Strain/species identification in metagenomes using genome-specific markers

Shotgun metagenome sequencing has become a fast, cheap and high-throughput technology for characterizing microbial communities in complex environments and human body sites. However, accurate identification of microorganisms at the strain/species level remains extremely challenging. We present a novel k-mer-based approach, termed GSMer, that identifies genome-specific markers (GSMs) from currently sequenced microbial genomes, which were then used for strain/species-level identification in metagenomes. Using 5390 sequenced microbial genomes, 8 770 321 50-mer strain-specific and 11 736 360 species-specific GSMs were identified for 4088 strains and 2005 species (4933 strains), respectively. The GSMs were first evaluated against mock community metagenomes, recently sequenced genomes and real metagenomes from different body sites, suggesting that the identified GSMs were specific to their targeting genomes. Sensitivity evaluation against synthetic metagenomes with different coverage suggested that 50 GSMs per strain were sufficient to identify most microbial strains with ≥0.25× coverage, and 10% of selected GSMs in a database should be detected for confident positive callings. Application of GSMs identified 45 and 74 microbial strains/species significantly associated with type 2 diabetes patients and obese/lean individuals from corresponding gastrointestinal tract metagenomes, respectively. Our result agreed with previous studies but provided strain-level information. The approach can be directly applied to identify microbial strains/species from raw metagenomes, without the effort of complex data pre-processing.     

Searching Biomarkers in the Sequenced Genomes of <Emphasis Type="Italic">Staphylococcus</Emphasis> for their Rapid Identification

Bacterial identification using rrs (16S rRNA) gene is widely reported. Bacteria possessing multiple copies of rrs lead to overestimation of its diversity. Staphylococcus genomes carries 5–6 copies of rrs showing high similarity in their nucleotide sequences, which lead to ambiguous results. The genomes of 31 strains of Staphylococcus representing 7 species were searched for the presence of common genes. In silico digestion of 34 common genes using 10 restriction endonucleases (REs) lead to select gene-RE combinations, which could be used as biomarkers. RE digestion of recA allowed unambiguous identification of 13 genomes representing all the 7 species. In addition, a few more genes (argH, argR, cysS, gyrB, purH, and pyrE) and RE combinations permitted further identification of 12 strains. By employing additional RE and genes unique to a particular strain, it was possible to identify the rest 6 Staphylococcus aureus strains. This approach has the potential to be utilized for rapid detection of Staphylococcus strains.

Electronic supplementary material

The online version of this article (doi:10.1007/s12088-016-0565-9) contains supplementary material, which is available to authorized users.     

A novel method for screening species-specific gDNA probes for species identification

We report a method called SSH array which combines the suppression subtraction hybridization (SSH) and DNA array techniques to find species-specific DNA probes from genomic DNA (gDNA) for species identification. The method first obtains the differential gDNA fragments between two species by SSH and then hybridizes the differential gDNA fragments with arrays made of multiple whole genomes from several species to screen the unique gDNA fragments for one species. The screened unique gDNA fragments can be used as species-specific probes to differentiate the species they represent from all other species. We used five species of the genus Dendrobrium, D.aurantiacum Kerr, D.officinale Kimura et Migo, D.nobile Lindl., D.chrysotoxum Lindl. and D.fimbriatum Hook., as experimental materials to study the feasibility of the method. The results showed that the method could efficiently obtain different species-specific probes for each of the five species.     

