Subscribe to our newsletter to receive the monthly program

Past Events·

Saturday, November 20, 2021

MVIF.3 | 9 & 11 November 2021

with Keynote talk by Prof. Frederic Bushman


Frederic Bushman is a Professor of Microbiology at the Perelman School of Medicine at the University of Pennsylvania. Dr. Bushman received his bachelor’s degree in Biology and English from Amherst College.  He received his PhD in cellular and developmental biology from Harvard University. His research focuses on host/microbe interactions, with specific projects focusing on the human microbiome, HIV pathogenesis, and human genome therapy.  Many recent projects in the laboratory involve developing new methods for studying microbiology using deep sequencing and bioinformatics. Dr. Bushman has published over 200 papers and two books.


Castalagin, a natural polyphenol, exerts antitumor activity and circumvents anti-PD-1 resistance through effects on the gut microbiome

Preclinical and clinical evidence indicate that the gut microbiome bacteria such as Akkermansia muciniphila modulate the clinical efficacy of immune checkpoint inhibitors (ICI). Therefore, strategies to beneficially shift microbiota composition represent a novel therapeutic avenue. Myrciaria dubia (MD) is a polyphenol-rich berry that has previously shown to increase A. muciniphila abundance in mice. Consequently, we sought to isolate the active polyphenol present in MD leading to a shift in microbiome and capable of increasing anti-PD-1 activity.

Materials and methods:
Daily oral gavage with MD or water was performed in MCA-205 (anti-PD-1 sensitive) and E0771 (anti-PD-1 resistant) murine tumor models treated with anti-PD-1. RP-HPLC was used to isolate the polyphenol’s bioactive compounds that were then tested in both tumor models. 16s rRNA sequencing and RT-PCR were performed to profile the murine fecal microbiome. Tumor immune infiltration was analyzed by flow cytometry, immunofluorescence and RNA sequencing. MD or isolated polyphenols were also tested in germ-free or antibiotic-treated mice that had undergone fecal microbiota transplantation (FMT) using feces from non-responder (NR) and responder (R) non-small cell lung cancer patients. Finally, fluorescence microscopy analysis was performed to examine labelled bacteria and their interaction with various polyphenols.

In this study, we showed that oral supplementation with MD in mice induced a shift in the gut microbiome, which translated into antitumor activity and restored the efficacy of anti-PD-1 therapy. We identified castalagin, a polyphenol, as the active compound in MD. Oral administration of castalagin enriched Ruminococcaceae, Alistipes and A.muciniphila, expanded TCM CD8+ cells and improved CD8+/Treg ratio within the tumor microenvironment. Oral supplementation of castalagin following FMT using NR patient feces in mice restored anti-PD-1 activity and favorably shifted the microbiome with enrichment of Alistipes, Christensenellaceae R-7 group and Ruminococcus. Using a fluorescein-castalagin conjugate, we showed that castalagin preferentially binds to R. bromii through direct interaction with its cellular envelope. Conclusion: Altogether, we isolated castalagin a comestible prebiotic with direct action on the microbiome and capable to circumvent anti-PD-1 resistance. Clinical trials combining castalagin to anti-PD-1 are currently ongoing.

Meriem Messaoudene

CRCHUM University of Montreal

Characterizing microbial community viability with RNA-based high-throughput sequencing

