how human gut microbes diversified with and adapted to their hosts
Coevolution of humans with gut bacteria requires a shared evolutionary history. Host-microbial cophylogeny has been shown for hominids, but within humans, only the stomach bacterium Helicobacter pylori is known to have codiversified with its host. Human populations are known to harbor distinct strains of bacterial species and this is generally attributed to differences in environment and diet, however, codiversification through long-term associations could also contribute to these patterns. To test for cophylogeny between gut microbiota and their human hosts, we analyzed paired human genotypes and bacterial strain genotypes from fecal metagenomes obtained from five countries. Our results indicate that strains of common gut bacteria have transmitted vertically for thousands of generations. In accord, strains displaying patterns of cophylogeny are also shared between mothers and infants. Patterns of strain transfer between populations are consistent with an African origin for taxa showing cophylogeny. Our results indicate long-term fidelity of gut microbial strains with human populations, creating opportunities for coevolution.
prof Ruth E. Ley
Ruth Ley is the Director of the Department of Microbiome Science at the Max Planck Institute for Developmental Biology and a Speaker for the Cluster of Excellence “Controlling Microbiomes to Fight Infection” with the University of Tuebingen, Germany. Since 2020 Ley is on the Scientific Council of the Institut Pasteur.
Critical Assessment of Metagenome Interpretation - the second round of challenges
Evaluating metagenomic software is key for optimizing metagenome interpretation and focus of the community-driven initiative for the Critical Assessment of Metagenome Interpretation (CAMI). In its second challenge, CAMI engaged the community to assess their methods on realistic and complex metagenomic datasets with long and short reads, created from ∼1,700 novel and known microbial genomes, as well as ∼600 novel plasmids and viruses. Altogether 5,002 results by 76 program versions were analyzed, representing a 22x increase in results.
Substantial improvements were seen in metagenome assembly, some due to using long-read data. The presence of related strains still was challenging for assembly and genome binning, as was assembly quality for the latter. Taxon profilers demonstrated a marked maturation, with taxon profilers and binners excelling at higher bacterial taxonomic ranks, but underperforming for viruses and archaea. Assessment of clinical pathogen detection techniques revealed a need to improve reproducibility. Analysis of program runtimes and memory usage identified highly efficient programs, including some top performers with other metrics. The CAMI II results identify current challenges, but also guide researchers in selecting methods for specific analyses
Quantification of gut bacterial strains following fecal transplantation
Fecal Microbiota Transplantation (FMT) while successful, lacks a quantitative identification of the discrete bacterial strains that stably engraft in recipients, and their association with clinical outcomes. Using the largest collection of >1,000 unique bacterial strains cultured from a combination of 22 FMT donors and recipients with recurrent Clostridioides difficile infection (rCDI), we develop an approach Strainer for detection and tracking of bacterial strains from metagenomic sequencing data. On application to 13 FMT interventions, we detect stable and high engraftment of 71% of microbiota strains in recipients at even 5-years post-transplant, a remarkably durable therapeutic from a single administration. Although ~80% of the original pre-FMT recipient strains were eliminated by the FMT, those strains that remain persist even 5 years later, along with newer strains acquired from the environment. The precise quantification of donor bacterial strain engraftment in recipients independently explained (precision 100%, recall 95%) the clinical outcomes of relapse or success after both initial and repeat-FMT. We next apply this to the largest successful FMT trial for patients with Ulcerative Colitis (FOCUS study), and found both overall and specific high engraftment of certain bacterial strains and species in patients that achieved the primary endpoint of steroid-free clinical remission with endoscopic remission or response (p val < 0.01). We provide a list of bacterial species and strains that are present in multiple donors and consistently engraft in recipients over time, for use in Live Biotherapeutic Products (LBP) as a safer alternative to FMT. Our framework can enable the systematic evaluation of different FMT and LBP study designs by quantification of strain engraftment in recipients.
Klebsiella oxytoca mediates colonization resistance against multi-drug resistant Klebsiella pneumoniae in the gut via cooperative carbohydrate competition
Gut colonization with multi-drug resistant (MDR) bacteria enhances the risk of bloodstream infections in susceptible individuals. We demonstrate highly-variable degrees of ex vivo colonization resistance against a carbapenem-resistant Klebsiella pneumoniae strain in human feces samples and subsequently isolated from protected donors diverse K. oxytoca strains. Several of these K. oxytoca strains reduced gut colonization of MDR K. pneumoniae strains in antibiotic-treated and gnotobiotic mouse models. Comparative analysis of K. oxytoca strains identified competition for specific carbohydrates as a key factor for colonization resistance. In addition to direct competition between K. oxytoca and K. pneumoniae, cooperation with additional commensals was required to reestablish full colonization resistance and gut decolonization. Finally, K. oxytoca also protected humanized microbiota mice generated from susceptible donors against K. pneumoniae colonization demonstrating the potential of commensal K. oxytoca strains as next-generation probiotics.
