Genomic description and prevalence of two new Candidatus Saccharibacteria species from the human gut in different samples and countries

doi:10.21203/rs.3.rs-1920999/v1

PDF

Article

Genomic description and prevalence of two new Candidatus Saccharibacteria species from the human gut in different samples and countries

https://doi.org/10.21203/rs.3.rs-1920999/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The candidate phyla radiation (CPR) has been described as an obligatory group of ultrasmall bacteria associated with host bacteria. They phylogenetically represent a subdivision of bacteria distinct from other living organisms.

Using polyphasic approaches, we screened human faecal samples for the detection of Saccharibacteria.

The new sequences obtained by sequencing were compared to the complete CPR genomes available to date. Then, we attempted a co-culture of CPR-bacteria and non-CPR bacteria from human faecal samples. We finally aimed to evaluate the prevalence and distribution of these Saccharibacteria sequences in human sources in 16S amplicon datasets.

We were able to reconstitute two high-quality Saccharibacteria genomes named Minimicrobia massiliensis and Minimicrobia timonensis. We have established, for the first time in human digestive samples, the coculture of Candidatus Saccharibacteria with two different bacterial hosts. Finally, we showed that 12.8% (610/4,756) of samples sequenced in our laboratory were positive for operational taxonomic units (OTUs) assigned to M.massiliensis. and significantly enriched in human respiratory and oral microbiota.

Here, we reported the first genomes and coculture of Saccharibacteria from human gut specimens. This study opens a new field, particularly in the study of the involvement of CPR in the human intestinal microbiota.

Minimicrobia massiliensis

Minimicrobia timonensis

Candidate Phyla Radiation

human

faecal microbiota

Until now, members of the candidate phyla radiation (CPR) have been unidentifiable by conventional techniques, but their detection has been made possible by the advent of molecular techniques and the massive use of metagenomics ¹. Indeed, it was estimated that CPR members represented approximately 15% of all bacteria ¹. The CPR members is a clade distinct from other organisms and represents a subdivision of bacteria, which has been suggested to include 35 phyla ^1,2. Furthermore, CPR members were not simply DNA inside host cells but actually represented a cellular life form ³. Indeed, the CPR genomes encoded genetic systems involved in cell division ³. It was shown that CPR members have a symbiotic lifestyle with bacteria ^4–6. This is exemplified by their small genome size (approximately 1 Mb) ^3,7 and a reduction in their essential metabolic activity compared to the usual metabolic pathways of non-CPR bacteria ³. Indeed, with few exceptions, the lack of metabolic pathways involved in amino and fatty acid biosynthesis was previously described ^3,8–10. In addition, the lack of electron transport pathway complexes and an absence of ATP synthase compared to those of non-CPR bacteria were also reported ^3,11.

In parallel, due to their particular morphology, the use of electron microscopy ^4,12,13 and fluorescence light microscopy (FISH) ^9,13 have allowed clear visualization of the Candidatus Saccharibacteria phylum which has been described as a nanoscale organisms ⁹. Despite the difficulty of culturing these microorganisms in axenic culture, several studies have shown the possibility of culturing Candidatus Saccharibacteria with their host by coculture ^9,13 or using reverse genomics ¹⁴ techniques, mainly from human oral cavity specimens. Indeed, their presence in various microbiomes, including mainly environmental microbiomes ^4,15,16 but also animal ¹⁷ and human microbiota ^9,18, has been described by metagenomic studies. Few studies have shown the presence of CPR members in the digestive tract ^19–22, but none of these studies have identified novel genomic sequences or attempt to cultivate Candidatus Saccharibacteria from the human intestinal microbiota.

Herein, using a polyphasic approach, we detected two novel species of Candidatus Saccharibacteria from the human intestinal microbiota. We provide their genomic description and estimate their prevalence and distribution in various microbiomes. Finally, we describe the first attempts to co-culture Candidatus Saccharibacteria with bacteria previously described as hosts of these ultrasmall bacteria in the human oral microbiome ^9,12,23,24.

1. Screening of Candidatus Saccharibacteria in human fecal samples

1.1 Standard polymerase chain reaction (PCR)

Standard PCR targeting Candidatus Saccharibacteria enabled the generation of an approximate 600 bp amplicon from the 2 studied fecal samples. We assembled and cleaned sequences using ChromasPro, and we obtained a first amplicon of 441 bp and a second amplicon of 468 bp. These sequences both had 100% homology with the GenBank sequence LR761313.1 belonging to the Candidatus Saccharibacteria bacterium partial 16S rRNA gene. Indeed, this GenBank sequence LR761313.1 corresponded to the first 16S ribosomal RNA gene sequence of Candidatus M.timonensis, originating from the faecal sample, which we deposited in the NCBI database.

1.2 Electron microscopy

The examination of symbionts of bacteria by scanning electron microscopy from the faaecal specimen enabled the capture of images compatible with the presence of Candidatus Saccharibacteria. Cells with a diameter of less than 300 nm associated with the extremity or on the side of the bacteria were observed in both faecal specimens (Fig. 2A et 2B). As the two specimens were positive for Candidatus Saccharibacteria (standard PCR, Sanger sequencing, and microscopy), whole genome sequencing was then carried out on these two fecal samples.

2. Genome sequencing and description of Candidatus M.massiliensis and Candidatus M. timonensis.

Furthermore, based on the cut-off used to describe two different bacterial species, we concluded that they were two different species. Indeed, the ANI values (44) and the dDDH values (45, 46) calculated between Candidatus M.massiliensis and Candidatus M.timonensis were far below the threshold values recommended to delineate bacterial species (i.e., 95–96% and 70%, respectively) (see "Genomic comparison" section below).

2.1 Genomic properties of Candidatus M.massiliensis

A total of 3,516,594 Illumina paired-end reads were assembled. A BLASTn of 45,409 contigs versus nr was performed, and we kept only 191 contigs that belonged to “Candidatus Saccharibacteria.” The deglycosylation pretreatment combined with extraction protocol 5 enabled us to cover the largest percentage of the Candidatus Saccharibacteria genome (70%) (Table S2). In this study, deglycosylation treatment had an important impact on the results obtained. Indeed, this deglycosylation was also found in extraction protocol 5 (excluding pre-treatment), which could explain why this extraction protocol 5 was more adapted to our research than the other protocols used. Here, we highlight that the removal of the biofilm and the release of the DNA by deglycosylation improved the DNA extraction and sequencing results.

Then, we obtained a genome of 1 scaffold and 91 contigs that were 971.417 bp long with 43.2% GC content. A graphical circular map of Candidatus M.massiliensis was acquired and was shown in Fig. 2C. The gene prediction of Candidatus M.massiliensis allowed us to count a total of 1,073 ORFs, and functional annotation showed that 991 (92.3%) were coding DNA sequences, 62 (5.8%) were tRNA-(amino acid) sequences and 3 (0.3%) were rRNA sequences, and 565 (52.6%) were assigned to clusters of orthologous groups (COGs). In addition, we found that the numbers of enzymes involved in protein metabolism (27.8% =68/245) and DNA metabolism (16.3% =40/245) were the highest among the represented subsystems (Figure S1).

