CoChat
Sign Up
Cited by 330Gut microbiota and healthMicrobial Community Ecology and PhysiologyOral microbiology and periodontitis research

The DOI matches its Crossref record; 78 of the 78 references it cites were found in a registry or an index.

RapidAIM: a culture- and metaproteomics-based Rapid Assay of Individual Microbiome responses to drugs

Open access
Li, Leyuan, Ning, Zhibin, Zhang, Xu, Mayne, Janice, Cheng, Kai, Stintzi, Alain, Figeys, Daniel·2020·Microbiome
DOI 10.1186/s40168-020-00806-zPMC PMC7066843

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Human-targeted drugs may exert off-target effects or can be repurposed to modulate the gut microbiota. However, our understanding of such effects is limited due to a lack of rapid and scalable assay to comprehensively assess microbiome responses to drugs. Drugs and other compounds can drastically change the overall abundance, taxonomic composition, and functions of a gut microbiome. Results: Here, we developed an approach to screen compounds against individual microbiomes in vitro, using metaproteomics to both measure absolute bacterial abundances and to functionally profile the microbiome. Our approach was evaluated by testing 43 compounds (including 4 antibiotics) against 5 individual microbiomes. The method generated technically highly reproducible readouts, including changes of overall microbiome abundance, microbiome composition, and functional pathways. Results show that besides the antibiotics, the compounds berberine and ibuprofen inhibited the accumulation of biomass during in vitro growth of the microbiota. By comparing genus and species level-biomass contributions, selective antibacterial-like activities were found with 35 of the 39 non-antibiotic compounds. Seven of the compounds led to a global alteration of the metaproteome, with apparent compound-specific patterns of functional responses. The taxonomic distributions of altered proteins varied among drugs, i.e., different drugs affect functions of different members of the microbiome. We also showed that bacterial function can shift in response to drugs without a change in the abundance of the bacteria. Conclusions: Current drug-microbiome interaction studies largely focus on relative microbiome composition and microbial drug metabolism. In contrast, our workflow enables multiple insights into microbiome absolute abundance and functional responses to drugs. The workflow is robust, reproducible, and quantitative and is scalable for personalized high-throughput drug screening applications.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s40168-020-00806-z