Species-specific oligonucleotide probes for the detection and identification of Lactobacillus isolated from mouse faeces

AIMS: To develop species-specific monitoring techniques for rapid detection and identification of Lactobacillus isolated from mouse faeces. METHODS AND RESULTS: The specificity of oligonucleotide probes was evaluated by dot blot hybridization to 16S rDNA and 23S rDNA amplified by PCR from 12 Lactobacillus type strains and 100 strains of Lactobacillus isolated from mouse faeces. Oligonucleotide probes specific for each Lactobacillus species hybridized only with targeted rDNA. The Lactobacillus strains isolated from mouse faeces were identified mainly as Lactobacillus intestinalis, L. johnsonii, L. murinus and L. reuteri using species-specific probes. 16S rDNA of eight unidentified isolates were sequenced and two new probes were designed. Four of eight strains of unhybridized Lactobacillus were identified as L. johnsonii/gasseri group, and the remaining four strains as L. vaginalis. CONCLUSIONS: The species-specific probe set of L. intestinalis, L. johnsonii, L. murinus, L. reuteri and L. vaginalis in this study was efficient for rapid identification of Lactobacillus isolated from mouse faeces. SIGNIFICANCE AND IMPACT OF THE STUDY: The oligonucleotide probe set for Lactobacillus species harboured in the mouse intestine, can be used for rapid identification of lactobacilli and monitoring of the faecal Lactobacillus community.     

Quantification of Gordona amarae Strains in Foaming Activated Sludge and Anaerobic Digester Systems with Oligonucleotide Hybridization Probes

Previous studies have shown the predominance of mycolic acid-containing filamentous actinomycetes (mycolata) in foam layers in activated sludge systems. Gordona (formerly Nocardia) amarae often is considered the major representative of this group in activated sludge foam. In this study, small-subunit rRNA genes of four G. amarae strains were sequenced, and the resulting sequences were compared to the sequence of G. amarae type strain SE-6. Comparative sequence analysis showed that the five strains used represent two lines of evolutionary descent; group 1 consists of strains NM23 and ASAC1, and group 2 contains strains SE-6, SE-102, and ASF3. The following three oligonucleotide probes were designed: a species-specific probe for G. amarae, a probe specific for group 1, and a probe targeting group 2. The probes were characterized by dissociation temperature and specificity studies, and the species-specific probe was evaluated for use in fluorescent in situ hybridizations. By using the group-specific probes, it was possible to place additional G. amarae isolates in their respective groups. The probes were used along with previously designed probes in membrane hybridizations to determine the abundance of G. amarae, group 1, group 2, bacterial, mycolata, and Gordona rRNAs in samples obtained from foaming activated sludge systems in California, Illinois, and Wisconsin. The target groups were present in significantly greater concentrations in activated sludge foam than in mixed liquor and persisted in anaerobic digesters. Hybridization results indicated that the presence of certain G. amarae strains may be regional or treatment plant specific and that previously uncharacterized G. amarae strains may be present in some systems.     

Quantification of Different Eubacterium spp. in Human Fecal Samples with Species-Specific 16S rRNA-Targeted Oligonucleotide Probes

Species-specific 16S rRNA-targeted, Cy3 (indocarbocyanine)-labeled oligonucleotide probes were designed and validated to quantify different Eubacterium species in human fecal samples. Probes were directed at Eubacterium barkeri, E. biforme, E. contortum, E. cylindroides (two probes), E. dolichum, E. hadrum, E. lentum, E. limosum, E. moniliforme, and E. ventriosum. The specificity of the probes was tested with the type strains and a range of common intestinal bacteria. With one exception, none of the probes showed cross-hybridization under stringent conditions. The species-specific probes were applied to fecal samples obtained from 12 healthy volunteers. E. biforme, E. cylindroides, E. hadrum, E. lentum, and E. ventriosum could be determined. All other Eubacterium species for which probes had been designed were under the detection limit of 10⁷ cells g (dry weight) of feces⁻¹. The cell counts obtained are essentially in accordance with the literature data, which are based on colony counts. This shows that whole-cell in situ hybridization with species-specific probes is a valuable tool for the enumeration of Eubacterium species in feces.     