The mosaic representation of the Candidatus M.massiliensis genome showed that 63.8% of the genes overlapped with bacteria, 26.4% with other CPR groups, 0.3% with archaea and 9.32% were ORFans (Fig. 2E). Moreover, we also detect one gene with viral origin in this genome. Thereafter, after the application of the same antibiotic resistance adapted strategy ⁴⁰, we found that Candidatus M.massiliensis was resistant to tetracycline, Glycopeptide and MLS (for Macrolide – Lincosamide - Streptogramin).

2.2 Genomic properties of Candidatus M.timonensis

A total of 17,610,272 Illumina paired-end reads and 67,937 Nanopore reads were assembled. A BLASTn of 50,159 contigs versus nr was performed, and we kept only 400 contigs that belong to “Candidatus Saccharibacteria.” The combination of deglycosylation and 5% Tween 80 pretreatment combined with extraction protocol 5 enabled us to cover the largest percentage of the genome (79%) (Table S2). As mentioned above, we confirm that deglycosylation has a beneficial impact on the genome coverage. Still with the aim of reducing the presence of biofilm and freeing the DNA, we had the idea of using Tween 80 known to be an emulsifier/dispersant. In this work, we also demonstrated that the Tween 80 treatment complemented with deglycosylation allowed to obtain a better coverage of the genome compared to the other treatments.

A genome of 3 scaffolds and 197 contigs was obtained with genomic size was 755,196 bp long with 43.7% GC content. In addition, we showed that Candidatus M. timonensis had 839 ORFs, including 777 (92.6%) coding DNA sequences. In parallel, 40 genes coded for (4.8%) tRNA-(amino acid), 3 (0.4%) for rRNA, and 478 (56.9%) were assigned to COGs. Moreover, we highlighted that the number of enzymes involved in protein metabolism (27.7% =54/195) and DNA metabolism (15.3% =30/195), as for Candidatus M.massiliensis, were the most common subsystems found in the genome of Candidatus M. timonensis (Figure S1). The graphical circular map of Candidatus M. timonensis is shown in Fig. 2D. The mosaic representation of the rhizome showed that 63.8% of the genes overlapped with bacteria, 26.1% overlapped with CPR, 0.29% with archaea, and 9.8% with ORFans (Fig. 2F). In parallel, one gene with viral origin were detected. Finally, Candidatus M. timonensis was resistant to beta-lactam, tetracycline and glycopeptide.

3. Genomic comparison

3.1 Taxono-genomics classification

The 16S rRNA-based phylogenetic tree highlighting the position of Candidatus M.massiliensis and Candidatus M. timonensis relative to other closely related taxa shows that these genomes belonged to the cluster of Candidatus Saccharibacteria (Fig. 3A). The maximum ANI value was 94 between Candidatus M.massiliensis and Candidatus M. timonensis (Fig. 3C). In addition, estimation by digital DNA-DNA hybridization (dDDH) ^46,47 was performed, and the maximum value of the genomes of interest was 56.8 (confidence interval = [54.0–59.5]) between Candidatus M.massiliensis and Candidatus M. timonensis (Table S3). Then, a phylogenetic tree based on whole genome sequences was constructed and confirmed that Candidatus M.massiliensis and Candidatus M. timonensis belong to the phylum Candidatus Saccharibacteria (Fig. 3B). A heatmap representing the single nucleotide polymorphisms (SNPs) between the CPR genomes studied in this work is shown in Fig. 3D. A SNP is a germline substitution of a nucleotide at a specific position in the genome. Regarding CPR genomes, we found that the number of SNPs shared between the CPR genomes studied here varied from 162 between Candidatus Gracilibacteria bacterium 28_42_T64 chromosome (CP042461.1) and Candidatus Gracilibacteria bacterium HOT-871 (CP017714) to 2,213 between Candidatus Saccharibacteria bacterium oral taxon 488 strain AC001 (CP040003) and Candidatus Woesebacteria bacterium GW2011_GWF1_31_35 (CP011214). The genomes of Candidatus M.massiliensis and Candidatus M.timonensis shared 184 SNPs between them, highlighting the close link between these strains.

3.2 Comparison of genomic properties and metabolism of CPR

The genomic properties of the CPR genomes studied in this study are listed in Table 1. Briefly, the size of the available CPR genomes varied between 1,450,269 bp (for Candidatus Saccharibacteria division) and 602,646 bp (for Candidatus Kazan division), and the coding radio varied between 20.9% (for Candidatus Gracilibacteria) and 94.8% (for Candidatus Beckwithbacteria). For Candidatus M.massiliensis and Candidatus M.timonensis, their genomic sizes were 971.417 bp and 755.196 bp and the coding radio were 85.7% and 78.7%, respectively.

Furthermore, the number of coding proteins in these genomes ranged from 2,293 (for Candidatus SR1 division) to 642 (for Candidatus Kazan division), and the average size of the proteins ranged from 311.2 amino acids (for Candidatus Kazan division) to 74.1 (for Cadidatus Gracilibacteria phylum). Regarding Candidatus M.massiliensis and Candidatus M.timonensis, the number of coding proteins were 1009 and 796, and the average size of the proteins were 280 amino acids and 255 amino acids, respectively.

Moreover, the proportion of ORFan genes ranged from 35.37% (= 504/1515) for the Candidatus Gracilibacteria division to 0% for five complete genomes from the environmental microbiome (Table 1). Regarding Candidatus M.massiliensis and Candidatus M. timonensis, the ORFan genes represented 5.0% (= 40/796) and 2.7% (= 27/1,008), respectively. In parallel, all the genomes described here possessed rRNA and tRNA sequences, but only two human CPR genomes showed the presence of CRISPR systems. In terms of CG%, there was a variation of 26.7% between the different CPR genomes studied here from 51.8% for the Candidatus Kazan division to 25.1% for the phylum Candidatus Gracilibacteria (Table 1). For Candidatus M.massiliensis and Candidatus M.timonensis, the GC% were 43.2% and 43.7%, respectively.

In addition, the distribution of functional classes of predicted genes based on orthologous protein groups was mainly in the following groups: 1) translation, 2) replication, recombination, and repair, 3) general function prediction only, and 4) cell wall/membrane biogenesis. Indeed, in Candidatus M.massiliensis and Candidatus M. timonensis, these groups included more than half of the COGs: 55.75% (315/565) and 55.02% (263/478), respectively (Figure S2). The most represented Kyoto Encyclopedia of Genes and Genomes (KEGG) categories of the CPR genomes described here were “translation,” followed by "carbohydrate metabolism” and "replication and repair” (Fig. 3E). In Candidatus M.massiliensis and Candidatus M. timonensis, these KEGG categories included 48.0% (160/333) and 50.6% (154/304) of the genes, respectively. In parallel, the categories underrepresented in the CPR genome were "signaling molecules and interaction" and "lipid metabolism" (Fig. 3E).