Characterization of microbial community viability is of great importance: essentially all sequence-based technologies do not differentiate living from dead microbes, whereas the functions and phenotypes of microbial communities are defined by biochemically active (“viable”) organisms. As a result, our understanding of microbial community structures and their potential transmission mechanisms remains incomplete. As a potential solution to this issue, 16S RNA-based amplicon sequencing (not rRNA gene, but the rRNA transcripts) has been proposed as a method to quantify the viable fraction of a microbial community, but its reliability has not been evaluated systematically. Here, we present our work to benchmark 16S-RNA-seq (targeting 16S rRNA transcripts and genes for parallel RNA and DNA sequencing) for viability assessment in synthetic and realistic microbial communities. In synthetic communities, we found that 16S-RNA-seq successfully reconstructed the mixtures of heat-killed Escherichia coli and Streptococcus sanguinis. We then applied this technique to natural communities (computer screens and mice, soil, and saliva) spiked with known concentrations of living and heat-killed E. coli to evaluate its performance in realistic conditions. No significant compositional differences were explained by the 16S-RNA assessment, suggesting that 16S-RNA-seq is not appropriate for viability assessment in complex communities. Results were slightly different when evaluated in environmental samples of similar origins (i.e. from Boston subway systems), samples were differentiated both by environments as well as by library type. Overall, these results show that 16S-RNA-seq has promise, but that previous literature assuming that 16S-rRNA amplicons are directly, quantitatively enriched for viable microbes is likely incorrect, as the technique does not reflect microbial viability outside of very simple, synthetic “communities.” We are currently continuing this work to develop new RNA amplicon-seq markers based on protein-encoding genes cpn60, as well as to improve viability assessment with multi-omic integration, e.g. combining amplicon-seq with functional indicators such as metatranscriptomic and metaproteomic profiles. These can circumvent some of the current limitations of 16S-RNA-seq, providing a complementary definition of viability, and directly observing activities such as virulence, pathogenicity, or antimicrobial resistance that are not captured by amplicon sequencing.

Ya Wang

Harvard University

Longitudinal disease-associated gut microbiome differences in infants with early food allergic manifestations

Complex interactions between the gut microbiome and immune cells in infancy are thought to be part of the pathogenesis for the marked rise in pediatric allergic diseases, particularly food allergies. Food protein-induced allergic proctocolitis (FPIAP) is commonly the earliest recognized non-IgE-mediated food allergy in infancy and is associated with atopic dermatitis and subsequent IgE-mediated food allergy later in childhood. Yet, a large prospective longitudinal study of the microbiome of infants with FPIAP (including samples prior to symptom onset) has not been done. Here we analyzed 954 longitudinal samples from 160 infants in a nested case-control study (81 who developed FPIAP, and 79 matched controls) from 1 week to 1 year of age by 16S rRNA ribosomal gene sequencing as part of the Gastrointestinal Microbiome and Allergic Proctocolitis (GMAP) Study. We confirmed that vaginally delivered infants had a greater abundance of Bacteroides, infants who received any breast milk had a greater abundance of Bifidobacterium, and that overall bacterial richness rose over the first year. We found key differences in the microbiome of infants with FPIAP, most strongly a higher abundance of a genus of Enterobacteriaceae and a lower abundance of a family of Clostridiales during the symptomatic period, as well as other key taxonomic differences across symptom states including prior to symptom onset. This study contributes to the larger body of literature examining structural development of the early life gut microbiome and provides a foundation for more mechanistic investigation into the pathogenesis and microbial effects on FPIAP and subsequent food allergic diseases in childhood.

Moran Yassour

The Hebrew University of Jerusalem


Engineering Genetically Intractable Bacteria: Targeted elimination of restriction-modification motifs in plasmid sequences with the SyngenicDNA Tool Generator (SyToGen)

The vast majority of bacteria that can be grown in a laboratory remain genetically intractable, beyond the power of genetics for elucidating function or engineering for human use. Inherent diversity of genetic defenses, primarily Restriction-Modification (RM) systems, across bacterial species and individual strains remains a fundamental barrier to human-made DNA constructs during genetic engineering of bacteria. Previously, we described an approach to evade RM systems through the creation of SyngenicDNA; de-novo synthesized DNA sequences that are precisely altered to achieve epigenetic compatibility with a desired bacterial host and to prevent RM defense activation upon artificial transformation. SyngenicDNA-based genetic tools have increased transformation efficiencies by several orders of magnitude, yet still require manual ad-hoc sequence alterations for their generation. To remove the need for manual introduction of site-specific DNA alterations in silico, and to improve the accessibility, speed of generation, and reproducibility of SyngenicDNA, we introduce the SyngenicDNA Tool Generator (SyToGen). SyToGen is a computational workbench designed to assist users who work with bacteria that are currently genetically intractable. The software adheres to the manual workflow for generating SyngenicDNA, namely target identification, in silico tool assembly and sequence adaptation, but rapidly and reproducibly provides users with RM-silent DNA templates for de novo synthesis and assembly. SyToGen output plasmid sequences retain the form and functionality of the original genetic tool, but are now uniquely re-designed at the DNA level to bypass RM barriers and operate within the specific bacterium of interest.