Integrative analysis of gut microbiota, microbial gene expression and function coupled with metabolome to understand mechanisms that influence the development of non-alcoholic fatty liver disease
Nonalcoholic fatty liver disease (NAFLD) affects a quarter of adults in North America and is a leading cause of liver-related mortality worldwide. Among those affected by NAFLD, 30% progress to a more severe and dangerous form, known as non-alcoholic steatohepatitis (NASH). Currently, there are no FDA-approved therapeutics for NASH. The human gut microbiome is a critical component of digestion, breaking down complex carbohydrates, proteins, and extent fats. Among other functions, gut microbiome is responsible for the conversion of primary bile acids produced in the liver to secondary bile acids via a multi-step process that involves multiple bacterial genes. Prior literature described changes in the microbiome in NAFLD and its progression to NASH, and identified taxa and metabolites associated with the condition. However, bacterial transcriptomic data are needed to understand how microbial functionality changes with changes in microbial composition. The physiological relevance of such changes in microbial functionality further require confirmation by interrogation of the microbial metabolome. To better understand the changes in microbiome function and its relationship to development and progression of NAFLD, we interrogated the intestinal ecosystem using a multi-omics (16S, metatransriptomic, metabolomic) approach to generate a comprehensive view of the functional dysbiosis across the histological spectrum of NAFLD. We identified microbial taxa expressing critical enzymes involved in fecal microbial bile acid transformations, and linked changes in taxa containing these enzymes to changes in bile acids. Specifically, among all secondary bile acids, deoxycholate (DCA)-derivatives were significantly increased with disease progression, which was related to increased expression of bile salt hydrolase, hydroxysteroid dehydrogenase and multiple genes encoded in the bai operon including baiCD the key enzyme in deoxycholate synthesis. Catenobacterium, Anerobifustis, Eggerthella and Holdemania emerged as key taxa with bile acid metabolic functions that were increased with the severity of NAFLD. These novel meta-transcriptomic metabolomic data provide evidence for widespread metabolic reprogramming and loss of function of the intestinal microbiome with development and progression of NASH. It also sets the stage of functional modulation of the microbiome to prevent and treat NASH.
Phage induction gives insight on the Anaerobic Digestion metavirome
Anaerobic Digestion (AD) is a process carried out by a microbial community composed of Bacteria and Archaea which degrade organic matter and produce biogas. This is a common process occurring in natural environments, but it is also a widely employed biotechnological approach for the production of methane and other chemical compounds. While the microbial community of AD has been thoroughly studied, the underlying viral community remains almost uninvestigated. Viruses have a great impact on the evolutionary dynamics of microbial communities across the globe, in terms of selective pressure, horizontal gene transfer and recycling of nutrients. For these reasons, it is likely that they play an important role in AD, and a better understanding about the virome composition and behavior would lead to improve the efficiency of the AD process. To investigate the role of phages in the AD microbial community, an induction experiment was run by setting up four batches under different stress conditions. Each sample was separated into a virus-enriched component and a microbe-enriched one via centrifugation and filtration. Methane production was measured at the end of the experiment in order to monitor the methanogenic activity. The supernatants and the pellets were sequenced separately on an Illumina platform and co-assembled. Metagenomic analysis of the assembly, including binning and the employment of a set of viral prediction tools, retrieved 120 bacterial metagenome assembled genomes (MAGs) and 705 putative viral genomes. Furthermore, integrated viral genomes were detected in 64 out of 120 MAGs. Taxonomic assignment of non-integrated phages showed a net prevalence of the order Caudovirales, and especially of the families Podoviridae and Siphoviridae. Functional annotation performed via EggNOG revealed that phage tail proteins are often associated with a glycosyl-hydrolase function (CAZy family GH33, sialidases) that could play a role in the mechanism of infection. Overall, these data offer a first glimpse about the complexity of the AD virome composition and dynamicity, showing how it reacts to environmental changes. Host predictions and the impact of viruses on gene-level evolution in the AD microbial community are still under investigation.