3.3 Pangenomic comparison

Pangenome analysis of 10 genomes of Candidatus Saccharibacteria, including Candidatus M.massiliensis and Candidatus M. timonensis, showed that 136 common genes were shared (percentage identity = 50%) (Fig. 3F). However, pangenome analysis against 18 complete genomes of various phyla of CPR described only 3 common genes: rpoB, tuf, and rpsJ coding for DNA-directed RNA polymerase subunit beta, elongation factor Tu, and 30S ribosomal protein S10, respectively (percentage blastp identity = 50%) (Fig. 3G). Moreover, other blastp identity percentages (65%, 80%, and 95%) are also shown in Figures S3.

4. Culture of Candidatus Saccharibacteria from a human faecal sample

The culture of Candidatus Saccharibacteria was cocultured with potential host bacteria (Schaalia odontolytica and Arachnia propionica) as suggested by Murugkar et al. ¹².

Follow-up by standard PCR targeting Candidatus Saccharibacteria showed positive results after filtration at 0.22 µm and ultracentrifugation of the two human faecal samples studied here. In addition, all initial cocultures were negative in standard PCR targeting the 16S rRNA gene (primers TM7-1177R and TM7-580F).

4.1. Attempts to cultivate Candidatus M.massiliensis from faecal sample 1

For the human digestive sample from which the Candidatus M.massiliensis genome was reconstructed, all the subcultures performed from the “Candidatus Saccharibacteria"- Schaalia odontolytica coculture were negative by standard PCR (Table S4). Nevertheless, subcultures P1 and P5 under anaerobic conditions and subculture P4 under aerobic conditions of the potential CPR coculture with their host Arachnia propionica were positive by molecular biology techniques (Table S4). Sanger sequencing was able to prove the infection of host cells with Candidatus Saccharibacteria (Table 2). Surprisingly, each coculture a different top hit suggesting the presence of more than one species of Candidatus Saccharibacteria in the faecal sample 1 (Table 2). In parallel, the electron micrographs showed the presence of these structures in postinfection in the subcultures that were positive by PCR (Figure S4).

Table 2

**– Summary table of blastn results obtained from positive amplicons of** Candidatus **Saccharibacteria-host coculture from human digestive samples (best hits).** P1: passage 1, P3: passage 3, P4: passage 4, P5: passage 5., S1: Digestive sample 1 (*M. massiliensis*), S2: Digestive sample 2 (*M. timonensis*)
Positive coculture in standard PCR and electron microscopy	Blastn results	Accession number	%Coverage	%Identity
P1-S1- A.propionica (Anaerobic condition)	Uncultured bacterium clone 071072_210 16S ribosomal RNA gene, partial sequence	JQ477137.1	98%	99.54%
P5-S1- A.propionica (Anaerobic condition)	Uncultured bacterium clone TM7 BBM10 16S ribosomal RNA gene, partial sequence	KP326382.1	100%	99.78%
P4-S1- A.propionica (Aerobic condition)	Uncultured candidate division TM7 bacterium clone 09_3_D06 16S ribosomal RNA gene	GU227156.1	100%	98.85%
P3-S2- A.propionica (Aerobic condition)	Candidatus Saccharibacteria bacterium oral taxon 488 strain CM002 chromosome, complete genome	CP039998.1	100%	95.96%
P5-S2- A.propionica (Aerobic condition)	Uncultured bacterium clone TM7 BBM10 16S ribosomal RNA gene, partial sequence	KP326382.1	100%	100%
P4-S2- S.odontolytica (Anaerobic condition)	Candidatus Saccharibacteria bacterium partial 16S rRNA gene (= Minimicrobia timonensis)	LR761313.1	100%	99.78%

4.2 Attempts to cultivate Candidatus M.timonensis from faecal sample 2

For the second human digestive sample from which the Candidatus M. timonensis genome was constructed, all the subcultures of the “Candidatus Saccharibacteria" and Arachnia propionica cocultures in an anaerobic atmosphere were negative by standard PCR (Table S4). However, subcultures P3 and P5 of the " Candidatus Saccharibacteria"- Arachnia propionica coculture under aerobic conditions and subculture P4 of the coculture of potential “Candidatus Saccharibacteria” with Schaalia odontolytica were positive by molecular techniques (Table S4). Blasting of DNA bands obtained on agarose gels revealed the infection of host strains by Candidatus Saccharibacteria (Table 2). Surprisingly, the blastn search of the sequence obtained after subculturing P4 of the coculture with Schaalia odontolytica gave a result with 100% coverage and 99.78% identity with Candidatus M. timonensis that we targeted (Table 2). Once again, each coculture a different top hit suggesting the presence of more than one species of Candidatus Saccharibacteria (Table 2). In parallel, microscopic observation highlighted the presence of symbionts of bacteria in postinfection that were absent preinfection (Figure S4).

5. Metagenomic analysis

The MetaGX database is based on 16S rRNA, and these two genomes described here are very close to each other (99.7% identity of the 16S rRNA gene). Therefore, in order to have a preliminary view on the prevalence and distribution, we used only the reference sequence of Candidatus M.massiliensis. The number of reads corresponding to OTUs associated with Candidatus M.massiliensis ranged from 20 to 41,243. The maximum relative abundance was 9.31% in human respiratory samples from France. In total, we found 12.8% (= 610/4,756) positive samples in the MetaGX database ⁴⁴. More precisely, the frequency of Candidatus M.massiliensis OTUs was 85.71% (= 24/28) in the human oral samples, 17.7% (= 396/2,233) in human gut samples, 13.87% (= 117/843) in human respiratory samples, 12% (= 42/350) in human breast milk samples, 11.9% (= 14/117) in human skin samples, 1.97% (= 15/760) in human urine samples and 1.5% (= 2/128) in environmental samples. However, this OTUs were absent from human blood, animal digestive tract, human genital tract, human brain abscess, human bone, insects, planaria, and yogurt microbiota (Figure S5). The frequency in positive samples (= 610) was 64.9% (= 396/610) in the human digestive tract, 19.1% (= 117/610) in the human respiratory system, 6.88% (= 42/610) in human breast milk, 3.93% (= 24/610) in human oral samples, 2.45% (= 15/610) in human urine, 2.29% (= 14/610) in human skin and 0.32% (= 2/610) in environment specimens (Fig. 4A). The highest relative abundances were found in the oral (1.76%) and respiratory (0.53%) microbiota (Fig. 4B). In parallel, there were significant differences between the relative abundance in the oral samples and breast milk (p < 0.0001), urine (p < 0.05), skin (p < 0.01), and human digestive samples (p < 0.0001) (Fig. 4B). In addition, we found a significant difference between the relative abundances in respiratory samples and human digestive (p < 0.0001) and breast milk samples (p = 0.0001) (Fig. 4B). Candidatus M.massiliensis OTUs were more frequently detected in samples from France (51.63% =315/610) than in those from Mali (38.85% =237/610) and Senegal (9.50% =58/610) but were absent in microbiomes from Vietnam, Niger and Côte d’Ivoire (Figure S5). The relative abundance found in samples from France was significantly higher than in those from Mali (p < 0.0001) (Figure S5). In addition, the frequencies of positive samples from Europe and Africa were 51.64% (= 315/610) and 48.36% (= 295/610), respectively (Fig. 4C), but these OTUs were absent from the samples from Asia (Figure S5). We found a significant difference between the relative abundance of samples from Europe and those from Africa, with abundances in European samples being higher (p < 0.0001) (Fig. 4D). Conversely, a significant difference was found between the relative abundances in human breast milk samples from Europe and Africa, with higher values in African human breast milk samples (p < 0.01) (Figure S5). In parallel, no significant differences were found between the relative abundances of other microbiota according to their continental origin (Figure S5).

