimonas or other periodontal pathogens, arrive to the gut, they can settle at the base of villi or in intestinal crypts of Lieberkühn as well as in dysplastic structures of adenomas or adenocarcinomas where the oxygen pressure is low [95]. Consequently, the original oral P. micra could undergo different genomic rearrangements, accumulate mutations and promote changes in gene expression to adapt to its new niche such as the ones observed after the adenocarcinoma genome analysis performed. One of the most apparent genomic changes include the deletion or inactivation of different genes, such as several transporters, as well as the increase of insertion elements, which are both typical features of bacterial adaptation to a new niche [96, 97].

In addition to the genetic differences, it is important to note that the colony appearance of the PM89KC isolates was quite different when gingival strains and the adenocarcinoma isolate were compared. Different phenotypes of P. micra have been previously described, being one type with hemolytic capacity and high adherent properties the most prevalent in CRC feces samples [25]. The PM89KC-AC-1 isolate had translucent, bright and undefined border colonies with no hemolytic activity whereas the oral colonies were white and compact with well-defined borders, which could be related to its adaptation and survival in the colon mucosa.

Periodontitis is a chronic inflammatory disease caused by a multispecies community of periodontal pathogens affecting dental supporting tissues [98]. Although bacterial plaque is the primary etiologic factor, progression and clinical characteristics of these diseases are influenced by acquired, local, systemic, and genetic factors that can modify susceptibility to this polymicrobial infection [87, 99]. In the concrete case of P89, the oral health assessment performed in this work by a dentist allowed us to know that this patient presented a periodontal disease in stage IVB. In the preliminary interview, P89 reported having had several oral surgeries in recent years, as a consequence of her periodontal disorders. Previous studies have proven 3 potential pathways for oral microbes to reach the gut niche: (A) as a transient bacteremia during dental procedures, such as tooth extraction, dental cleaning or different oral surgeries; (B) as a consequence of the direct contact between the subgingival biofilm and the blood vessels in the deep pockets of patients with periodontitis; or (C) through the oral-gut axis during meals or during daily swallows of saliva, considering that a human could swallow about 1500–2000 times a day [26, 100]. Therefore, P. micra PM89KC translocation from the oral cavity to the colon, could have been facilitated during P89 oral surgeries or during bleeding caused through daily dental cleaning, due to her advanced stage of periodontal disease.

In multiple studies, periodontal conditions have been related to other systemic diseases such as diabetes [101], cardiovascular diseases [102], Alzheimer’s disease [103, 104], and rheumatoid arthritis [105] and also to certain types of cancer [3, 5, 23, 26, 27, 95]. It is increasingly being proven that the dissemination of pathobionts from the oral cavity to distal areas of the human body implies that these bacteria can exert harmful functions on human cells by colonizing these new ecological niches. In 2020, a study demonstrated in a mouse model that F. nucleatum, reached colorectal tumors coming from the oral cavity via intravenous route [26]. In addition, authors isolated different F. nucleatum strains from saliva and adenocarcinoma of CRC patients, demonstrating that fusobacteria found in carcinomas migrated from the oral cavity [26] as we have proposed for P. micra in the present study.

Moreover, several genomic features of P. micra, like the small genome size or the high A + T content, suggest that this species could have an intracellular lifestyle [85, 86]. Thus, P. micra could be able to migrate from the oral cavity to other localizations of the human body inside epithelial cells avoiding the human immune system as previously suggested [84]. Besides, it is interesting to note that nine genes related to cell membrane transport were lost by the adenocarcinoma isolate (PM89KC-AC) in comparison to gingival isolates. This genomic process, linked to the genome reduction that was detected when both oral and colon P. micra isolates were compared, indicate that this oral pathobiont changed its original ecological niche for a new colon niche, supporting our hypothesis that P. micra could be intracellular [96, 97].

Furthermore, oral, gut and carcinoma microbiota of P89 was analyzed by 16S rRNA sequencing using samples of different nature. The analysis of adenocarcinoma and oral samples performed in the current study showed a clear over-representation of DNA from oral bacteria in CRC tissue, as reported in previous studies [4, 13, 27, 106]. Additionally, we analyzed and compared the transcriptomic profile of P. micra in the control, non-neoplastic region vs the colon adenocarcinoma with the aim of identifying genes that could be related to cancer development and progression. Two lines of evidence support that the detected DNA corresponded to active, viable oral pathogen bacteria. Firstly, bacterial isolates of P. micra were obtained from fresh tumor samples, and secondly, the metatranscriptomic analysis of the adenocarcinoma sample confirmed that these bacteria are not only transcriptionally active but also at higher levels in the RNA pool than on the DNA-based analysis. Moreover, it was found that many of the bacteria which are more active in the tumor are oral pathogens. Taking the 16S rRNA metabarcoding and the metatranscriptomic analysis into account, we suggest that P. micra could arrive to the gut accompanied by other oral pathobionts, such as Fusobacterium, through the formation of polymicrobial aggregates. Micro-communities of bacteria are naturally found in human saliva and are composed by both aerobe and anaerobe microbes [107–109]. These bacterial aggregates appear to be able to grow and form biofilms more efficiently than if they travel as individual sessile cells [109]. Metabolic interactions and co-operation between different genera could increase survival of fastidious bacteria, such as P. micra, during these translocation processes [107, 108]. This, together with the fact that oral bacteria were found as transcriptionally active in adenocarcinoma could explain why different subgingival pathogens (such as Parvimonas, Fusobacterium, Actinomyces, Campylobacter or Dialister among others) were detected in adenoma and adenocarcinoma samples through 16S rRNA metabarcoding bioanalysis. It has to be borne in mind that these oral microbes could promote a gut inflammatory effect, individually, as reported previously for P. micra [110] and for other bacteria [5, 9, 95, 111, 112] or synergistically when living in communities or aggregates [107].