Gianmarco Piccinno

University of Trento

Microbiota links to neural dynamics supporting threat processing

There is growing recognition that the composition of the gut microbiota influences behaviour, including responses to threat. The cognitive-interoceptive appraisal of threat-related stimuli relies on dynamic neural computations between the anterior insular (AIC) and the dorsal anterior cingulate (dACC) cortices. If, to what extent, and how microbial consortia influence the activity of this cortical threat processing circuitry is unclear. We addressed this question by combining a threat processing task, neuroimaging, 16S rRNA profiling, and computational modelling in healthy participants. Results showed interactions between high-level ecological indices with threat-related AIC-dACC neural dynamics. At finer taxonomic resolutions, the abundance of Ruminococcus was differentially linked to connectivity between, and activity within the AIC and dACC during threat updating. Functional inference analysis provides a strong rationale to motivate future investigations of microbiota-derived metabolites in the observed relationship with threat-related brain processes.

Caitlin Hall

QIMR Berghofer Medical Research Institute

Healthy urogenital microbiome as a source of putative uropathogenic strains - Escherichia coli in the spotlight

To date, the human gut is recognized as primary reservoir of extraintestinal pathogenic Escherichia coli (ExPEC) strains, followed by vagina. However, since the discovery of urinary microbiome, the origin of microbes associated with UTI is of high interest. We aimed to explore population diversity of E. coli strains identified in urogenital microbiome of asymptomatic and recurrent UTI (rUTI) women. Additionally, we investigated genomic relation between E. coli strains isolated from healthy and diseased host. We performed thorough characterization [41 virulence-associated genes (VAGs) and phylogenetic groups] of 70 E. coli recovered from voided urine and vaginal samples. We further performed whole genome sequencing and antimicrobial susceptibility testing of 11 representative B2 ExPEC clones. Following, we performed SNP-based phylogenetic analysis and comparative genomics with E. coli genomes available in public databases, focusing on ST131 lineage. We identified E. coli in 48% of asymptomatic women and in all rUTI women. We observed different E. coli prevalence in urine (29%) and in vaginal samples (37%). Most of women were colonized by only one strain, while in 4 women we found more than one E. coli strain in urine and/or vagina. The most frequently identified phylogenetic group was B2 (56%) followed mostly by F and D. Most B2 and all F strains were classified as ExPEC and possess an array of VAGs, independently of host health status. Hierarchical clustering based on VAGs presence/absence confirmed that it is not possible to distinguish strains isolated from healthy and rUTI host based on their VAG profiles. We identified well recognized widespread lineages among our isolates e.g., sequence types (ST) 127, ST131, ST140 (asymptomatic) and ST12, ST73, ST131 (rUTI). Our isolates were occasionally resistant to antibiotics used in the clinic, with ST131 strain (asymptomatic woman) showing the highest resistance rate (6 antibiotics from 4 classes). Phylogenomics of ST131 and other analyzed lineages revealed close relatedness of genomes from healthy and diseased host and no phylogenetic distinction according to host health status. Our findings demonstrate that healthy urogenital microbiome is a source of potentially pathogenic E. coli strains, including lineages causing UTI e.g., ST131. A high prevalence of ExPEC clones, occasionally resistant to antibiotics should be of clinical concern and likely raises questions regarding etiology of UTI-associated clones.