Furthermore, following a thorough metagenomic analysis, we found a significant proportion of Candidatus M.massiliensis (7,024 reads with a relative abundance of 0.9%) in a fecal metagenome from a patient who had spontaneously been cured of HIV infection ²⁵. We were able to cover only 88% of the genome by mapping the Candidatus M.massiliensis genome and we detected a second novel genome of Candidatus Saccharibacteria species that we named Candidatus M.timonensis.

In this study, we demonstrated the presence of two novel Candidatus Saccharibacteria species from the human intestinal microbiota using a polyphasic approach. First, we observed images compatible with the presence of ultrasmall bacteria using electron microscopy. Secondly, molecular biology methods confirmed the presence of Candidatus Saccharibacteria DNA in human faecal specimens, as amplified sequences displayed 100% similarity with Candidatus Saccharibacteria. In addition, using next-generation sequencing on both specimens, we obtained two novel genomic sequences related to sequences of Candidatus Saccharibacteria from human stool samples. For the first time from human digestive samples, we succeeded in establishing various cocultures of Candidatus Saccharibacteria with two bacterial hosts. In addition, a meta-analysis of the distribution of OTUs assigned to Candidatus M.massiliensis highlighted the prevalence of Candidatus Saccharibacteria in various human microbiota. Finally, significant differences according to the origin and the type of sample were highlighted.

To date, 33 complete genomes of Candidatus Saccharibacteria are available online in the NCBI database (https://www.ncbi.nlm.nih.gov/), including 14 of human origin. However, none of these genomes comes from the human gut. Therefore, we compared the two genomes studied here with a random subset of 18 complete genomes available on NCBI, including 8 genomes of Candidatus Saccharibacteria and 10 genomes of other CPR phyla. Due to the lack of data on the classification of CPRs and the thresholds used delineating one CPR species from another, we privileged the thresholds used for the non-CPR bacteria classification. As a result, we demonstrated that the OrthoANI values ⁴⁵ and the dDDH values ^46,47 calculated between the genomes of the species identified in the present study and those included were far below the threshold values recommended to delineate bacterial species (i.e., 95–96% and 70%, respectively) ⁵². All these data support the creation of two new Candidatus Saccharibacteria species from human faecal samples: Candidatus M.massiliensis and Candidatus M. timonensis.

Moreover, we showed herein the mosaic representation of the genomes of Candidatus M.massiliensis and Candidatus M. timonensis, which confirms the parasitic relationship of Candidatus Saccharibacteria with non-CPR bacteria, which is in line with the findings of previous studies ⁴. Indeed, Luef et al. showed that CPR members had metabolic processes specific to parasites ⁴ including causing host cell lysis ²³. Moreover, the ORFan genes of Candidatus M.massiliensis and Candidatus M. timonensis were numerous (i.e., 27 and 40, respectively), and we do not currently know if these genes are involved in the production of alternative metabolic pathways ^7,53,54, such as antibiotic resistance. In parallel, various previous studies have shown that CPRs were limited in or even totally devoid of defense systems against bacteriophages, particularly CRISPR systems ⁵⁵. Nevertheless, we have highlighted the presence of CRISPR systems in only 2/20 complete genomes studied here. Indeed, these systems could be acquired in CPR genomes through a mammalian host, such as humans ⁵⁵.

In addition, our metagenome selection highlighted that Candidatus M.massiliensis species appeared to be very commonly found in human microbiomes, as recently described ^14,56–59. Interestingly, the proportion of Candidatus M.massiliensis was significantly higher in human respiratory and oral samples. Moreover, to our knowledge, this study was the first work showing the presence of genomic sequences of Candidatus Saccharibacteria, and more generally of CPR members, in human breast milk and human urine. On the one hand, the relative abundances of Candidatus M.massiliensis OTUs were significantly higher in Europe than in Africa. On the other hand, we found a significant difference between the relative abundances of these OTUs found in human breast milk samples from Africa compared to human breast milk samples from Europe (with breast milk samples of African origin showing higher values (p = 0.01)). Therefore, the study of the Candidatus Saccharibacteria impact on the human breast milk microbiota, their transfer into the microbiota of the breastfed child ^60,61, and the possible human pathologies associated with this transfer ⁶² are currently unexplored items that would be of interest for future study.

Due to their particular lifestyle, the culture of Candidatus Saccharibacteria bacteria is currently difficult to achieve because they requires fastidious cultivation, such as coculture ^9,13 or reverse genomics methods ¹⁴. In this work, we cocultivated Candidatus M. timonensis in a human feacal sample with Schaalia odontolytica, for which the genomic sequence had been previously obtained by metagenomic (Blastn results = 100% coverage and 99.78% identity with Candidatus M. timonensis). The coculture technique described by Murugkar et al. ¹² initially used for human oral samples could also be used for other microbiota, such as the human gut microbiota, as described here.

However, one of the limitations of this work was that we were not able to obtain complete genomes of Candidatus Saccharibacteria from the faecal microbiota. Indeed, the sequences obtained were fragmented. Furthermore, despite pure co-culture, we could not be sure that Candidatus M.massiliensis and Candidatus M.timonensis were alone on the micrographs. This was supported by the sequencing results which confirmed that different species of Candidatus Saccharibacteria could be found. Indeed, studies have previously established that the same host can be host to different Candidatus Sacchairbacteria ^10,14.

The cultivation of these mysterious symbionts and the discovery of new genomic sequences in various human microbiomes could help us to better understand their physiological and pathological roles, which are currently poorly understood.