Focusing only on P. micra results, transcriptomic bioanalysis confirmed that the PM89KC strain was more active in the adenocarcinoma than in the non-neoplastic distant tissue, supporting that this pathobiont is not only present in dysplastic tissues (revealed by 16S rRNA metabarcoding analysis) but also viable (cultured at the laboratory) and active (demonstrated by metatranscriptomic analysis), so its potential role in cancer development should be further explored. Remarkably, the gene most expressed by P. micra in the adenocarcinoma was a probable lipoprotein nlpC that was also among the more overexpressed in the carcinoma tissue globally. NlpC/P60 domains are bacterial peptidoglycan hydrolases that cleave noncanonical peptide linkages and contribute to cell wall remodeling [113], and its potential role in the adenocarcinoma tissue should be further explored. In addition, another highly expressed gene from Parvimonas was mltA (membrane-bound lytic murein transglycosylase A), which also degrades murein. This suggests that Parvimonas was remodeling its peptidoglycan to facilitate cell growth or it could be also a strategy of pathogenesis to resist human degradative enzymes or release of cytotoxic muropeptides.

Among the potential virulence effects of P. micra, its proteolytic potential could be relevant. Previous studies reported that endogenous proteolytic activity of P. micra facilitates bacteria dissemination into periodontal tissues but also to blood vessels [114]. Furthermore, this pathobiont can regulate and activate the proteolytic activity of other key oral pathogens (belonging to the red and orange complexes). For example, an in vitro study showed that P. micra stimulates the biosynthesis of proteolytic gingipains, enhances growth and coaggregates with the well-known pathogen P. gingivalis [17]. Moreover P. micra showed synergic biofilm formation and coaggregation with F. nucleatum, another periodontal pathogen associated with CRC [15]. An in vivo research work showed that P. micra has pathogenic synergy with P. intermedia and Prevotella nigrescens, showing higher transmissibility of infection, and that also enhanced Prevotella growth and oral abscesses aggravation [115]. Furthermore, another study demonstrated that P. micra coaggregates with another well-characterized oral pathogen: Treponema denticola [116]. P. micra was also shown to shoot up carcinogenesis in ApcMin/+ mice and also in germ free mice, by increasing the expression of pro-inflammatory cytokines and promoting the proliferation of colon cells [24]. This interesting work also reported that P. micra could act as a poor survival biomarker in CRC patients, after the analysis of the association of P. micra abundance in fecal samples with clinical data [24].

Besides, in the present study we detected the presence of Parvimonas DNA, among other periodontal pathogens (such as Capnocytophaga, Prevotella and Eubacterium) in liver tissues of P89. A recent study also reported the presence of typical gut microbes in liver samples of CRC patients that underwent liver metastasis, suggesting that a pre-metastatic niche may be constructed by gut bacteria, allowing tumoral cells to develop secondary tumors in the liver [9]. While it is true that Parvimonas was not detected by 16S rRNA metabarcoding in neoplastic liver tissues, Parvimonas among other periodontal pathogens were present in the non-neoplastic region of the liver tissue. There, P. micra and other pathobionts could promote liver inflammation by disrupting the normal function of the NOD2 signaling pathway [16]. However, it has to be highlighted that the methodology applied has serious limitations for small samples. Therefore the presence of P. micra in the metastatic liver tissue cannot be totally discarded. Additionally, it was previously reported that periodontal pathogens could arrive from the colon to the liver via circulatory system (by the portal vein) [9].

Considering that P. micra was associated with CRC [13, 14, 23, 117–119] and taking into account our results, we propose that P. micra may promote pro-tumoral colon inflammation and/or adenocarcinoma development in susceptible patients. The tumors appear to be a stressful environment (as shown by the overexpression of stress response genes in the tumor metatranscriptome) where bacteria produce toxins among different virulence factors. The isolation of highly related P. micra clones in the oral cavity and in the gut provides a unique opportunity to study its activity in the oral cavity vs the gut tissue. The potential role of these anaerobic bacteria to generate an inflammatory microenvironment and in the initiation or progression of the tumor should be investigated in-depth. More in vitro and in vivo studies are needed to elucidate the cell mechanisms of P. micra in cancer development, since its pathogenicity remains unclear. However, P. micra could be considered as a CRC biomarker detected in non-invasive samples such as saliva or feces. To conclude, it should be pointed out that periodontitis could be a risk factor for CRC development, and therefore we propose that early periodontal treatments and/or probiotics, prebiotics and antibiotics could contribute to reducing the risk of cancer development.

P. micra appears to be able to translocate, possibly aggregated with other oral pathobionts such as Fusobacterium, from the subgingival sulcus of the oral cavity to the colon, through the circulatory system or the oral-gut axis, where it has to adapt to survive in a new niche, undergoing genomic and transcriptomic rearrangements. The findings of the present study support the relation between periodontal pathogens and the development of CRC. This may be of great importance and demonstrates that proper oral health maintenance and an early detection of periodontal diseases could reduce the risk of CRC. Finally, we suggest that P. micra can be an interesting CRC biomarker. We hope that the current work stimulates further investigations about the P. micra role in cancer.

ASVs: Amplicon Sequence Variants

CARD: Comprehensive Antibiotic Resistance Database

Cas: CRISPR associated proteins

CBCT: Cone Beam Computed Tomography

CHUAC: University Hospital Complex of A Coruña

CMS1: Consensus Molecular Subtype 1

CRC: Colorectal Cancer

CRISPR: Clustered Regularly Interspaced Short Palindromic Repeats

EC: Enzymatic Cocktail

FFPE: Formalin-Fixed Paraffin-Embedded

gDNA: Genomic DNA

GI: Gastrointestinal

GPAC: Gram-Positive Anaerobic Cocci

GPD: Gut Phage Database

HUAC: Hospital Universitario de A Coruña

HUAC: University Hospital of A Coruña

iTOL: Interactive Tree of Life

KEGG: Kyoto Encyclopedia of Genes and Genomes

LSTs: Laterally Spreading Tumors

M89-F: Colorectal cancer patient 89, faeces microbiome

M89-FT-Ac: Colorectal cancer patient 89, adenocarcinoma microbiome

M89-FT-MetL: Colorectal cancer patient 89, metastatic liver tissue microbiome

M89-FT-NC: Colorectal cancer patient 89, normal colon tissue microbiome

M89-FT-NL: Colorectal cancer patient 89, normal liver tissue microbiome

M89-FT-Tr: Colorectal cancer patient 89, transition tissue microbiome

M89-G: Colorectal cancer patient 89, gingival crevicular fluid microbiome

M89-PT-A: Colorectal cancer patient 89, transition tissue microbiome (from FFPE tissue)