Human Lung Microbiota-driven Innate Immune Landscape
Human deep lung microbiota has been associated with lung health. Our recent study show a dynamic microbial ecosystem in lung harbouring discrete pneumotypes associated with specific immune response and clinical stability (Das, S. et al. Nat Commun 2021). We show that the predominant “balanced” pneumotype resembles a healthy lung and consists of a diverse bacterial community. We also establish a large collection of primary lung isolates called the Lung Microbiota culture Collection (LuMiCol) containing bacteria from major phyla in human lower respiratory tract, including the most prevalent bacteria and important lung pathogens. But despite sharing the same niche, we know little about direct interaction between these bacteria and human macrophages. Alveolar Macrophages (AM) are surveying cells in the human alveolar space with a unique phenotype. However, there is a lack of a standardized in vitro model for studying the largely unknown AM-lung bacteria interactions. Therefore, we repurpose THP-1 monocytes harboring a reporter for NFkB activation, to differentiate into macrophages. Using mass cytometry and cell-based assays, we show that these macrophages are phenotypically similar to AMs i.e. no background NFkB activation and similar surface markers. Thereby, we establish an AM-like macrophage model in vitro and a fast, easy, scalable screening strategy to investigate the inflammatory potential of diverse bacteria from human lung. By doing so, we discovered differential recognition of lung bacteria by pattern recognition receptors, varied NFkB activation and contact-dependent cytokine production in AM-like macrophages (Das et al. Unpublished). We also observe differential priming of macrophages by bacteria belonging to the balanced pneumotype, prior to pathogen challenge. Next, we show that the lung bacterial repertoire is mainly recognised by human TLR2. Finally, we used Small Artificial Communities (SACs) resembling the balanced pneumotype that confered varied degree of protection against inflammatory burst induced by pathogens such as Staphylococcus aureus and Pseudomonas aeruginosa. Amongst these, commensal Streptococcus mitis was observed to have a decisive role in dampening inflammation. Overall, we discover novel interactions of human lung bacteria and the innate immune system. We also uncover the potential of LUMICOL as a resource for individual bacteria or tailor-made communities with therapeutic potential in lung disease.
The aging mouse microbiome has obesogenic characteristics
Background: During aging, there is a physiological decline, an increase of morbidity and mortality, and a natural change in the gut microbiome. In this study, we investigated the influence of the gut microbiome on different metabolic parameters in adult and aged mice. Methods: Fecal and blood samples from adult (n = 42, 100‚Äì300 days) and aging (n = 32, 550‚Äì750 days) mice were collected. Microbiome analysis was done using QIIME2. Mouse weight and body composition were measured using NMR, and insulin and leptin levels in the blood were measured with Mouse Adipokine Magnetic Bead Panel kit. Fecal microbiota transplantation experiments from adult and aged mice into young germ-free mice were carried out in order to examine the effect of the gut microbiome of adult and aging mice on weight, body composition, insulin, and leptin. Results: We demonstrate that the microbiomes from adult and aged mice are distinguishable. We also report changes in metabolic parameters as we observed significantly higher weight and fat mass and low lean mass in aged compared to adult mice along with high insulin and leptin levels in the blood. The transplanted gut microbiome from aged mice transferred part of the phenotypes seen in aged mice. Fat body mass and insulin levels were higher in the mice who received feces from aged mice than mice receiving feces from adult mice. In addition, they consumed more food and had a higher respiratory quotient compared to mice receiving adult feces. Conclusions: We conclude that aged mice have a gut microbiota with obesogenic characteristics. In addition, the gut bacterial population itself is sufficient to induce some of the manifestations of obesity.
The gut microbiotas influence in the development of Foetal Alcohol Spectrum Disorders
Prenatal alcohol exposure (PAE) is the most preventable cause of birth defects, developmental disorders and intellectual disability, yet the prevalence of Foetal Alcohol Spectrum Disorders (FASD) in the Western Cape of South Africa is estimated to be between 16% and 31%, significantly higher than the global prevalence of 0.77%. Neurocognitive development is dependent on microbial composition and the corresponding microbial metabolic outputs, therefore alcohol-induced microbial alterations may alter infant gut microbiota functioning, increasing the risk of FASD development. This study compared the gut microbial composition of infants diagnosed with and without FASD by performing 16S ribosomal RNA V1-V2 sequencing on DNA extracted from 211 infant stool samples. The dada2 pipeline was used to pre-process the fastq sequencing files, create an amplicon sequence variant table, and assign taxonomy. Differential compositional analyses were performed using PhyloSeq, while vegan was used to compute the statistical analyses of microbial composition and calculate alpha- and beta-diversity. The infant gut microbiota was dominated by the gram-negative anaerobes Prevotella, Bacteroides and Faecalibacterium, with significant representations of the gram-positive anaerobe Bifidobacterium and facultative microorganisms Eshcerichia/Shigella. Bifidobacteria was found to be higher (p = 0.017) in infants diagnosed with FASD. A lower abundance of Bifidobacteria has been observed in children with Autism Spectrum Disorder (ASD), making this finding unexpected. Prevotella was higher (p = 0.003) in infants diagnosed with FASD, a finding that mirrors findings in individuals diagnosed with ASD in other low- and middle-income countries. Prevotella readily breaks down mucin – a structural component of mucus which protects the colon. An increased abundance of Prevotella may result in a compromised intestinal barrier, allowing bacteria and their metabolic outputs to enter the bloodstream and influence neurodevelopment. Although further studies are required, these findings are promising for microbe-based therapeutic interventions to reduce the extent of neurocognitive deficits and the debilitating symptoms associated with FASD. Additionally, identification of a gut microbial signature associated with FASD will hopefully lead to the development of a FASD predictor allowing for early identification of FASD cases which may facilitate maternal- and infant-directed intervention and support.