Minimicrobia (Mini – Microbe), a small microbe (Latin noun). Minimicrobia massiliensis sp. nov. (mas.si' li.en.sis. L. gen. adj. massiliensis, from Massilia, the Latin name of Marseille, France, where the organism was first identified). Minimicrobia timonensis sp. nov. (ti.mo.nen’sis L. adj. fem refer to Timone, the name of the main hospital in Marseille, France, where the organism was characterized).

1. Patients and specimens

The first subject was a 62-year-old man with an HIV infection acquired 34 years ago for which he regularly received consulting at the infectious diseases department. The patient has an artificial intestinal outlet. Therefore, the first faecal sample studied in this work was taken directly from the device. In addition, fecal microbiota transplantation was planned due to virological failure (viral load = 2.1 Log₁₀ copies of HIV RNA/mL and CD4 T cell count = 594 cells/mm³) despite antiretroviral therapy that comprised emricitabine, etavirine, ritonavir, darunavir, and dolutegravir. A specimen was collected preceding the fecal microbiota transplantation. In parallel, the second subject was a 39-year-old woman who was spontaneously cured of HIV ²⁵. Unlike the previous faecal sample, this one was obtained directly from the stool.

The patients gave informed and signed consent. The study was approved by the IHU Méditerranée Infection under agreement number N ° 2019-002.

2. Electron microscopy detection

Fecal samples were fixed in a 2.5% glutaraldehyde fixative solution. Then, the samples were incubated, and the wells were cytocentrifuged on glass slides (Thermo Electron Corporation - Shandon Cytospin 4 R) at 800 rpm for 7 min. In addition, SEM images were contrasted using 1% phosphotungstic acid aqueous solution (pH 7.0) for 2 minutes. The slide was then air dried. Electron micrographs were acquired manually using a Hitachi TM400 Plus tabletop SEM. The accelerating voltage was 10 kV or 15 kV, and magnifications ranged between 4000× and 10,000×, as previously described ²⁶.

3. Molecular biology detection

3.1 Amplification and revelation of Candidatus Saccharibacteria DNA

Candidatus Saccharibacteria DNA was extracted from fecal specimens using the EZ1 biorobot (Qiagen, Courtaboeuf, France) complemented by the EZ1 DNA tissue kit, following the “DNA bacteria” extraction program. PCR targeting the 16S rRNA gene was performed using the M7580F primer and 1177R primer ²⁷. The electrophoresis of the PCR product was carried out on an 1.5% agarose gel with SYBR™ Safe DNA Gel Stain. Sanger sequencing was performed as previously described ²⁸ using the same primers used for standard PCR in this study. A summary diagram of the methodology is presented in Fig. 1.

3.2 Next Generation Sequencing

a. Pretreatments.

By electron microscopy, we identified the presence of a large biofilm in the fecal samples. We then hypothesised that this biofilm was hindering the extraction of DNA from Candidatus Saccharibacteria. In order to reduce this biofilm, we used proteolytic enzymes to release the DNA as previously described ²⁹. Therefore, we tested various pretreatments on human fecal samples: i) a deglycosylation pretreatment using the Endo Hf kit P0703L (New England Biolabs, Evry, France). Briefly, 20 µL/ml of each reagent was deposited in a diluted sample followed by incubation for 1 hour at room temperature for 1 hour at 37°C. ii) DNase treatment was performed using a TURBO DNA-free™ Kit (ThermoFisher, Life Technologies, Illkirch-Graffenstaden, France). We added 0.1 V/ml in the sample that was incubated twice for 30 minutes at 37°C and DNase activity was deactivated at 65°C for 10 minutes. iii) Pretreatment with Trypsin (20µL/ml) and Proteinase K (20µL/ml). Indeed, the pretreated sample was incubated at 37°C for 1 hour. iv) Pretreatment using 5% Tween 80. The pretreated sample was then incubated at 37°C for 1 hour.

b. DNA extraction.

For faecal sample sequencing, we extracted the bacterial DNA using three different extraction protocols, regularly used in our laboratory and named the “classic protocol,” “protocol 1.” and “protocol 5”. We used these different extraction protocols to compare their efficiency in DNA extraction from Candidatus Saccharibacteria from human faecal samples.

Classic protocol.

The classic protocol was carried out using 200 µL of diluted fecal sample and the EZ1 biorobot (Qiagen, Courtaboeuf, France) complemented by the EZ1 DNA tissue kit, following the “DNA bacteria” extraction program. The elution volume was 100 µL.

Protocol 1.

Protocol 1 extraction was performed using the EZNA Tissue DNA Kit (Omega Bio-Tek Norcross, USA). First, the sample was mixed with 250 µL of phosphate-buffered saline (PBS) and a tip of glass powder. Then, Fastprep was carried out (90 seconds, 6.5 m/s), and the sample was heated to 100°C for 5 minutes. We added 200 µL of TL buffer and 25 µL of OB protease solution to the mix. This step was followed by incubation at 55°C overnight. A Fastprep was performed a second time, followed by centrifugation at 10,000g for 5 minutes. Then, we added 200 µL of BL buffer to the supernatant and incubated it at 70°C for 10. Then, 220 µL of ethanol was added, and the sample was vortexed. The mixture was centrifuged (1 minute; 10,000g) using a nucleospin column. We performed 3 washes: a first wash with 500 µL of HBS buffer and isopropanol and a second wash with 700 µL of DNA wash buffer supplemented with ethanol. This last wash was performed twice. The column was centrifugally dried (2 minutes; 10,000g). Then, we added 25 µL of EB buffer preheated to 70°C. The extracted DNA was incubated for 2 minutes and centrifuged (1 minute at 10,000g).

Protocol 5.

Protocol 5 extraction started with mechanical lysis. For this, a tip of glass powder was added to 500 µL of a diluted stool sample. Fastprep was then carried out (90 s; 6.5 m/s), followed by decantation of the sample. The supernatant was centrifuged to recover the pellet. Then, the pellet was resuspended in 2 µL of 10X Glycoprotein denaturation buffer (New England Biolabs) and 17 µL of water. The pellet was heated at 100°C for 10 minutes. We added 2 µL of 10X G3 reaction buffer, 2 µL of EndoHF (New England Biolabs), 2 µL of cellulase (Sigma), and 10 µL of H2O. We then incubated this at 37°C for a minimum of 1 h and maximum overnight. DNA extraction was performed by adding 25 µL of Proteinase K and 180 µL of G2 buffer. This step was followed by incubation from 45 minutes to 1 hour at 56°C. The sample was centrifuged (1 minute; 10,000g) and 200 µL was recovered for extraction by an EZ1 biorobot (Qiagen, Courtaboeuf, France). The elution volume was 50 µL.

c. Genome sequencing.