M89-PT-Ac: Colorectal cancer patient 89, adenocarcinoma microbiome (from FFPE tissue)

M89-PT-NC: Colorectal cancer patient 89, normal colon tissue microbiome (from FFPE tissue)

M89-S: Colorectal cancer patient 89, saliva microbiome

MALDI-TOF MS: Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

MTT: Percentage of transcripts (metatranscriptome)

OrthoANI: Average Nucleotide Identity by Orthology

OVD: Oral Virus Database

P102: Colorectal cancer patient number 102

P114: Colorectal cancer patient number 114

P79: Colorectal cancer patient number 79

P89: Colorectal cancer patient number 89

P94: Colorectal cancer patient number 94

RGI: Resistance Gene Identifier

RPKM: Number of reads normalized by length of the gene (Kb) and size of the dataset (Mb)

rRNA: Ribosomal ribonucleic acid

SMURF: Short MUltiple Regions Framework

SNPs: Single nucleotide polymorphisms

T89-FT-Ac: Colorectal cancer patient 89, adenocarcinoma transcriptome

T89-FT-NC: Colorectal cancer patient 89, normal colon tissue transcriptome

T89-FT-Tr: Colorectal cancer patient 89, transition tissue transcriptome

UHGG: Unified Human Gastrointestinal Genome

VFDB: Virulence Factor Database

AVAILABILITY OF SUPPORTING DATA

Genomes from each P. micra isolate sequenced in this work were deposited in the NCBI database with GenBank Assembly accession code shown in Table 1. The datasets analyzed in this research project for 16S rRNA metabarcoding and metatranscriptome are available from the corresponding authors on reasonable request, not publicly available due to ethical reasons.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

To satisfy any ethical or legal consideration, the study was carried out adhering to the standards of good clinical practice and current research regulations included in Law of Biomedical Research 14/2007, in accordance with the principles derived from the latest version of the Declaration of Helsinki and of the Convention on Human Rights and Biomedicine (the Oviedo Convention). Compliance with the protection of personal data of all those involved in the RGPD – UE 2016/679, LOPDGDD 3/2018 Ley 41/2002 and its implementing regulations, Royal Decree 1720/2007, were enforced. This study, that belongs to the project PI20/00413 (ISCIII, Spain), has been approved by the autonomic Research Ethical Committee of Galicia (CEIm-G 2018/609) and the Spanish Agency for Medicines and Healthcare Products (AEMPS) for the use of human samples from CRC patients of the University Hospital Complex of A Coruña (CHUAC, A Coruña, Galicia, Spain). Signed informed Biobank consents and sample storage were managed by the Biobank of the University Hospital of A Coruña, UNE-EN ISO 9001-2015 certified, which ensured the traceability and quality of samples for research use. Clinical data presented in this work was obtained from the repository of Servizo Galego de Saúde (SERGAS) by medical personnel of HUAC.

CONSENT FOR PUBLICATION

We have the consent of all the CRC patients included in this project for the publication of results in scientific articles. All authors consent to the publication of this manuscript.

COMPETING INTERESTS

No conflicts of interest to declare by the authors of this research project.

FUNDING

This study has been funded by Instituto de Salud Carlos III through the project PI20/00413 to MP (Co-funded by European Regional Development Fund/European Social Fund "A way to make Europe"/"Investing in your future"), by project RTI2018-102032-B-I00 from the Spanish Ministry of Science and Innovation to AM and by the CIBER INF ISCIII (CB21/13/00055) to GB and MP. Biobank of A Coruña was supported by Instituto de Salud Carlos III – Fondos FEDER (EU) by grant PT20/00128. K. Conde-Pérez was financially supported with a predoctoral fellowship by the Asociación Española Contra el Cáncer (AECC). E. Buetas is supported by a grant from the Spanish Ministry of Science and Innovation with the reference PRE2019-088126. S. Ladra, E. Martin-De Arribas and Iago Iglesias-Corrás are supported by grants from GAIN (Xunta de Galicia, Spain) with references ED431C 2021/53 and ED431G 2019/01.

AUTHOR CONTRIBUTIONS

J.A.V, M.P, A.M, S.L. and G.B conceived and designed the study. M.P, J.A.V and A.M. supervised the study. K.C-P, J.A.V and M.P obtained fecal and saliva samples. M.P managed the informed consent from patients. J.F.N obtained fresh tissue samples from surgery. L.S.E, B.O-A and A.C collected and studied FFPE samples. K.C-P, N.T-T, M.N-A, J.A.V and S.R-F processed samples and performed 16S rRNA metabarcoding and whole genome sequencing. E.B, M.C-D and A.M performed transcriptomic analysis. P.A-M, E.M-D.A, I.I-C, K.C-P, J.A.V and S.L made bioinformatics analysis related to DNA sequencing. K.C-P, P.A-M, E.B. and E.M-D.A designed and created tables, figures 1-9, additional files and graphical abstract of the manuscript. S.P-L obtained subgingival samples and made dental checkups for patients. I.G-R, N.M-L and L.A.A (RIP) followed up oncological patients. K.C-P, J.A.V, M.P, M.C-D, A.M, E.B, P.A-M, A.C, B.O-A, S.P-L, E.M-D.A, I.G-R, N.M-L and S.L wrote the main manuscript text. All authors reviewed the manuscript.