Cites

78 of 78 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Rapid Assay of Individual Microbiome (RapidAIM) workflow and performance. a Experimental, analytical, and bioinformatics components of the RapidAIM workflow. Each individual’s gut microbiome samples are cultured with the test compounds in a 96-well deep-well plate at 37 °C in strict anaerobic conditions for 24 h followed by high-throughput sample preparation and rapid LC-MS/MS analysis. Peptide and protein identification and quantification, taxonomic profiling, and functional annotation were performed using the automated MetaLab software [27]. b A series of six dilutions (dilution gradients: GRD1~6) of a same microbiome sample was tested in triplicates through the equal-volume digestion and equal-volume MS loading protocol; the summed peptide intensity was compared with a set of protein concentration standards provided with DC protein concentration assay and showed good linearity (center points and error bars represent mean ± SD). c Stacked bars of clusters of orthologous groups (COG) category levels across the six concentrations showing no bias at the functional quantifications. d PCA based on LFQ intensities of protein groups for all POC samples
Fig. 1 Rapid Assay of Individual Microbiome (RapidAIM) workflow and performance. a Experimental, analytical, and bioinformatics components of the RapidAIM workflow. Each individual’s gut microbiome samples are cultured with the test compounds in a 96-well deep-well plate at 37 °C in strict anaerobic conditions for 24 h followed by high-throughput sample preparation and rapid LC-MS/MS analysis. Peptide and protein identification and quantification, taxonomic profiling, and functional annotation were performed using the automated MetaLab software [27]. b A series of six dilutions (dilution gradients: GRD1~6) of a same microbiome sample was tested in triplicates through the equal-volume digestion and equal-volume MS loading protocol; the summed peptide intensity was compared with a set of protein concentration standards provided with DC protein concentration assay and showed good linearity (center points and error bars represent mean ± SD). c Stacked bars of clusters of orthologous groups (COG) category levels across the six concentrations showing no bias at the functional quantifications. d PCA based on LFQ intensities of protein groups for all POC samplesCC BY · PubMed Central
Fig. 2 Response of microbiome abundance and composition to compounds. a Biomass responses of individuals’ microbiomes to compounds relative to DMSO control. Ratio of peptide intensity between compound and DMSO control samples was calculated for each individual microbiome. b Log2 fold-change of absolute abundance at the genus level in response to each drug compared with the DMSO control. Genera that existed in ≥ 80% of the volunteers are shown. Asterisk indicates significantly changed bacterial abundance by Wilcoxon test, p < 0.05. c Bray-Curtis distance of genus-level composition between drug-treated microbiomes and the corresponding DMSO control samples. Heatmap colors are generated with average of log2 fold-changes among the five individual samples. Statistical significance was calculated by pairwise Wilcoxon test (FDR-adjusted p < 0.05). Box spans interquartile range (25th to 75th percentile), and line within box denotes median. For full compound names, see abbreviation list in Supplementary Table S1
Fig. 2 Response of microbiome abundance and composition to compounds. a Biomass responses of individuals’ microbiomes to compounds relative to DMSO control. Ratio of peptide intensity between compound and DMSO control samples was calculated for each individual microbiome. b Log2 fold-change of absolute abundance at the genus level in response to each drug compared with the DMSO control. Genera that existed in ≥ 80% of the volunteers are shown. Asterisk indicates significantly changed bacterial abundance by Wilcoxon test, p < 0.05. c Bray-Curtis distance of genus-level composition between drug-treated microbiomes and the corresponding DMSO control samples. Heatmap colors are generated with average of log2 fold-changes among the five individual samples. Statistical significance was calculated by pairwise Wilcoxon test (FDR-adjusted p < 0.05). Box spans interquartile range (25th to 75th percentile), and line within box denotes median. For full compound names, see abbreviation list in Supplementary Table S1CC BY · PubMed Central
Fig. 3 Effect of compounds on metaproteomic profiles of the microbiome. a Bray-Curtis distance of protein groups between drug-treated microbiomes and the corresponding DMSO control samples. Statistical significance was calculated by pairwise Wilcoxon test (FDR-adjusted p < 0.05). b Unsupervised dimensionality reduction analysis showing compound-specific patterns of responses to class I compounds. c Enrichment analysis of all significantly different COGs in the POC dataset. Significantly altered COGs with a p value cutoff of 0.05 are shown
Fig. 3 Effect of compounds on metaproteomic profiles of the microbiome. a Bray-Curtis distance of protein groups between drug-treated microbiomes and the corresponding DMSO control samples. Statistical significance was calculated by pairwise Wilcoxon test (FDR-adjusted p < 0.05). b Unsupervised dimensionality reduction analysis showing compound-specific patterns of responses to class I compounds. c Enrichment analysis of all significantly different COGs in the POC dataset. Significantly altered COGs with a p value cutoff of 0.05 are shownCC BY · PubMed Central
Fig. 4 Response of specific proteins and enzymatic pathways to drug treatment. a Response of proteins related to drug resistance (red asterisk denotes p < 0.05 by Wilcoxon test); boxplots of all functional responses of the RapidAIM POC dataset are available at https://shiny.imetalab.ca/RapidAIM_functional_response/. b Effect of FOS treatment on enzymes involved in fructose and mannose metabolism, GABA production, and sulfide metabolism pathways. c Effect of ciprofloxacin treatment on enzymes involved in the glycolysis/gluconeogenesis and pentose phosphate pathway. The five blocks of each enzyme represent the five individual microbiomes. Colors in the blocks represent differences between normalized KEGG enzyme intensities with drug treatment versus DMSO (log2 transformation of the original intensity followed by a quotient normalization (x/mean))
Fig. 4 Response of specific proteins and enzymatic pathways to drug treatment. a Response of proteins related to drug resistance (red asterisk denotes p < 0.05 by Wilcoxon test); boxplots of all functional responses of the RapidAIM POC dataset are available at https://shiny.imetalab.ca/RapidAIM_functional_response/. b Effect of FOS treatment on enzymes involved in fructose and mannose metabolism, GABA production, and sulfide metabolism pathways. c Effect of ciprofloxacin treatment on enzymes involved in the glycolysis/gluconeogenesis and pentose phosphate pathway. The five blocks of each enzyme represent the five individual microbiomes. Colors in the blocks represent differences between normalized KEGG enzyme intensities with drug treatment versus DMSO (log2 transformation of the original intensity followed by a quotient normalization (x/mean))CC BY · PubMed Central
Fig. 5 Global functional effects of berberine, ciprofloxacin, FOS and diclofenac. a Taxon-function distribution of protein groups responding to berberine, FOS, ciprofloxacin, and diclofenac. Responding protein groups were selected by PLS-DA based on ComBat-corrected data. The semicircle diameter represents the number of PLS-DA VIP > 1 protein groups corresponding to each phyla-COG category pair. b Genus level shifts in functional activities in response to berberine and the alterations in biomass of the corresponding genera. Functional shifts (differential protein groups) were identified by PLS-DA. For each genus, the percentages of the total numbers of up- and downregulated protein groups corresponding to each COG category are shown. Shifts in bacterial biomass in the five microbiomes are shown in box plots with the boxes spanning interquartile range (25th to 75th percentile) and the vertical lines denoting the median for each genus
Fig. 5 Global functional effects of berberine, ciprofloxacin, FOS and diclofenac. a Taxon-function distribution of protein groups responding to berberine, FOS, ciprofloxacin, and diclofenac. Responding protein groups were selected by PLS-DA based on ComBat-corrected data. The semicircle diameter represents the number of PLS-DA VIP > 1 protein groups corresponding to each phyla-COG category pair. b Genus level shifts in functional activities in response to berberine and the alterations in biomass of the corresponding genera. Functional shifts (differential protein groups) were identified by PLS-DA. For each genus, the percentages of the total numbers of up- and downregulated protein groups corresponding to each COG category are shown. Shifts in bacterial biomass in the five microbiomes are shown in box plots with the boxes spanning interquartile range (25th to 75th percentile) and the vertical lines denoting the median for each genusCC BY · PubMed Central
Fig. 6 Individual functional responses to indomethacin. a When visualizing the responses of several individual microbiomes with PCA (based on LFQ intensities of protein groups), inter-individual variability can be greater than drug-induced functional shifts. b PCA clearly differentiated the response of microbiome V1 treated in triplicates using RapidAIM. c Three hundred three significant protein group responders were found by t test (FDR-adjusted p < 0.05). d Taxon enrichment analysis based on the differential protein groups, (p-adjusted = 0.05). e Taxon-function coupled enrichment analysis of upregulated protein groups
Fig. 6 Individual functional responses to indomethacin. a When visualizing the responses of several individual microbiomes with PCA (based on LFQ intensities of protein groups), inter-individual variability can be greater than drug-induced functional shifts. b PCA clearly differentiated the response of microbiome V1 treated in triplicates using RapidAIM. c Three hundred three significant protein group responders were found by t test (FDR-adjusted p < 0.05). d Taxon enrichment analysis based on the differential protein groups, (p-adjusted = 0.05). e Taxon-function coupled enrichment analysis of upregulated protein groupsCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 1