DNA extracts were sequenced using MiSeq technology (Illumina, San Diego, CA, USA) with the paired-end strategy and prepared with the Nextera XT Paired-end kit (Illumina), as previously described ³⁰. In parallel, using the SQK-LSK109 kit, the Oxford Nanopore approach was performed on 1D genomic DNA sequencing for the MinIon device. Then, 1.5 µg of genomic DNA without fragmentation and end repair was used for the construction of the library. Adapters were ligated to both ends of genomic DNA. After purification on AMPure XP beads (Beckman Coulter Inc., Fullerton, CA, USA), the library was quantified by a Qubit assay with a high sensitivity kit (Life Technologies, Carlsbad, CA, USA). For sequencing, we detected active pores, and we used the workflow WIMP for live bioinformatic analysis.

4. Bioinformatic analysis

4.1 Genome assembly, scaffolding and finishing

Reads were quality-checked using FastQC and trimmed using Trimmomatic version 0.36.6 ³¹. The total reads were assembled using SPAdes software version 3.13.0 ³². The default options were used. We kept only contigs with a minimum of 400 bp. We performed a BLASTn of contigs versus nr database. The scaffolding was performed using MeDuSa scaffolds ³³.

4.2 Genome annotation and genes prediction

The genomes were annotated as previously described ³⁴. The CDS, tRNA, rRNA, and CRISPR sequences of CPR genomes were predicted with Dfast (https://dfast.ddbj.nig.ac.jp/) ³⁵. In addition, Rast annotation was performed ³⁶. At the same time, an analysis with the Kyoto Encyclopedia of Genes and Genomes (KEGG) Automated Annotation Server (KAAS) ³⁷ was performed to obtain an overview of the metabolic pathways predicted from the coding sequences. Functional annotation was performed by a comparison of protein sequences with public databases (nr and Cluster of orthologous groups database (COG)) ^38,39. Antibiotic resistance profiles were performed using the application of the antibiotic resistance adapted strategy, as previously described ⁴⁰. Open reading frames (ORFs) were predicted using Prokka ⁴¹ for each of the genomes studied. In order to show the mosaicism of these two described genome, we constructed a rhizomal representation of each genome, as a previously described ^10,42,43. This representation shows the origin of each amino acid sequences and determine the possible lateral gene transfer between our CPR genomes and other microbes ⁴³. Rhizome representation was constructed using circos tool Version 0.69-9 (http://www.circos.ca/).

5. Taxonogenomic comparison.

We compared two new genomes of Candidatus Saccharibacteria studied in the present study (CADDWL000000000 and CAJGBL000000000) to the 18 complete genomes of CPR reference strains, including 8 Candidatus Saccharibacteria reference strains (CP040004, CP011211, CP006915, CP040003, CP040008, CP007496, CP025011, CP023991) and 10 complete genomes of non-Sacchairbacteria CPR phyla (accession numbers: CP006913, CP006914, CP011209, CP011210, CP011214, CP011215, CP011268, CP017714, CP042461, CP011216). Two databases were used for 16S rRNA-based phylogenetic tree production: the online database nucleotide collection (nr/nt) on NCBI Blast and a database created in our lab entitled MetaGX ⁴⁴. Sequences were aligned using Muscle v3.8.31 with default parameters, and phylogenetic inferences were obtained using the maximum likelihood method (Hasegawa-Kishino-Yano model/method) with 1000 bootstrap replicates in MEGA software version 7.0. The degree of genomic similarity was estimated using ANI values ⁴⁵. Digital DNA-DNA hybridization (dDDH) ^46,47 was performed using Type (Strain) Genome Server (TYGS) (https://tygs.dsmz.de/) for a whole genome-based taxonomic analysis ⁴⁸. Moreover, pangenomic comparison was realized using Roary ⁴⁹ and visualized using online phandango (https://jameshadfield.github.io/phandango/#/). Finally, SNP results were obtained using a homemade command line and visualized using Morpheus software (https://software.broadinstitute.org/morpheus/). The homemade command lines are available in supplementary materials.

6. MetaGX database and statistics

The metagenomic analysis were performed using the MetaGX database, as previously described ⁴⁴. The table used for the metagenomic analyses is available in Table S1. In this work, we used the same OTU definition used by Diakité et al ⁴⁴, specifically targeting OTUs that were assigned to Candidatus M.massiliensis. The graphic representations and statistical comparisons were made using GraphPad Prism version 8.0.1. The Mann–Whitney test was used when two unmatched unpaired and nonparametric variables were involved. The Kruskal–Wallis test was used when more than two unmatched unpaired and nonparametric variables were involved.

7. Attempts to cultivate Saccharibacteria from human digestive samples

The cultivation of Candidatus Saccharibacteria was performed using the protocol of Murugkar et al. ¹², which we adapted for human digestive samples. Briefly, we diluted the respective stool in 20 ml of MRD medium to filter this type of sample. Unlike the standard protocol, we performed successive filtration of the digestive samples to 1.2 µm, 0.8 µm, 0.45 µm, and 0.22 µm. Then, the filtrate was ultracentrifuged as previously described ¹² and resuspended in 1 ml of MRD medium. Two hundred milliliters of the ultracentrifuged filtrate was cultured in 2 ml of Schaalia odontolytica Q4727 under anaerobic conditions in BHI medium, as well as with Arachnia propionica P6184 under anaerobic and aerobic conditions in TSBY medium as in the standard protocol. We selected these hosts because they have previously shown efficiency in coculture with Candidatus Saccharibacteria ^12,23. These cultures were subcultured 5 times every other day in a new medium containing the newly described strains dated 48 h. Follow-up by standard PCR targeting the 16S rRNA gene (primers TM7-1177R and TM7-580F ^27,50,51), Sanger sequencing, and electron microscopy was performed.

ACKNOWLEDGMENTS

We sincerely thank Vincent Bossi, Ludivine Brechard, Claudia Andrieu, Olivia Ardizzoni, and Madeleine Carrara for their help in collecting sequencing data and for their technical help regarding the genomic analysis. The authors thank Jacques Boukhalilfor his help with the microscopic analysis and Camille Valles for her help with the culture technique. We sincerely thank Takashi Irie, Kyoko Imai, Shigeki Matsubara, Taku Sakazume, Toshihide Agemura, Yusuke Ominami, Akiko Hisada, and the Hitachi High-Tech Corporation team in Japan (Hitachi High-Tech Corporation, Toranomon Hills Business Tower, 1–17–1 Toranomon, Minato-ku, Tokyo 105–6409, Japan) for the collaborative study conducted together with IHU Méditerranée Infection and the installation of TM4000 Plus microscopes at the IHU Méditerranée Infection facility.