NotI passporting to identify species composition of complex microbial systems

We describe here a new method for large-scale scanning of microbial genomes on a quantitative and qualitative basis. To achieve this aim we propose to create NotI passports: databases containing NotI tags. We demonstrated that these tags comprising 19 bp of sequence information could be successfully generated using DNA isolated from intestinal or fecal samples. Such NotI passports allow the discrimination between closely related bacterial species and even strains. This procedure for generating restriction site tagged sequences (RSTS) is called passporting and can be adapted to any other rare cutting restriction enzyme. A comparison of 1312 tags from available sequenced Escherichia coli genomes, generated with the NotI, PmeI and SbfI restriction enzymes, revealed only 219 tags that were not unique. None of these tags matched human or rodent sequences. Therefore the approach allows analysis of complex microbial mixtures such as in human gut and identification with high accuracy of a particular bacterial strain on a quantitative and qualitative basis.     

Phylogeny in Aid of the Present and Novel Microbial Lineages: Diversity in Bacillus

Bacillus represents microbes of high economic, medical and biodefense importance. Bacillus strain identification based on 16S rRNA sequence analyses is invariably limited to species level. Secondly, certain discrepancies exist in the segregation of Bacillus subtilis strains. In the RDP/NCBI databases, out of a total of 2611 individual 16S rDNA sequences belonging to the 175 different species of the genus Bacillus, only 1586 have been identified up to species level. 16S rRNA sequences of Bacillus anthracis (153 strains), B. cereus (211 strains), B. thuringiensis (108 strains), B. subtilis (271 strains), B. licheniformis (131 strains), B. pumilus (83 strains), B. megaterium (47 strains), B. sphaericus (42 strains), B. clausii (39 strains) and B. halodurans (36 strains) were considered for generating species-specific framework and probes as tools for their rapid identification. Phylogenetic segregation of 1121, 16S rDNA sequences of 10 different Bacillus species in to 89 clusters enabled us to develop a phylogenetic frame work of 34 representative sequences. Using this phylogenetic framework, 305 out of 1025, 16S rDNA sequences presently classified as Bacillus sp. could be identified up to species level. This identification was supported by 20 to 30 nucleotides long signature sequences and in silico restriction enzyme analysis specific to the 10 Bacillus species. This integrated approach resulted in identifying around 30% of Bacillus sp. up to species level and revealed that B. subtilis strains can be segregated into two phylogenetically distinct groups, such that one of them may be renamed.     

Genetic Identification of Members of the Genus Corynebacterium at Genus and Species Levels with 16S rDNA-Targeted Probes

16S rRNA gene-targeted probes were designed for the identification of corynebacteria at the genus and species levels. The genus-specific probe hybridized all clinically important members of the genus Corynebacterium and could distinguish them from other coryneform bacteria and phylogenetically related high G + C% gram-positive bacteria, including Actinomyces, Rhodococcus, Gordona, Nocardia, Streptomyces, Brevibacterium and Mycobacterium. The species-specific probes for C. jeikeium and C. diphtheriae could differentiate these two species from other members of this genus. The probes were used to select corynebacteria among gram-positive clinical isolates which had been tentatively identified as corynebacteria by biochemical tests. We screened 59 strains with the genus-specific probe; 51 strains hybridized to the genus-specific probe, 8 did not. Of the 51 strains that hybridized to the genus-specific probe, 1 hybridized to the C. diphtheriae species probe and 13 hybridized to the C. jeikeium species probe. The 8 strains that did not hybridize to the genus probe were further characterized by analyzing cell wall diaminopimelic acid and partial 16S rRNA sequencing. The results indicated that these strains were distributed in the genera Arthrobacter and Brevibacterium.     

Whole Genome Sequence of the Treponema Fribourg-Blanc: Unspecified Simian Isolate Is Highly Similar to the Yaws Subspecies

Background

Unclassified simian strain Treponema Fribourg-Blanc was isolated in 1966 from baboons (Papio cynocephalus) in West Africa. This strain was morphologically indistinguishable from T. pallidum ssp. pallidum or ssp. pertenue strains, and it was shown to cause human infections.