Magdalena Ksiezarek

UCIBIO – Applied Molecular Biosciences Unit, REQUIMTE, Faculty of Pharmacy, Department of Biological Sciences, Laboratory of Microbiology, University of Porto, 4050-313 Porto, Portugal

Associate Laboratory i4HB – Institute for Health and Bioeconomy, Faculty of Pharmacy, University of Porto, 4050-313 Porto, Portugal

Microdiversity of the Vaginal Microbiome in Preterm Birth

Preterm birth (PTB) is the leading cause of neonatal mortality and morbidity. The vaginal microbiome has been associated with PTB, yet the mechanisms underlying this association remain largely unexplained. Our understanding of these associations could be informed by detecting microbial genomic adaptation to host-related stresses. To this end, we analyzed metagenomic data from 705 vaginal samples collected longitudinally during pregnancy from 40 women who delivered preterm spontaneously and 135 term controls from the Multi-Omic Microbiome Study-Pregnancy Initiative (MOMS-PI). We assembled 1,078 metagenome-assembled genomes, representing 157 species-level taxa from at least 8 bacterial phyla. We find that the vaginal microbiome of pregnancies that ended in PTB exhibits unique genetic profiles. Compared to pregnancies ending at term, it is more genetically diverse and harbors more virulence genes. It also harbors higher richness and diversity of antimicrobial resistant genes, likely promoted by plasmids. Interestingly, we found that Gardnerella vaginalis, a central vaginal pathobiont, is driving this high genetic diversity, particularly its genes involved in ribosome pathways and metabolism. We present evidence that this species undergoes more frequent recombination and stronger purifying selection in pregnancies which ended preterm. We further showed that the PTB-associated population structure of G. vaginalis may be related to optimization of growth rates. Overall, our results reveal novel associations between the vaginal microbiome and PTB from a population genetics perspective, and suggest that evolutionary processes acting on the vaginal microbiome may play a vital role in adverse pregnancy outcomes.

Jingqiu Liao

Columbia University

Rapid and accurate identification of ribosomal RNA sequences using deep learning

Advances in transcriptomic and translatomic techniques enable in-depth studies of RNA activity profiles and RNA-based regulatory mechanisms. Ribosomal RNA (rRNA) sequences are highly abundant among cellular RNA, but if the target sequences do not include polyadenylation, these cannot be easily removed in library preparation, requiring their post-hoc removal with computational techniques to accelerate and improve downstream analyses. Here we describe RiboDetector, a novel software based on a Bi-directional Long Short-Term Memory (BiLSTM) neural network, which rapidly and accurately identifies rRNA reads from transcriptomic, metagenomic, metatranscriptomic, noncoding RNA, and ribosome profiling sequence data. Compared with state-of-the-art approaches, Ribodetector can produce one order of magnitude fewer misclassifications. Importantly, the few false positives of RiboDetector on the benchmark dataset were not enriched in certain Gene Ontology (GO) terms, suggesting a low bias for downstream functional profiling. RiboDetector also demonstrated a remarkable generalization ability for detecting novel rRNA sequences divergent from the training data with sequence identities of less than 90%. On a personal computer, RiboDetector can process 40M reads in less than 6 minutes, which is ~50 times faster in GPU mode and ~15 times in CPU mode than other methods.

RiboDetector is available under a GPL v3.0 license at

Zhi-Luo Deng

Helmholtz Centre for Infection Research

A Microbiome Restoration Strategy Modulates the Gut Microbiome and Metabolic Markers in Healthy Adults