AUTHOR CONTRIBUTIONS

Investigation: SN, AC, HA, AI, AL, MM, MB, AD, CD. Formal analysis: SN, AC, HA, AI, AL, MM, and AD. Funding acquisition: DR. Methodology: DR and GD. Supervision: DR and GD. Validation: DR and GD. Visualization: SN, AC, HA, AI, and MM. Writing—original draft: SN, AC, GD, and DR. Writing—Review and Editing: SN, AC, GD, and DR. All authors have read and agreed to the published version of the manuscript.

COMPETING INTERESTS

DR was a consultant in microbiology for the Hitachi High-Tech Corporation until March 2021. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

DATA AVAILABILITY

The genome sequences of Candidatus M.massiliensis and Candidatus M. timonensis are available in Genbank under accession numbers CADDWL010000000 and CAJGBL010000000, respectively. In addition, here the accession numbers of the sequences obtained in this work by Sanger sequencing that are available on NCBI (https://www.ncbi.nlm.nih.gov/):

ETHICS DECLARATIONS

The study is approved by the ethical committee of IHU Méditerranée Infection (Marseille, France) (project n° 2019–002). All experiments were performed in accordance with the recommendations and informed consent was obtained from all participants in this study.

FUNDING

This work was funded by the “IHU Méditerranée Infection” (Marseille, France) and by the Fresh Government under the “Investissements d’avenir” (Investments for the future) program managed by Agence Nationale de la Recherche (ANR, fr: National Agency for Research) (reference: Méditerranée Infection 10-IAHU-03). This work was also supported by Region Provence Alpes Côte d’Azur” and European funding FEDER PRIMI.

Brown, C. T. et al. Unusual biology across a group comprising more than 15% of domain Bacteria. Nature 523, 208–211 (2015).
Hug, L. A. et al. A new view of the tree of life. Nat Microbiol 1, 16048 (2016).
Castelle, C. J. et al. Biosynthetic capacity, metabolic variety and unusual biology in the CPR and DPANN radiations. Nature Reviews Microbiology 16, 629–645 (2018).
Luef, B. et al. Diverse uncultivated ultra-small bacterial cells in groundwater. Nat Commun 6, 6372 (2015).
Chen, I. & Dubnau, D. DNA uptake during bacterial transformation. Nat Rev Microbiol 2, 241–249 (2004).
Proft, T. & Baker, E. N. Pili in Gram-negative and Gram-positive bacteria - structure, assembly and their role in disease. Cell Mol Life Sci 66, 613–635 (2009).
Kantor, R. S. et al. Small genomes and sparse metabolisms of sediment-associated bacteria from four candidate phyla. mBio 4, e00708-00713 (2013).
Meng, P. et al. A computational protocol to characterize elusive Candidate Phyla Radiation bacteria in oral environments using metagenomic data. bioRxiv 358812 (2018) doi:10.1101/358812.
He, X. et al. Cultivation of a human-associated TM7 phylotype reveals a reduced genome and epibiotic parasitic lifestyle. Proc. Natl. Acad. Sci. U.S.A. 112, 244–249 (2015).
Naud, S. et al. Candidate Phyla Radiation, an Underappreciated Division of the Human Microbiome, and Its Impact on Health and Disease. Clin Microbiol Rev e00140-21 (2022) doi:10.1128/cmr.00140-21.
Castelle, C. J. & Banfield, J. F. Major new microbial groups expand diversity and alter our understanding of the tree of life. Cell 172, 1181–1197 (2018).
Murugkar, P. P., Collins, A. J., Chen, T. & Dewhirst, F. E. Isolation and cultivation of candidate phyla radiation Saccharibacteria (TM7) bacteria in coculture with bacterial hosts. Journal of Oral Microbiology 12, 1814666 (2020).
Bor, B. et al. Insights Obtained by Culturing Saccharibacteria With Their Bacterial Hosts. J. Dent. Res. 22034520905792 (2020) doi:10.1177/0022034520905792.
Cross, K. L. et al. Targeted isolation and cultivation of uncultivated bacteria by reverse genomics. Nat. Biotechnol. 37, 1314–1321 (2019).
Parks, D. H. et al. Recovery of nearly 8,000 metagenome-assembled genomes substantially expands the tree of life. Nat Microbiol 2, 1533–1542 (2017).
Rinke, C. et al. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499, 431–437 (2013).
Dudek, N. K. et al. Novel microbial diversity and functional potential in the marine mammal oral microbiome. Current Biology 27, 3752–3762 (2017).
Dewhirst, F. E. et al. The human oral microbiome. J Bacteriol 192, 5002–5017 (2010).
Ling, Z. et al. Altered fecal microbiota composition associated with food allergy in infants. Appl Environ Microbiol 80, 2546–2554 (2014).
Liu, Y. et al. Effects of dihydroartemisinin on the gut microbiome of mice. Mol Med Rep 22, 707–714 (2020).
Kuehbacher, T. et al. Intestinal TM7 bacterial phylogenies in active inflammatory bowel disease. J. Med. Microbiol. 57, 1569–1576 (2008).
Stearns, J. C. et al. Bacterial biogeography of the human digestive tract. Sci Rep 1, 170 (2011).
Bor, B. et al. Phenotypic and physiological characterization of the epibiotic interaction between TM7x and its basibiont Actinomyces. Microbial ecology 71, 243–255 (2016).
Bedree, J. K. et al. Quorum Sensing Modulates the Epibiotic-Parasitic Relationship Between Actinomyces odontolyticus and Its Saccharibacteria epibiont, a Nanosynbacter lyticus Strain, TM7x. Front Microbiol 9, 2049 (2018).
Colson, P. et al. Full-length title: Dramatic HIV DNA degradation associated with spontaneous HIV suppression and disease-free outcome in a young seropositive woman following her infection. Sci Rep 10, (2020).
Belkacemi, S. et al. Passive Filtration, Rapid Scanning Electron Microscopy, and Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry for Treponema Culture and Identification from the Oral Cavity. J. Clin. Microbiol. 57, (2019).
Sizova, M. V., Doerfert, S. N., Gavrish, E. & Epstein, S. S. TM7 detection in human microbiome: Are PCR primers and FISH probes specific enough? J. Microbiol. Methods 114, 51–53 (2015).
Guindo, C. O., Davoust, B., Drancourt, M. & Grine, G. Diversity of Methanogens in Animals’ Gut. Microorganisms 9, 13 (2020).
Angelakis, E. et al. Glycans affect DNA extraction and induce substantial differences in gut metagenomic studies. Sci Rep 6, 26276 (2016).
Lo, C. I. et al. High-quality draft genome sequence and description of Haemophilus massiliensis sp. nov. Stand Genomic Sci 11, 31 (2016).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Bankevich, A. et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J. Comput. Biol. 19, 455–477 (2012).
Bosi, E. et al. MeDuSa: a multi-draft based scaffolder. Bioinformatics 31, 2443–2451 (2015).
Hadjadj, L. et al. Non contiguous-finished genome sequence and description of Microbacterium gorillae sp. nov. Stand Genomic Sci 11, (2016).
Tanizawa, Y., Fujisawa, T. & Nakamura, Y. DFAST: a flexible prokaryotic genome annotation pipeline for faster genome publication. Bioinformatics 34, 1037–1039 (2018).
Aziz, R. K. et al. The RAST Server: Rapid Annotations using Subsystems Technology. BMC Genomics 9, 75 (2008).
Kanehisa, M. & Goto, S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28, 27–30 (2000).
Database resources of the National Center for Biotechnology Information. Nucleic Acids Res 43, D6–D17 (2015).
Tatusov, R. L., Galperin, M. Y., Natale, D. A. & Koonin, E. V. The COG database: a tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res 28, 33–36 (2000).
Maatouk, M., Ibrahim, A., Rolain, J.-M., Merhej, V. & Bittar, F. Small and equipped: the rich repertoire of antibiotic resistance genes in Candidate Phyla Radiation genomes. http://biorxiv.org/lookup/doi/10.1101/2021.07.02.450847 (2021) doi:10.1101/2021.07.02.450847.
Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069 (2014).
Ibrahim, A. et al. Rhizomal Reclassification of Living Organisms. IJMS 22, 5643 (2021).
Ibrahim, A. et al. Adapted protocol for Saccharibacteria co-cultivation: two new members join the club of Candidate Phyla radiation. http://biorxiv.org/lookup/doi/10.1101/2021.07.23.453610 (2021) doi:10.1101/2021.07.23.453610.
Diakite, A. et al. Extensive culturomics of 8 healthy samples enhances metagenomics efficiency. PLoS One 14, (2019).
Lee, I., Ouk Kim, Y., Park, S.-C. & Chun, J. OrthoANI: An improved algorithm and software for calculating average nucleotide identity. Int. J. Syst. Evol. Microbiol. 66, 1100–1103 (2016).
Auch, A. F., von Jan, M., Klenk, H.-P. & Göker, M. Digital DNA-DNA hybridization for microbial species delineation by means of genome-to-genome sequence comparison. Stand Genomic Sci 2, 117–134 (2010).
Meier-Kolthoff, J. P., Auch, A. F., Klenk, H.-P. & Göker, M. Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics 14, 60 (2013).
Meier-Kolthoff, J. P. & Göker, M. TYGS is an automated high-throughput platform for state-of-the-art genome-based taxonomy. Nature Communications 10, 1–10 (2019).
Page, A. J. et al. Roary: rapid large-scale prokaryote pan genome analysis. Bioinformatics 31, 3691–3693 (2015).
Brinig, M. M., Lepp, P. W., Ouverney, C. C., Armitage, G. C. & Relman, D. A. Prevalence of bacteria of division TM7 in human subgingival plaque and their association with disease. Appl. Environ. Microbiol. 69, 1687–1694 (2003).
Hugenholtz, P., Tyson, G. W., Webb, R. I., Wagner, A. M. & Blackall, L. L. Investigation of candidate division TM7, a recently recognized major lineage of the domain Bacteria with no known pure-culture representatives. Appl. Environ. Microbiol. 67, 411–419 (2001).
Konstantinidis, K. T. & Tiedje, J. M. Towards a Genome-Based Taxonomy for Prokaryotes. J Bacteriol 187, 6258–6264 (2005).
Wrighton, K. C. et al. Fermentation, hydrogen, and sulfur metabolism in multiple uncultivated bacterial phyla. Science 337, 1661–1665 (2012).
Rappé, M. S. & Giovannoni, S. J. The uncultured microbial majority. Annual Reviews in Microbiology 57, 369–394 (2003).
Burstein, D. et al. Major bacterial lineages are essentially devoid of CRISPR-Cas viral defence systems. Nature communications 7, 10613 (2016).
Gomes, A. C., Hoffmann, C. & Mota, J. F. Gut microbiota is associated with adiposity markers and probiotics may impact specific genera. Eur J Nutr 59, 1751–1762 (2020).
Zhu, Q. et al. Visualization-assisted binning of metagenome assemblies reveals potential new pathogenic profiles in idiopathic travelers’ diarrhea. Microbiome 6, 201 (2018).
McLean, J. S. et al. Acquisition and Adaptation of Ultra-small Parasitic Reduced Genome Bacteria to Mammalian Hosts. Cell Rep 32, 107939 (2020).
Bernstein, D., Dewhirst, F. & Segre, D. Quantifying biosynthetic network robustness across the human oral microbiome. BioRxiv 392621 (2018).
Gómez-Gallego, C. et al. Human Breast Milk NMR Metabolomic Profile across Specific Geographical Locations and Its Association with the Milk Microbiota. Nutrients 10, 1355 (2018).
Togo, A. et al. Repertoire of human breast and milk microbiota: a systematic review. Future Microbiol 14, 623–641 (2019).
Fernández, L. et al. The human milk microbiota: Origin and potential roles in health and disease. Pharmacological Research 69, 1–10 (2013).
Grant, J. R. & Stothard, P. The CGView Server: a comparative genomics tool for circular genomes. Nucleic Acids Res. 36, W181-184 (2008).
Lefort, V., Desper, R. & Gascuel, O. FastME 2.0: A Comprehensive, Accurate, and Fast Distance-Based Phylogeny Inference Program. Mol Biol Evol 32, 2798–2800 (2015).