  1. Maier L, Pruteanu M, Kuhn M, Zeller G, Telzerow A, Anderson EE, Brochado AR, Fernandez KC, Dose H, Mori H, et al. Extensive impact of non-antibiotic drugs on human gut bacteria. Nature. 2018;555:623. · DOI · Paper page

Cited by

Sorted by citations

See all 330

Related

See all

Metadata

Authors
Li, L., Ning, Z., Zhang, X., Mayne, J., Cheng, K., Stintzi, A., & Figeys, D.
Affiliations
  • University of Ottawa (CA)
  • Canadian Institute for Advanced Research (CA)
Corresponding author
  • Stintzi, Alain
  • Figeys, Daniel
ORCID
Venue
Microbiome
Publisher
Springer Science and Business Media LLC
Published
2020-03-11
Volume
8
Issue
1
Pages
33-33
ISSN
2049-2618 (ISSN-L)
DOI
10.1186/s40168-020-00806-z
PMC
PMC7066843
OpenAlex
W3010746544
Cited by
328
References
78
Citations by year
2026: 28 · 2025: 60 · 2024: 60 · 2023: 60 · 2022: 53 · 2021: 49 · 2020: 14
FWCI
14.28
Type
journal article
Language
en
Open access
Licence
Locations
  • Microbiome (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • UEA Digital Repository (University of East Anglia) (acceptedVersion)
  • uO Research (University of Ottawa) (submittedVersion, public-domain)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Genome Canada (OGI-114 and OGI-149)
  • Canadian Institutes of Health Research (ECD-144627)
  • Natural Sciences and Engineering Research Council of Canada (210034)
  • Ontario Ministry of Economic Development and Innovation (REG1-4450)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gut microbiota and health
  • Physical Sciences › Environmental Science › Ecology › Microbial Community Ecology and Physiology
  • Health Sciences › Dentistry › Periodontics › Oral microbiology and periodontitis research
Keywords
gut microbiota, drug-microbiome interactions, metaproteomics, antibiotic effects on microbiota, selective antibacterial activity, bacterial biomass, berberine, ibuprofen, drug repurposing, personalized drug screening
MeSH terms
Proof of Concept Study, Adult, Anti-Bacterial Agents, Bacteria, Nonprescription Drugs, Feces, Female, Humans, Male, RNA, Ribosomal, 16S, Colony Count, Microbial, Proteomics, Prescription Drugs, Microbiota

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review