The dramatic increase of chronic diseases in industrialized societies might be driven in part by a disruption of gut microbiome composition and function (e.g., reduced fiber fermentation) and impaired intestinal barrier integrity. We developed a microbiome restoration strategy based on a diet that resembles key aspects of a non-industrialized diet (Non-Ind) and a probiotic Limosilactobacillus reuteri (a species rarely found in microbiome from industrialized populations). Using a randomized controlled pilot study, 30 participants consumed either the Non-Ind diet or their usual diet in a crossover fashion for three weeks each. Participants were also divided into three groups and consumed either a single dose of one of two Lm. reuteri strains or a placebo on day four of each diet period. The Non-Ind diet enhanced the temporal persistence of one of the Lm. reuteri strains, but had no measurable effects on microbiome and host. In contrast, the Non-Ind diet shifted overall fecal microbiome composition (R2=0.015, p=0.001; ADONIS), and significantly altered 56% of Amplicon Sequence Variants obtained from 16S dataset (FDR<0.05), 22% of species-level genome bins, and 21% of metabolic pathways obtained from metagenomic dataset (FDR<0.05). Bacteria fermentation was enhanced, exhibited by increased concentration of total fecal short-chain fatty acids (+10.7%, p=0.03) and reduced fecal pH (-3.8%, p=0.002). Furthermore, we observed reductions in six chronic disease risk markers (all p<0.01): total cholesterol (-14.1%), low-density lipoprotein cholesterol (-16.8%), high-density lipoprotein (HDL) cholesterol (-11.3%), non-HDL cholesterol (-15.2%), glucose (-6.3%), and C-reactive protein (-14.2%). The diet also reduced fecal calprotectin (-21.0%, p=0.002) and zonulin levels (-14.9%, p=0.025) that are markers of gut inflammation and impaired barrier functions, respectively. Our study demonstrates pronounced beneficial effects of a microbiome restoration strategy based on the Non-Ind diet on metabolic markers of health. The findings provide valuable information for improving human health in modern societies. Ongoing analyses are exploring the mechanistic links between diet-induced changes in the gut microbiome and the physiological effects on the host.

Fuyong Li

Department of Agricultural, Food, and Nutritional Science, University of Alberta, Edmonton, Alberta, Canada


Large-scale genome-wide analysis links lactic acid bacteria from food with the gut microbiome

Lactic acid bacteria (LAB) are fundamental in the production of fermented foods and several strains are regarded as probiotics. Large quantities of live LAB are consumed within fermented foods, but it is not yet known to what extent the LAB we ingest become members of the gut microbiome. By analysis of 9445 metagenomes from human samples, we demonstrate that the prevalence and abundance of LAB species in stool samples is generally low and linked to age, lifestyle, and geography, with Streptococcus thermophilus and Lactococcus lactis being most prevalent. Moreover, we identify genome-based differences between food and gut microbes by considering 666 metagenome-assembled genomes (MAGs) newly reconstructed from fermented food microbiomes along with 154,723 human MAGs and 193,078 reference genomes. Our large-scale genome-wide analysis demonstrates that closely related LAB strains occur in both food and gut environments and provides unprecedented evidence that fermented foods can be indeed regarded as a possible source of LAB for the gut microbiome.

Edoardo Pasolli

Department of Agricultural Sciences, University of Naples Federico II, Portici, Italy

Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3

Culture-independent analyses of microbial communities dramatically improved in the last decade, thanks to advances in methods for biological profiling via shotgun metagenomics and other molecular methods. This opens new opportunities for large-scale assembly efforts and the improvement of analysis tools with greater access to multi-omics, microbial reference genomes, and strain-level diversity. To leverage these, we present bioBakery 3, a set of integrated and improved methods for taxonomic, strain-level, functional, and phylogenetic profiling of metagenomes based on the largest set of reference sequences now available. MetaPhlAn 3 and HUMAnN 3 provides more accurate taxonomic and functional potential and activity profiles, respectively. These methods detected novel disease-microbiome links in CRC and IBD cohorts. Strain-level profiling of 4,077 metagenomes with StrainPhlAn 3 and PanPhlAn 3 unravelled the phylogenetic and functional structure of the gut microbe Ruminococcus bromii, previously described by only 15 isolate genomes. Phylogenetic analysis with PhyloPhlAn 3 supports both genomic and metagenomic data, by assigning genomes from isolate sequencing or metagenomic assemblies to species-level genome bins defined on >230,000 publically available sequences. It also accurately reconstructs phylogenies at different resolutions (from strain-level to microbial tree-of-life) using maximally informative markers. The bioBakery 3 is open-source and includes documentation, training data, and cloud-deployable reproducible workflows to help researchers deepen the resolution, scale, and accuracy of multi-omic profiling for microbial communities. The ability to use metagenomic assemblies further help to study uncharacterized microbial species not represented by isolate genomes.


Francesco Beghini

University of Trento