Table 1 is available in the Supplementary Files section.

Competing interest reported. DR was a consultant in microbiology for the Hitachi High-Tech Corporation until March 2021. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

PDF

Editorial decision: Major revision
06 Oct, 2022
Reviews received at journal
28 Sep, 2022
Reviewers agreed at journal
20 Sep, 2022
Reviewers invited by journal
19 Sep, 2022
Editor assigned by journal
14 Sep, 2022
Editor invited by journal
06 Sep, 2022
Submission checks completed at journal
06 Sep, 2022
First submitted to journal
02 Aug, 2022

You are reading this latest preprint version

Genomic description and prevalence of two new Candidatus Saccharibacteria species from the human gut in different samples and countries

Status:

Version 1

Abstract

Figures

Introduction

Results

1.1 Standard polymerase chain reaction (PCR)

1.2 Electron microscopy

3. Genomic comparison

3.1 Taxono-genomics classification

3.3 Pangenomic comparison

5. Metagenomic analysis

Discussion

Ethymology

Methods

1. Patients and specimens

2. Electron microscopy detection

3. Molecular biology detection

3.2 Next Generation Sequencing

4. Bioinformatic analysis

4.1 Genome assembly, scaffolding and finishing

4.2 Genome annotation and genes prediction

6. MetaGX database and statistics

7. Attempts to cultivate Saccharibacteria from human digestive samples

Declarations

References

Table 1

Additional Declarations

Supplementary Files

Status:

Version 1