ACKNOWLEDGMENTS

The present authors want to thank all the colorectal cancer patients of the University Hospital of A Coruña for participating in this study despite their health problems. We warmly want to thank Gema Carro Díaz and Montserrat Ingelmo Sánchez (surgical service nurses of HUAC) for their support during patient’s recruitment time and also with sample collection. Moreover, we want to appreciate the valuable assistance received in anaerobic culturing by David Velasco Fernández (microbiologist) and Ana María Fernández Liñares (technician), and in FFPE samples processing by the pathology service technicians: Diego Barco Díaz, Cristina Vázquez Costa and Ana María Mejuto Rial. This work would not be possible without all the professionals of the Microbiology, Surgery, Pathology and Oncology services and Biobank from CHUAC who support this microbiome and cancer research project.

Wong SH, Yu J. Gut microbiota in colorectal cancer: mechanisms of action and clinical applications. Nature Reviews Gastroenterology & Hepatology. 2019;16:690–704. doi: 10.1038/s41575-019-0209-8.
Fan X, Jin Y, Chen G, Ma X, Zhang L. Gut Microbiota Dysbiosis Drives the Development of Colorectal Cancer. Digestion. 2021;102:508–15. doi: 10.1159/000508328.
Flemer B, Warren RD, Barrett MP, Cisek K, Das A, Jeffery IB, et al. The oral microbiota in colorectal cancer is distinctive and predictive. Gut. 2018;67:1454–63. doi: 10.1136/gutjnl-2017-314814.
Mohammadi M, Mirzaei H, Motallebi M. The role of anaerobic bacteria in the development and prevention of colorectal cancer: A review study. Anaerobe. 2021:102501. doi: 10.1016/j.anaerobe.2021.102501.
Parhi L, Alon-Maimon T, Sol A, Nejman D, Shhadeh A, Fainsod-Levi T, et al. Breast cancer colonization by Fusobacterium nucleatum accelerates tumor growth and metastatic progression. Nature Communications. 2020;11:3259. doi: 10.1038/s41467-020-16967-2.
Purcell RV, Visnovska M, Biggs PJ, Schmeier S, Frizelle FA. Distinct gut microbiome patterns associate with consensus molecular subtypes of colorectal cancer. Scientific Reports. 2017;7:11590. doi: 10.1038/s41598-017-11237-6.
Zhang J, Tao J, Gao RN, Wei ZY, He YS, Ren CY, et al. Cytotoxic T-Cell Trafficking Chemokine Profiles Correlate With Defined Mucosal Microbial Communities in Colorectal Cancer. Frontiers in Immunology. 2021;12:715559. doi: 10.3389/fimmu.2021.715559.
Nejman D, Livyatan I, Fuks G, Gavert N, Zwang Y, Geller LT, et al. The human tumor microbiome is composed of tumor type–specific intracellular bacteria. Science. 2020;368:973–80. doi: 10.1126/science.aay9189.
Bertocchi A, Carloni S, Ravenda PS, Bertalot G, Spadoni I, Lo Cascio A, et al. Gut vascular barrier impairment leads to intestinal bacteria dissemination and colorectal cancer metastasis to liver. Cancer Cell. 2021;39:708 – 24 e11. doi: 10.1016/j.ccell.2021.03.004.
Yang Y, Nguyen M, Khetrapal V, Sonnert ND, Martin AL, Chen H, et al. Within-host evolution of a gut pathobiont facilitates liver translocation. Nature. 2022. doi: 10.1038/s41586-022-04949-x.
Xi Y, Xu P. Global colorectal cancer burden in 2020 and projections to 2040. Translational Oncology. 2021;14:101174. doi: 10.1016/j.tranon.2021.101174.
Alves JM, Prado-López S, Tomás L, Valecha M, Estévez-Gómez N, Alvariño P, et al. Clonality and timing of relapsing colorectal cancer metastasis revealed through whole-genome single-cell sequencing. Cancer letters. 2022;543:215767. doi: 10.1016/j.canlet.2022.215767.
Osman MA, Neoh HM, Ab Mutalib NS, Chin SF, Mazlan L, Raja Ali RA, et al. Parvimonas micra, Peptostreptococcus stomatis, Fusobacterium nucleatum and Akkermansia muciniphila as a four-bacteria biomarker panel of colorectal cancer. Scientific Reports. 2021;11:2925. doi: 10.1038/s41598-021-82465-0.
Xu J, Yang M, Wang D, Zhang S, Yan S, Zhu Y, et al. Alteration of the abundance of Parvimonas micra in the gut along the adenoma-carcinoma sequence. Oncology Letters. 2020;20:106. doi: 10.3892/ol.2020.11967.
Horiuchi A, Kokubu E, Warita T, Ishihara K. Synergistic biofilm formation by Parvimonas micra and Fusobacterium nucleatum. Anaerobe. 2020;62:102100. doi: 10.1016/j.anaerobe.2019.102100.
Marchesan J, Jiao Y, Schaff RA, Hao J, Morelli T, Kinney JS, et al. TLR4, NOD1 and NOD2 mediate immune recognition of putative newly identified periodontal pathogens. Molecular Oral Microbiology. 2016;31:243–58. doi: 10.1111/omi.12116.
Neilands J, Davies JR, Bikker FJ, Svensäter G. Parvimonas micra stimulates expression of gingipains from Porphyromonas gingivalis in multi-species communities. Anaerobe. 2019;55:54–60. doi: 10.1016/j.anaerobe.2018.10.007.
Ryan PM, Shin CP. Native joint infections caused by Parvimonas micra. Anaerobe. 2021;71:102412. doi: 10.1016/j.anaerobe.2021.102412.
Gomes BP, Berber VB, Kokaras AS, Chen T, Paster BJ. Microbiomes of Endodontic-Periodontal Lesions before and after Chemomechanical Preparation. Journal of endodontics. 2015;41:1975–84. doi: 10.1016/j.joen.2015.08.022.