Methodology/Principal Findings

To precisely define genetic differences between Treponema Fribourg-Blanc (unclassified simian isolate, FB) and T. pallidum ssp. pertenue strains (TPE), a high quality sequence of the whole Fribourg-Blanc genome was determined with 454-pyrosequencing and Illumina sequencing platforms. Combined average coverage of both methods was greater than 500×. Restriction target sites (n = 1,773), identified in silico, of selected restriction enzymes within the Fribourg-Blanc genome were verified experimentally and no discrepancies were found. When compared to the other three sequenced TPE genomes (Samoa D, CDC-2, Gauthier), no major genome rearrangements were found. The Fribourg-Blanc genome clustered with other TPE strains (especially with the TPE CDC-2 strain), while T. pallidum ssp. pallidum strains clustered separately as well as the genome of T. paraluiscuniculi strain Cuniculi A. Within coding regions, 6 deletions, 5 insertions and 117 substitutions differentiated Fribourg-Blanc from other TPE genomes.

Conclusions/Significance

The Fribourg-Blanc genome showed similar genetic characteristics as other TPE strains. Therefore, we propose to rename the unclassified simian isolate to Treponema pallidum ssp. pertenue strain Fribourg-Blanc. Since the Fribourg-Blanc strain was shown to cause experimental infection in human hosts, non-human primates could serve as possible reservoirs of TPE strains. This could considerably complicate recent efforts to eradicate yaws. Genetic differences specific for Fribourg-Blanc could then contribute for identification of cases of animal-derived yaws infections.     

Rapid identification and detection of pine pathogenic fungi associated with mountain pine beetles by padlock probes

Fifteen million hectares of pine forests in western Canada have been attacked by the mountain pine beetle (Dendroctonus ponderosae; MPB), leading to devastating economic losses. Grosmannia clavigera and Leptographium longiclavatum, are two fungi intimately associated with the beetles, and are crucial components of the epidemic. To detect and discriminate these two closely related pathogens, we utilized a method based on ligase-mediated nucleotide discrimination with padlock probe technology, and signal amplification by hyperbranched rolling circle amplification (HRCA). Two padlock probes were designed to target species-specific single nucleotide polymorphisms (SNPs) located at the inter-generic spacer 2 region and large subunit of the rRNA respectively, which allows discrimination between the two species. Thirty-four strains of G. clavigera and twenty-five strains of L. longiclavatum representing a broad geographic origin were tested with this assay. The HRCA results were largely in agreement with the conventional identification based on morphology or DNA-based methods. Both probes can also efficiently distinguish the two MPB-associated fungi from other fungi in the MPB, as well as other related fungi in the order Ophiostomatales. We also tested this diagnostic method for the direct detection of these fungi from the DNA of MPB. A nested PCR approach was used to enrich amplicons for signal detection. The results confirmed the presence of these two fungi in MPB. Thus, the padlock probe assay coupled with HRCA is a rapid, sensitive and reproducible method for the identification and detection of these ophiostomatoid fungi.     

Comparison of Streptococcus thermophilus strains by pulse field gel electrophoresis of genomic DNA

Pulse field gel electrophoresis (PFGE) was utilised to compare the genomes of 16 Streptococcus thermophilus cultures from yoghurt, cheese, laban and dahi after digestion with the restriction endonucleases, SfiI, SmaI and BssHII. PFGE profiles could be used for strain identification and were also useful in predicting relatedness of certain strains. Genetic variations between specific morphotypes of a highly proteolytic culture were not detectable by PFGE in this study. Statistical analysis of SmaI restriction patterns enabled the clustering of strains into two groups which corresponded with biochemical properties of the strains examined and suggested that PFGE profiles could be useful in predicting biochemical characteristics.     