When 16s rRNA outperforms WMGS metagenomics
Background: The biggest drawback of 16S rRNA gene sequencing is that the reads originate from a single short region and the resulting reads lack sufficient specificity for reliable species-level identification. Whole metagenomic sequencing (WMGS) is seen as a solution to this problem that should provide the highest degree of specificity. We argue that WMGS for metataxonomics can be inefficient since most parts of a typical microbial genome are non-specific and provide no value for species identification. Consequently, the sequencing budget is spent on useless parts of genomes and the process’s sensitivity is significantly reduced. This is a problem, especially in the low-abundant samples contaminated by eukaryotic DNA. Unlike WMGS, where, in theory, all organisms could be classified down to the species level, in 16S rRNA the set of identifiable species depends on a chosen primer combination. Methods: We performed an in silico analysis of 46,235 publicly available microbial genomes and 5,658 primer combinations to investigate if, contrary to the widespread opinion, it is possible to use 16S rRNA workflow for species identification. Results: Mean phylum-balanced capture rate of the primer pairs ranged from 0.10 to 0.97 (mean 0.51), the mean classification accuracy for only captured species ranged from 0.07 to 0.83 (mean 0.57) and the overall mean classification accuracy ranged from 0.07 to 0.83 (mean 0.48). The identification performance also significantly varied between primer pairs spanning approximately the same 16s rRNA region (median IQR 0.23 for bacteria, 0.33 archaea). This was caused by variance in the capture rate (median IQR 0.51, 0.43), while the variance of the product specificity was much lower (median IQR 0.07, 0.2). The best primer combination for the identification of bacterial species was the pair encompassing all 16s rRNA variable regions with a 83.8% phylum-balanced species-level mean identification rate and 97.1% phylum-balanced mean capture rate. Overall, the classification accuracy of primers varied significantly across different phyla with mean standard deviation of 0.41, which illustrates the need for careful primer choice. Conclusion: Our results show that using the right primer combination, it is possible to guarantee that relevant species will be identified. Choosing 16S rRNA sequencing over WMGS can mean considerably lower costs in selected applications, with increased sensitivity and comparable specificity.
OPEN-ACCES PAPER HIGHLIGHT
Dispersal strategies shape persistence and evolution of human gut bacteria
Human gut bacterial strains can co-exist with their hosts for decades, but little is known about how these microbes persist and disperse, and thereby evolve. Here we examined these factors in 5,278 adult and infant fecal metagenomes, longitudinally sampled in individuals and families. Our analyses revealed that a subset of gut species has extreme persistence in individuals, families and geographic regions, represented often by locally successful strains of phylum Bacteroidota. These ‚Äútenacious‚Äù bacteria show high levels of genetic adaptation to the human host, but have also the highest probability to be lost in response to antibiotic interventions. By contrast, we found that bacteria in phylum Firmicutes often rely on dispersal strategies with weak phylogeographic patterns but strong family transmissions, likely related to sporulation. Our analysis describes how different dispersal strategies can lead to long-term persistence of human gut microbes and how this can be used in gut flora modulations.