Colombo AP, Boches SK, Cotton SL, Goodson JM, Kent R, Haffajee AD, et al. Comparisons of subgingival microbial profiles of refractory periodontitis, severe periodontitis, and periodontal health using the human oral microbe identification microarray. Journal of periodontology. 2009;80:1421–32. doi: 10.1902/jop.2009.090185.
Watanabe T, Hara Y, Yoshimi Y, Fujita Y, Yokoe M, Noguchi Y. Clinical characteristics of bloodstream infection by Parvimonas micra: retrospective case series and literature review. BMC Infectious Disease. 2020;20:578. doi: 10.1186/s12879-020-05305-y.
Lowenmark T, Lofgren-Burstrom A, Zingmark C, Eklof V, Dahlberg M, Wai SN, et al. Parvimonas micra as a putative non-invasive faecal biomarker for colorectal cancer. Scientific Reports. 2020;10:15250. doi: 10.1038/s41598-020-72132-1.
Shen X, Li J, Li J, Zhang Y, Li X, Cui Y, et al. Fecal Enterotoxigenic Bacteroides fragilis - Peptostreptococcus stomatis - Parvimonas micra Biomarker for Noninvasive Diagnosis and Prognosis of Colorectal Laterally Spreading Tumor. Frontiers in Oncology. 2021;11:661048. doi: 10.3389/fonc.2021.661048.
Zhao L, Zhang X, Zhou Y, Fu K, Lau HC-H, Chun TW-Y, et al. Parvimonas micra promotes colorectal tumorigenesis and is associated with prognosis of colorectal cancer patients. Oncogene. 2022. doi: 10.1038/s41388-022-02395-7.
Bergsten E, Mestivier D, Donnadieu F, Pedron T, Tsoumtsa L, Lemichez E, et al. Parvimonas micra, an oral pathobiont associated with colorectal cancer, epigenetically reprograms human primary intestinal epithelial cells. bioRxiv. 2022. doi: 10.1101/2022.05.14.491935.
Abed J, Maalouf N, Manson AL, Earl AM, Parhi L, Emgard JEM, et al. Colon Cancer-Associated Fusobacterium nucleatum May Originate From the Oral Cavity and Reach Colon Tumors via the Circulatory System. Frontiers in Cellular and Infection Microbiology. 2020;10:400. doi: 10.3389/fcimb.2020.00400.
Komiya Y, Shimomura Y, Higurashi T, Sugi Y, Arimoto J, Umezawa S, et al. Patients with colorectal cancer have identical strains of Fusobacterium nucleatum in their colorectal cancer and oral cavity. Gut. 2019;68:1335–7. doi: 10.1136/gutjnl-2018-316661.
Zheng DW, Dong X, Pan P, Chen KW, Fan JX, Cheng SX, et al. Phage-guided modulation of the gut microbiota of mouse models of colorectal cancer augments their responses to chemotherapy. Nature Biomedical Engineering. 2019;3:717–28. doi: 10.1038/s41551-019-0423-2.
Mohammadi M, Mirzaei H, Motallebi M. The role of anaerobic bacteria in the development and prevention of colorectal cancer: A review study. Anaerobe. 2022;73:102501. doi: 10.1016/j.anaerobe.2021.102501.
Kim D, Jung J-Y, Oh H-S, Jee S-R, Park SJ, Lee S-H, et al. Comparison of sampling methods in assessing the microbiome from patients with ulcerative colitis. BMC Gastroenterology. 2021;21:396. doi: 10.1186/s12876-021-01975-3.
Durbán A, Abellán JJ, Jiménez-Hernández N, Ponce M, Ponce J, Sala T, et al. Assessing gut microbial diversity from feces and rectal mucosa. Microbial Ecology. 2011;61:123–33. doi: 10.1007/s00248-010-9738-y.
Yao QW, Zhou DS, Peng HJ, Ji P, Liu DS. Association of periodontal disease with oral cancer: a meta-analysis. Tumour biology: the journal of the International Society for Oncodevelopmental Biology and Medicine. 2014;35:7073–7. doi: 10.1007/s13277-014-1951-8.
Chang Y-S, Hsu M-H, Tu S-J, Yen J-C, Lee Y-T, Fang H-Y, et al. Metatranscriptomic Analysis of Human Lung Metagenomes from Patients with Lung Cancer. Genes. 2021;12:1458. doi: 10.3390/genes12091458.
Feng Y, Ramnarine VR, Bell R, Volik S, Davicioni E, Hayes VM, et al. Metagenomic and metatranscriptomic analysis of human prostate microbiota from patients with prostate cancer. BMC Genomics. 2019;20:146. doi: 10.1186/s12864-019-5457-z.
Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Research. 2013;41:e1. doi: 10.1093/nar/gks808.
Andrews S. FastQC. http://www.bioinformatics.babraham.ac.uk/projects/fastqc. Accessed 11 April 2022.
Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nature Biotechnology. 2019;37:852–7. doi: 10.1038/s41587-019-0209-9.
Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet journal. 2011;17:10–2. doi: 10.14806/ej.17.1.200.
Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nature Methods. 2016;13:581–3. doi: 10.1038/nmeth.3869.
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Research. 2013;41:D590-6. doi: 10.1093/nar/gks1219.
Fuks G, Elgart M, Amir A, Zeisel A, Turnbaugh PJ, Soen Y, et al. Combining 16S rRNA gene variable regions enables high-resolution microbial community profiling. Microbiome. 2018;6:17. doi: 10.1186/s40168-017-0396-x.
Debelius JW, Robeson M, Hugerth LW, Boulund F, Ye W, Engstrand L. A comparison of approaches to scaffolding multiple regions along the 16S rRNA gene for improved resolution. bioRxiv. 2021:2021.03.23.436606. doi: 10.1101/2021.03.23.436606.
Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, et al. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biology. 2014;12:87. doi: 10.1186/s12915-014-0087-z.
McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PloS one. 2013;8:e61217. doi: 10.1371/journal.pone.0061217.
Villanueva RAM, Chen ZJ. ggplot2: Elegant Graphics for Data Analysis (2nd ed.). Measurement: Interdisciplinary Research and Perspectives. 2019;17:160-7. doi: 10.1080/15366367.2019.1565254.