Isolation of low-copy-number sequences that neighbor satellite DNA in mammals

To investigate the role of satellite DNA in eukaryotic genomes, we isolated from an African green monkey (Cercopithecus aethiops) genomic library cloned segments containing the previously described deca-satellite linked to low-copy-number genomic sequences. Three such clones were obtained. The low-copy-number sequences in the three clones do not cross-hybridize suggesting that they derive from different genomic loci. The structure of one of the clones, λAMkA, is described in detail. Subcloned segments containing the low-copy-number sequences from λAMkA anneal to monkey, human and mouse genomic DNA. The subcloned probes were used to select clones containing homologous sequences from a second, independent monkey library as well as from human and mouse genomic libraries. Several of the newly isolated monkey clones hybridized to probes containing the species-specific deca- and -satellites, confirming the genomic association of the low-copy-number sequence in λAMkA with satellite DNA. Moreover, several of the human and mouse clones hybridized to species-specific human and mouse satellite DNAs, respectively. These experiments indicate that the low-copy-number sequence in λMkA and its association with satellite DNA is conserved in primates and rodents.     

Expanding the catalog of cas genes with metagenomes

The CRISPR (clusters of regularly interspaced short palindromic repeats)–Cas adaptive immune system is an important defense system in bacteria, providing targeted defense against invasions of foreign nucleic acids. CRISPR–Cas systems consist of CRISPR loci and cas (CRISPR-associated) genes: sequence segments of invaders are incorporated into host genomes at CRISPR loci to generate specificity, while adjacent cas genes encode proteins that mediate the defense process. We pursued an integrated approach to identifying putative cas genes from genomes and metagenomes, combining similarity searches with genomic neighborhood analysis. Application of our approach to bacterial genomes and human microbiome datasets allowed us to significantly expand the collection of cas genes: the sequence space of the Cas9 family, the key player in the recently engineered RNA-guided platforms for genome editing in eukaryotes, is expanded by at least two-fold with metagenomic datasets. We found genes in cas loci encoding other functions, for example, toxins and antitoxins, confirming the recently discovered potential of coupling between adaptive immunity and the dormancy/suicide systems. We further identified 24 novel Cas families; one novel family contains 20 proteins, all identified from the human microbiome datasets, illustrating the importance of metagenomics projects in expanding the diversity of cas genes.     

A direct method for the chromosomal assignment of DNA markers in Leishmania

A simple method for the chromosomal assignment of any DNA marker would be an important tool for the ongoing project to map the genome of the protozoan parasite Leishmania. The Leishmania chromosomes enter pulsed field gel electrophoresis (PFGE) gels under current electrophoretic conditions, but their direct identification in a given strain is hampered by their stacking in a few chromosomal bands, and by the very frequent size variations of the same chromosome among parasite strains. To overcome these problems, we determined the complete karyotypes of 12 Old World Leishmania cloned strains. This enabled us to select three of these strains that display great chromosome size polymorphisms, such that every chromosome can be individualized by a specific pattern after hybridization onto these three karyotypes. The complete resolution of the genomes of these three strains can be carried out with only three electrophoretic conditions. This makes a series of three blots sufficient for the assignment of any new marker on a particular Leishmania chromosome.     

Distinct signatures of host–microbial meta-metabolome and gut microbiome in two C57BL/6 strains under high-fat diet