Statistical approaches for differential expression analysis in metatranscriptomics
Motivation: Metatranscriptomics (MTX) has become an increasingly practical way to profile the functional activity of microbial communities in situ. However, MTX remains underutilized due to experimental and computational limitations. The latter are complicated by non-independent changes in both RNA transcript levels and their underlying genomic DNA copies (as microbes simultaneously change their overall abundance in the population and regulate individual transcripts), genetic plasticity (as whole loci are frequently gained and lost in microbial lineages), and measurement compositionality and zero-inflation. Here, we present a systematic evaluation of and recommendations for differential expression (DE) analysis in MTX. Results: We designed and assessed six statistical models for DE discovery in MTX that incorporate different combinations of DNA and RNA normalization and assumptions about the underlying changes of gene copies or species abundance within communities. We evaluated these models on multiple simulated and real multi-omic datasets. Models adjusting transcripts relative to their encoding gene copies as a covariate were significantly more accurate in identifying DE from MTX in both simulated and real datasets. Moreover, we show that when paired DNA measurements (meta-genomic data, MGX) are not available, models normalizing MTX measurements within-species while also adjusting for total-species RNA balance sensitivity, specificity, and interpretability of DE detection, as does filtering likely technical zeros. The efficiency and accuracy of these models pave the way for more effective MTX-based DE discovery in microbial communities. Availability: The analysis code and synthetic datasets used in this evaluation are available online at http://huttenhower.sph.harvard.edu/mtx2021.
Precise phylogenetic analysis of microbial isolates and genomes from metagenomes using PhyloPhlAn 3.0
Microbial genomes are available at an ever-increasing pace, as cultivation and sequencing become cheaper and obtaining metagenome-assembled genomes (MAGs) becomes more effective. Phylogenetic placement methods to contextualize hundreds of thousands of genomes must thus be efficiently scalable and sensitive from closely related strains to divergent phyla. We present PhyloPhlAn 3.0, an accurate, rapid, and easy-to-use method for large-scale microbial genome characterization and phylogenetic analysis at multiple levels of resolution. PhyloPhlAn 3.0 can assign genomes from isolate sequencing or MAGs to species-level genome bins built from >230,000 publically available sequences. For individual clades of interest, it reconstructs strain-level phylogenies from among the closest species using clade-specific maximally informative markers. At the other extreme of resolution, it scales to large phylogenies comprising >17,000 microbial species. Examples including Staphylococcus aureus isolates, gut metagenomes, and meta-analyses demonstrate the ability of PhyloPhlAn 3.0 to support genomic and metagenomic analyses.
Newly Explored Faecalibacterium Diversity Is Connected to Age, Lifestyle, Geography, and Disease
Faecalibacterium is prevalent in the human gut and a promising microbe for the development of next-generation probiotics (NGPs) or biotherapeutics. Analyzing reference Faecalibacterium genomes and almost 3,000 Faecalibacterium-like metagenome-assembled genomes (MAGs) reconstructed from 7,907 human and 203 non-human primate gut metagenomes, we identified the presence of 22 different Faecalibacterium-like species- level genome bins (SGBs), some further divided in different strains according to the subject geographical origin. Twelve SGBs are globally spread in the human gut and show different genomic potential in the utilization of complex polysaccharides, suggesting that higher SGB diversity may be related with increased utilization of plant-based foods. Moreover, up to 11 different species may co-occur in the same subject, with lower diversity in Western populations, as well as intestinal inflammatory states and obesity. The newly explored Faecalibacterium diversity will be able to support the choice of strains suitable as NGPs, guided by the consideration of the differences existing in their functional potential.
Francesca De Filippis
Dental Biofilm Microbiota Dysbiosis Is Associated With the Risk of Acute Graft-Versus-Host Disease After Allogeneic Hematopoietic Stem Cell Transplantation
Acute graft-versus-host disease (aGVHD) is one of the major causes of death after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Recently, aGVHD onset was linked to intestinal microbiota (IM) dysbiosis. However, other bacterial-rich gastrointestinal sites, such as the mouth, which hosts several distinctive microbiotas, may also impact the risk of GVHD. The dental biofilm microbiota (DBM) is highly diverse and, like the IM, interacts with host cells and modulates immune homeostasis. We characterized changes in the DBM of patients during allo-HSCT and evaluated whether the DBM could be associated with the risk of aGVHD. DBM dysbiosis during allo-HSCT was marked by a gradual loss of bacterial diversity and changes in DBM genera composition, with commensal genera reductions and potentially pathogenic bacteria overgrowths. High Streptococcus and high Corynebacterium relative abundance at preconditioning were associated with a higher risk of aGVHD (67% vs. 33%; HR = 2.89, P = 0.04 and 73% vs. 37%; HR = 2.74, P = 0.04, respectively), while high Veillonella relative abundance was associated with a lower risk of aGVHD (27% vs. 73%; HR = 0.24, P < 0.01). Enterococcus faecalis bloom during allo-HSCT was observed in 17% of allo-HSCT recipients and was associated with a higher risk of aGVHD (100% vs. 40%; HR = 4.07, P < 0.001) and severe aGVHD (60% vs. 12%; HR = 6.82, P = 0.01). To the best of our knowledge, this is the first study demonstrating that DBM dysbiosis is associated with the aGVHD risk after allo-HSCT.