Ihaka R, Gentleman R. R: A Language for Data Analysis and Graphics. Journal of Computational and Graphical Statistics. 1996;5:299–314. doi: 10.1080/10618600.1996.10474713.
Chen H, Boutros PC. VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics. 2011;12:35. doi: 10.1186/1471-2105-12-35.
Bushnell B. BBMap. https://sourceforge.net/projects/bbmap/. Accessed 10 April 2022.
Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30:2114–20. doi: 10.1093/bioinformatics/btu170.
Ewels P, Magnusson M, Lundin S, Käller M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics. 2016;32:3047–8. doi: 10.1093/bioinformatics/btw354.
Wick R. Porechop. https://github.com/rrwick/Porechop. Accessed 09 April 2022.
Wilk R. Filtlong. https://github.com/rrwick/Filtlong. Accessed 09 April 2022.
De Coster W, D’Hert S, Schultz DT, Cruts M, Van Broeckhoven C. NanoPack: visualizing and processing long-read sequencing data. Bioinformatics. 2018;34:2666–9. doi: 10.1093/bioinformatics/bty149.
Wick RR, Judd LM, Gorrie CL, Holt KE. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Computational Biology. 2017;13:e1005595. doi: 10.1371/journal.pcbi.1005595.
Hunt M, Silva ND, Otto TD, Parkhill J, Keane JA, Harris SR. Circlator: automated circularization of genome assemblies using long sequencing reads. Genome Biology. 2015;16:294. doi: 10.1186/s13059-015-0849-0.
Parks DH, Imelfort M, Skennerton CT, Hugenholtz P, Tyson GW. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. 2015;25:1043–55. doi: 10.1101/gr.186072.114.
Larsen MV, Cosentino S, Lukjancenko O, Saputra D, Rasmussen S, Hasman H, et al. Benchmarking of Methods for Genomic Taxonomy. 2014;52:1529–39. doi: doi:10.1128/JCM.02981-13.
Li W, O'Neill KR, Haft DH, DiCuccio M, Chetvernin V, Badretdin A, et al. RefSeq: expanding the Prokaryotic Genome Annotation Pipeline reach with protein family model curation. Nucleic acids research. 2021;49:D1020-d8. doi: 10.1093/nar/gkaa1105.
Alcock BP, Raphenya AR, Lau TTY, Tsang KK, Bouchard M, Edalatmand A, et al. CARD 2020: antibiotic resistome surveillance with the comprehensive antibiotic resistance database. Nucleic acids research. 2020;48:D517-D25. doi: 10.1093/nar/gkz935.
Bortolaia V, Kaas RS, Ruppe E, Roberts MC, Schwarz S, Cattoir V, et al. ResFinder 4.0 for predictions of phenotypes from genotypes. Journal of Antimicrobial Chemotherapy. 2020;75:3491–500. doi: 10.1093/jac/dkaa345.
Carattoli A, Zankari E, García-Fernández A, Larsen MV, Lund O, Villa L, et al. In silico detection and typing of plasmids using PlasmidFinder and plasmid multilocus sequence typing. Antimicrobial Agents and Chemotherapy. 2014;58:3895–903. doi: doi:10.1128/AAC.02412-14.
Chen L, Yang J, Yu J, Yao Z, Sun L, Shen Y, et al. VFDB: a reference database for bacterial virulence factors. Nucleic Acids Research. 2005;33:D325-D8. doi: 10.1093/nar/gki008.
Li S, Guo R, Zhang Y, Li P, Chen F, Wang X, et al. A catalog of 48,425 nonredundant viruses from oral metagenomes expands the horizon of the human oral virome. iScience. 2022;25. doi: 10.1016/j.isci.2022.104418.
Camarillo-Guerrero LF, Almeida A, Rangel-Pineros G, Finn RD, Lawley TD. Massive expansion of human gut bacteriophage diversity. Cell. 2021;184:1098 – 109.e9. doi: 10.1016/j.cell.2021.01.029.
Buchfink B, Reuter K, Drost H-G. Sensitive protein alignments at tree-of-life scale using DIAMOND. Nature Methods. 2021;18:366–8. doi: 10.1038/s41592-021-01101-x.
Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. doi: 10.1186/1471-2105-10-421.
Zhang F, Zhao S, Ren C, Zhu Y, Zhou H, Lai Y, et al. CRISPRminer is a knowledge base for exploring CRISPR-Cas systems in microbe and phage interactions. Communications Biology. 2018;1:180. doi: 10.1038/s42003-018-0184-6.
Mitrofanov A, Alkhnbashi OS, Shmakov SA, Makarova Kira S, Koonin Eugene V, Backofen R. CRISPRidentify: identification of CRISPR arrays using machine learning approach. Nucleic Acids Research. 2020;49:e20-e. doi: 10.1093/nar/gkaa1158.
Padilha VA, Alkhnbashi OS, Shah SA, de Carvalho ACPLF, Backofen R. CRISPRcasIdentifier: Machine learning for accurate identification and classification of CRISPR-Cas systems. GigaScience. 2020;9. doi: 10.1093/gigascience/giaa062.
Goris J, Konstantinidis KT, Klappenbach JA, Coenye T, Vandamme P, Tiedje JM. DNA-DNA hybridization values and their relationship to whole-genome sequence similarities. International journal of systematic and evolutionary microbiology. 2007;57:81–91. doi: 10.1099/ijs.0.64483-0.
Seemann T. Snippy. https://github.com/tseemann/snippy. Accessed 15 April 2022.
Bayliss SC, Thorpe HA, Coyle NM, Sheppard SK, Feil EJ. PIRATE: A fast and scalable pangenomics toolbox for clustering diverged orthologues in bacteria. GigaScience. 2019;8. doi: 10.1093/gigascience/giz119.
Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30:1312–3. doi: 10.1093/bioinformatics/btu033.
Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Research. 2021;49:W293-W6. doi: 10.1093/nar/gkab301.
Farrer RA. Synima: a Synteny imaging tool for annotated genome assemblies. BMC Bioinformatics. 2017;18:507. doi: 10.1186/s12859-017-1939-7.