A combinatory approach using metabolomics and gut microbiome analysis techniques was performed to unravel the nature and specificity of metabolic profiles related to gut ecology in obesity. This study focused on gut and liver metabolomics of two different mouse strains, the C57BL/6J (C57J) and the C57BL/6N (C57N) fed with high-fat diet (HFD) for 3 weeks, causing diet-induced obesity in C57N, but not in C57J mice. Furthermore, a 16S-ribosomal RNA comparative sequence analysis using 454 pyrosequencing detected significant differences between the microbiome of the two strains on phylum level for Firmicutes, Deferribacteres and Proteobacteria that propose an essential role of the microbiome in obesity susceptibility. Gut microbial and liver metabolomics were followed by a combinatory approach using Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) and ultra performance liquid chromatography time of tlight MS/MS with subsequent multivariate statistical analysis, revealing distinctive host and microbial metabolome patterns between the C57J and the C57N strain. Many taurine-conjugated bile acids (TBAs) were significantly elevated in the cecum and decreased in liver samples from the C57J phenotype likely displaying different energy utilization behavior by the bacterial community and the host. Furthermore, several metabolite groups could specifically be associated with the C57N phenotype involving fatty acids, eicosanoids and urobilinoids. The mass differences based metabolite network approach enabled to extend the range of known metabolites to important bile acids (BAs) and novel taurine conjugates specific for both strains. In summary, our study showed clear alterations of the metabolome in the gastrointestinal tract and liver within a HFD-induced obesity mouse model in relation to the host–microbial nutritional adaptation.     

Using high-density exon arrays to profile gene expression in closely related species

Global comparisons of gene expression profiles between species provide significant insight into gene regulation, evolutionary processes and disease mechanisms. In this work, we describe a flexible and intuitive approach for global expression profiling of closely related species, using high-density exon arrays designed for a single reference genome. The high-density probe coverage of exon arrays allows us to select identical sets of perfect-match probes to measure expression levels of orthologous genes. This eliminates a serious confounding factor in probe affinity effects of species-specific microarray probes, and enables direct comparisons of estimated expression indexes across species. Using a newly designed Affymetrix exon array, with eight probes per exon for approximately 315 000 exons in the human genome, we conducted expression profiling in corresponding tissues from humans, chimpanzees and rhesus macaques. Quantitative real-time PCR analysis of differentially expressed candidate genes is highly concordant with microarray data, yielding a validation rate of 21/22 for human versus chimpanzee differences, and 11/11 for human versus rhesus differences. This method has the potential to greatly facilitate biomedical and evolutionary studies of gene expression in nonhuman primates and can be easily extended to expression array design and comparative analysis of other animals and plants.     

Insertions or Deletions (Indels) in the rrn 16S-23S rRNA Gene Internal Transcribed Spacer Region (ITS) Compromise the Typing and Identification of Strains within the Acinetobacter calcoaceticus-baumannii (Acb) Complex and Closely Related Members

To determine whether ITS sequences in the rrn operon are suitable for identifying individual Acinetobacter Acb complex members, we analysed length and sequence differences between multiple ITS copies within the genomes of individual strains. Length differences in ITS reported previously between A. nosocomialis BCRC15417^T (615 bp) and other strains (607 bp) can be explained by presence of an insertion (indel 13i/1) in the longer ITS variant. The same Indel 13i/1 was also found in ITS sequences of ten strains of A. calcoaceticus, all 639 bp long, and the 628 bp ITS of Acinetobacter strain BENAB127. Four additional indels (13i/2–13i/5) were detected in Acinetobacter strain c/t13TU 10090 ITS length variants (608, 609, 620, 621 and 630 bp). These ITS variants appear to have resulted from horizontal gene transfer involving other Acinetobacter species or in some cases unrelated bacteria. Although some ITS copies in strain c/t13TU 10090 are of the same length (620 bp) as those in Acinetobacter strains b/n1&3, A. pittii (10 strains), A. calcoaceticus and A. oleivorans (not currently acknowledged as an Acb member), their individual ITS sequences differ. Thus ITS length by itself can not by itself be used to identify Acb complex strains. A shared indel in ITS copies in two separate Acinetobacter species compromises the specificity of ITS targeted probes, as shown with the Aun-3 probe designed to target the ITS in A. pitti. The presence of indel 13i/5 in the ITS of Acinetobacter strain c/t13TU means it too responded positively to this probe. Thus, neither ITS sequencing nor the currently available ITS targeted probes can distinguish reliably between Acb member species.