Minkin I, Patel A, Kolmogorov M, Vyahhi N, Pham S. Sibelia: A Scalable and Comprehensive Synteny Block Generation Tool for Closely Related Microbial Genomes. In: Springer, editor. Algorithms in Bioinformatics; Berlin, Heidelberg. WABI: Springer Berlin Heidelberg; 2013. p. 215–29.
Schmieder R, Edwards R. Quality control and preprocessing of metagenomic datasets. Bioinformatics. 2011;27:863–4. doi: 10.1093/bioinformatics/btr026.
Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nature Methods. 2012;9:357–9. doi: 10.1038/nmeth.1923.
Almeida A, Nayfach S, Boland M, Strozzi F, Beracochea M, Shi ZJ, et al. A unified catalog of 204,938 reference genomes from the human gut microbiome. Nature Biotechnology. 2021;39:105–14. doi: 10.1038/s41587-020-0603-3.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25:2078–9. doi: 10.1093/bioinformatics/btp352.
Lawrence M, Huber W, Pagès H, Aboyoun P, Carlson M, Gentleman R, et al. Software for computing and annotating genomic ranges. PLOS Computational Biology. 2013;9:e1003118. doi: 10.1371/journal.pcbi.1003118.
Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Research. 2000;28:27–30. doi: 10.1093/nar/28.1.27.
Rice P, Longden I, Bleasby A. EMBOSS: the European Molecular Biology Open Software Suite. Trends in Genetics. 2000;16:276–7. doi: 10.1016/s0168-9525(00)02024-2.
Durbin R, Eddy, S. R., Krogh, A., & Mitchison, G.. Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids. Cambridge: Cambridge University Press; 1998.
Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein modeling, prediction and analysis. Nature Protocols. 2015;10:845–58. doi: 10.1038/nprot.2015.053.
BioRender. 2022. https://biorender.com/.
Papapanou PN, Sanz M, Buduneli N, Dietrich T, Feres M, Fine DH, et al. Periodontitis: Consensus report of workgroup 2 of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. Journal of clinical periodontology. 2018;45 Suppl 20:S162-s70. doi: 10.1111/jcpe.12946.
Bondy-Denomy J, Qian J, Westra ER, Buckling A, Guttman DS, Davidson AR, et al. Prophages mediate defense against phage infection through diverse mechanisms. The ISME Journal. 2016;10:2854–66. doi: 10.1038/ismej.2016.79.
Fortier L-C, Sekulovic O. Importance of prophages to evolution and virulence of bacterial pathogens. Virulence. 2013;4:354–65. doi: 10.4161/viru.24498.
Benler S, Yutin N, Antipov D, Rayko M, Shmakov S, Gussow AB, et al. Thousands of previously unknown phages discovered in whole-community human gut metagenomes. Microbiome. 2021;9:78. doi: 10.1186/s40168-021-01017-w.
Varble A, Campisi E, Euler CW, Maguin P, Kozlova A, Fyodorova J, et al. Prophage integration into CRISPR loci enables evasion of antiviral immunity in Streptococcus pyogenes. Nature Microbiology. 2021;6:1516–25. doi: 10.1038/s41564-021-00996-8.
Bhatty M, Laverde Gomez JA, Christie PJ. The expanding bacterial type IV secretion lexicon. Research in Microbiology. 2013;164:620–39. doi: 10.1016/j.resmic.2013.03.012.
Pancholi V, Boël G, Jin H. Streptococcus pyogenes Ser/Thr Kinase-regulated Cell Wall Hydrolase Is a Cell Division Plane-recognizing and Chain-forming Virulence Factor. Journal of Biological Chemistry. 2010;285:30861–74. doi: 10.1074/jbc.M110.153825.
Sundarrajan S, Raghupatil J, Vipra A, Narasimhaswamy N, Saravanan S, Appaiah C, et al. Bacteriophage-derived CHAP domain protein, P128, kills Staphylococcus cells by cleaving interpeptide cross-bridge of peptidoglycan. 2014;160:2157–69. doi: 10.1099/mic.0.079111-0.
Mesa F, Mesa-López MJ, Egea-Valenzuela J, Benavides-Reyes C, Nibali L, Ide M, et al. A New Comorbidity in Periodontitis: Fusobacterium nucleatum and Colorectal Cancer. Medicina. 2022;58. doi: 10.3390/medicina58040546.
Pushker R, Mira A, Rodríguez-Valera F. Comparative genomics of gene-family size in closely related bacteria. Genome Biology. 2004;5:R27. doi: 10.1186/gb-2004-5-4-r27.
Mira A, Pushker R, Rodríguez-Valera F. The Neolithic revolution of bacterial genomes. Trends in microbiology. 2006;14:200–6. doi: 10.1016/j.tim.2006.03.001.
Hajishengallis G, Lamont RJ. Polymicrobial communities in periodontal disease: Their quasi-organismal nature and dialogue with the host. Periodontology 2000. 2021;86:210 – 30. doi: 10.1111/prd.12371.
Pardiñas-López S, Warren RN, Bromage TG, Matei IC, Khouly I. Treatment of an Unusual Non-Tooth Related Enamel Pearl (EP) and 3 Teeth-Related EPs with Localized Periodontal Disease Without Teeth Extractions: A Case Report. Compendium of Continuing Education in Dentistry. 2015;36:592–9.
Wilson W, Taubert KA, Gewitz M, Lockhart PB, Baddour LM, Levison M, et al. Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736–54. doi: 10.1161/circulationaha.106.183095.
Stanko P, Izakovicova Holla L. Bidirectional association between diabetes mellitus and inflammatory periodontal disease. A review. Biomedical papers of the Medical Faculty of the University Palacky, Olomouc, Czechoslovakia. 2014;158:35–8. doi: 10.5507/bp.2014.005.
Bahekar AA, Singh S, Saha S, Molnar J, Arora R. The prevalence and incidence of coronary heart disease is significantly increased in periodontitis: a meta-analysis. American heart journal. 2007;154:830–7. doi: 10.1016/j.ahj.2007.06.037.
Kamer AR, Craig RG, Dasanayake AP, Brys M, Glodzik-Sobanska L, de Leon MJ. Inflammation and Alzheimer's disease: possible role of periodontal diseases. Alzheimer's & dementia: the journal of the Alzheimer's Association. 2008;4:242–50. doi: 10.1016/j.jalz.2007.08.004.
Dominy SS, Lynch C, Ermini F, Benedyk M, Marczyk A, Konradi A, et al. Porphyromonas gingivalis in Alzheimer disease brains: Evidence for disease causation and treatment with small-molecule inhibitors. Science Advances. 2019;5:eaau3333. doi: doi:10.1126/sciadv.aau3333.
Wen S, Beltrán V, Chaparro A, Espinoza F, Riedemann JP. Association between chronic periodontitis and rheumatoid arthritis. A systematic review. Revista medica de Chile. 2019;147:762–75. doi: 10.4067/s0034-98872019000600762.
Drewes JL, White JR, Dejea CM, Fathi P, Iyadorai T, Vadivelu J, et al. High-resolution bacterial 16S rRNA gene profile meta-analysis and biofilm status reveal common colorectal cancer consortia. NPJ Biofilms Microbiomes. 2017;3:34. doi: 10.1038/s41522-017-0040-3.
Kolenbrander PE, Palmer Jr RJ, Rickard AH, Jakubovics NS, Chalmers NI, Diaz PI. Bacterial interactions and successions during plaque development. Periodontology 2000. 2006;42:47–79. doi: 10.1111/j.1600-0757.2006.00187.x.
Kolenbrander PE, Palmer RJ, Periasamy S, Jakubovics NS. Oral multispecies biofilm development and the key role of cell–cell distance. Nature Reviews Microbiology. 2010;8:471–80. doi: 10.1038/nrmicro2381.
Simon-Soro A, Ren Z, Krom BP, Hoogenkamp MA, Cabello-Yeves PJ, Daniel SG, et al. Polymicrobial Aggregates in Human Saliva Build the Oral Biofilm. mBio. 2022;13:e0013122. doi: 10.1128/mbio.00131-22.
Murphy EC, Frick IM. Gram-positive anaerobic cocci commensals and opportunistic pathogens. FEMS microbiology reviews. 2013;37:520–53. doi: 10.1111/1574-6976.12005.
Prorok-Hamon M, Friswell MK, Alswied A, Roberts CL, Song F, Flanagan PK, et al. Colonic mucosa-associated diffusely adherent afaC + Escherichia coli expressing lpfA and pks are increased in inflammatory bowel disease and colon cancer. Gut. 2014;63:761 – 70. doi: 10.1136/gutjnl-2013-304739%J Gut.
Yang Y, Nguyen M, Khetrapal V, Sonnert ND, Martin AL, Chen H, et al. Within-host evolution of a gut pathobiont facilitates liver translocation. Nature. 2022;607:563–70. doi: 10.1038/s41586-022-04949-x.
Humann J, Lenz LL. Bacterial peptidoglycan degrading enzymes and their impact on host muropeptide detection. Journal of Innate Immunity. 2009;1:88–97. doi: 10.1159/000181181.
Grenier D, Bouclin R. Contribution of proteases and plasmin-acquired activity in migration of Peptostreptococcus micros through a reconstituted basement membrane. Oral Microbiology and Immunology. 2006;21:319–25. doi: 10.1111/j.1399-302X.2006.00298.x.
Van Dalen PJ, Van Deutekom-Mulder EC, De Graaff J, Van Steenbergen TJ. Pathogenicity of Peptostreptococcus micros morphotypes and Prevotella species in pure and mixed culture. Journal of Medical Microbiology. 1998;47:135–40. doi: 10.1099/00222615-47-2-135.
Cogoni V, Morgan-Smith A, Fenno JC, Jenkinson HF, Dymock D. Treponema denticola chymotrypsin-like proteinase (CTLP) integrates spirochaetes within oral microbial communities. Microbiology. 2012;158:759–70. doi: 10.1099/mic.0.055939-0.
Dai W, Li C, Li T, Hu J, Zhang H. Super-taxon in human microbiome are identified to be associated with colorectal cancer. BMC Bioinformatics. 2022;23:243. doi: 10.1186/s12859-022-04786-9.
Löwenmark T, Löfgren-Burström A, Zingmark C, Eklöf V, Dahlberg M, Wai S, et al. Parvimonas micra as a putative non-invasive faecal biomarker for colorectal cancer. Scientific Reports. 2020;10:15250. doi: 10.1038/s41598-020-72132-1.
Yu J, Feng Q, Wong SH, Zhang D, Liang Qy, Qin Y, et al. Metagenomic analysis of faecal microbiome as a tool towards targeted non-invasive biomarkers for colorectal cancer. Gut. 2017;66:70–8. doi: 10.1136/gutjnl-2015-309800.

No competing interests reported.

Download PDF

Journal Publication

published 08 Aug, 2023

Read the published version in Molecular Oncology →

Version 1

posted

You are reading this latest preprint version

Evidence for translocation of oral Parvimonas micra from the subgingival sulcus of the human oral cavity to the colorectal adenocarcinoma

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Methods

Sample collection

MALDI-TOF MS identification

Bacterial nucleic acids extraction

DNA extraction from stool and saliva samples

DNA and RNA extraction from tissue samples and gingival crevicular fluids

DNA extraction from FFPE samples

Genomic DNA purification from P. micra

16S rRNA sequencing

16S rRNA sequencing for saliva, stool, tissues and gingival crevicular fluid samples

16S rRNA sequencing for FFPE samples

RNAseq of colon tissues

Bioinformatic analysis

16S rRNA sequencing data

Whole genome analysis of P. micra

Comparative genome analysis

Metatranscriptome analysis of gut tissues

Figures design

Results

Clinical background of patient 89

Microbiome of patient 89

Metatranscriptomics of tumor tissues

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1