key: cord-0703339-25erx6e6
authors: Cervino, A. C. L.; Fabre, R.; Plassais, J.; Gbikpi-Benissan, G.; Petat, E.; Le Quellenec, E.; Neuberger-Castillo, L.; Laurent, J.-M.; Iordache, L.; Bouchahda, M.; Marti, G.; Chapelet, G.
title: Results from EDIFICE : A French pilot study on COVID-19 and the gut microbiome in a hospital environment
date: 2022-02-08
journal: nan
DOI: 10.1101/2022.02.06.22269945
sha: f9eddd48f30781af873856379aabe3698657ecc9
doc_id: 703339
cord_uid: 25erx6e6

BACKGROUND & AIMS: Early reports suggest that both fecal shedding and dysbiosis of the gut microbiome are associated to disease severity in COVID-19 patients. We investigated the gut microbiome as well as the prevalence of SARS-CoV-2 in stool samples from two French populations: exposed healthcare workers and elderly hospitalized COVID-19 patients. The predictive power of bacterial loss of diversity and detection of SARS-CoV-2 in stool was assessed at 4 weeks against clinical outcomes in the patient group. METHODS: 79 healthcare workers in contact with COVID-19 patients and 64 elderly patients hospitalised in a COVID-19 unit in France were included in the EDIFICE trial from April 2020 until May 2021. Stool samples were collected at inclusion. Loss of bacterial diversity was diagnosed based on 16S rRNA gene sequencing. Stool positivity to SARS-CoV-2 was determined by RT-PCR. Clinical outcomes were recorded at a 4 weeks follow up visit. In particular, these include whether the patient had been put under oxygen during the 4 weeks follow up, whether he had been discharged with or without aggravation from initial symptoms or whether the patient had died. The primary end point was to validate the hypothesis that hospitalized COVID-19 patients had more often lost their bacterial diversity than highly exposed active healthcare workers. RESULTS: Elderly hospitalised patients with COVID-19 had more frequently lost their bacterial diversity when compared to exposed healthcare workers (p-value = 0.005), their severe dysbiosis was characterized by enrichment of the family Erysipelotrichaceae and depletion of beneficial bacteria at the genus level such as butyrate producers (Butyrivibrio, Roseburia, Faecalibacterium) and Bifidobacterium. The virus was detected in 61% of hospitalized patients and in only one healthcare workers (2%) who had previously been diagnosed with COVID-19 (p-value<0.001). No significant difference in the gut microbiome composition at the genus level of patients that tested positive in stool versus patients that tested negative was observed. Neither bacterial loss of diversity nor positivity to SARS-CoV-2 were associated to clinical outcome at 4 weeks. CONCLUSIONS: We report findings of the first French trial investigating the clinical interest of stool based diagnosis of SARS-CoV-2 and loss of bacterial diversity in a population of elderly hospitalised COVID-19 patients and highly exposed healthcare workers. Our findings of reduced bacterial diversity and a strong gut dysbiosis in elderly hospitalized COVID-19 patients are highly consistent with previous reports mostly from Chinese populations. A major limitation is that observed differences in the gut microbiome between the two studied groups cannot be attributed to COVID-19 per se given the large number of confounding factors. SARS-CoV-2 was detected in the stool of the majority of hospitalized patients even several weeks after initial diagnosis by nasopharyngeal swabs. This high prevalence warrants further investigation by the scientific community into mechanism.

The ongoing Coronavirus Disease 2019 pandemic, caused by the SARS-CoV-2 virus, is a worldwide pandemic. As of December 1 st , 2021, more than 260 million confirmed cases and 5 million deaths have been reported (World Health Organization). In France, more than 7 million cases have been confirmed so far, resulting in over 110,000 deaths.

Fortunately, the SARS-CoV-2 and its casualties have been partially controlled by implementation of nationwide sanitary measures, the widespread testing strategy and the fast development of vaccines. Interindividual contamination is thought to take place through inhalation of infected droplets or physical contact leading to the virus entering through the nose or the mouth. Patient diagnosis is mostly based on nasopharyngeal swabs. Fast and accurate diagnosis of the SARS-CoV-2 virus continues to take a central place in the pandemic control strategy and public health decision-making, with RT-PCR or antigenic tests but also with whole genome sequencing approach to understand transmission dynamics and follow new variants (Oude Munnink et al. 2020 ).

Main symptoms in hospitalized patients with mild to severe disease include cough, fever, f atigue and shortness of breath. Other symptoms are musculoskeletal, anosmia/dysgeusia and gastro-intestinal (GI) symptoms (Docherty et al. 2020 )(D. . Diarrhea is the most common GI symptom, reported in 2% to 50% of cases (Megyeri et al. 2021 ) mostly in association with respiratory symptoms (Docherty et al. 2020) . Other GI symptoms include nausea, vomiting or abdominal pain in 2% to 10% of cases (Wong, Lui, and Sung 2020; D. Wang et al. 2020 ).

SARS-CoV-2 has been detected by RT-PCR in stool samples of COVID-19 patients (D. Xu et al. 2020; Wu et al. 2020; Cheung et al. 2020; Vaselli et al. 2021; Y. Zhou et al. 2021; Zuo et al. 2020; Britton et al. 2021) . RT-PCR tests on faecal samples can remain positive about 50% longer than tests from nasopharyngal swabs after first symptom onset, meaning that the virus can be detected in stool samples even when it is no longer detectable by nasopharyngal swabs ). On average stool-based detection can be positive for as long as 28 days raising the question of whether stool can be a source of continued contamination. (Wölfel et al. 2020 ) did however show that the virus is not infectious through that route despite elevated viral loads in a study on 9 patients. Other reports though indicate isolating the infectious virus from a stool sample (Yong Zhang et al. 2020; Xiao et al. 2020) . Currently the infectious potential of virus titre found in faeces is still unclear (Guo et al. 2021 ).

Most research since 2019 has focused on the respiratory tract and to a much lesser extent has the gut been investigated. (Zuo et al. 2020 ) published the first cohort study investigating alterations in the gut microbiome and viral activity in stool. Since then, several other publications reported results on the differences between the gut bacterial microbiome in COVID-19 patients compared to controls Chen et al. 2022; Mazzarelli et al. 2021; Yeoh et al. 2021; Reinold et al. 2021; Newsome et al. 2021; Tao et al. 2020; Li et al. 2021 ). Most reported a decreased bacterial diversity in the gut of COVID-19 patients when compared to controls. The observed dysbiosis was linked to enrichment of opportunistic pathogens and decrease of beneficial bacteria such as Butyrate producers and Bifidobacteria.

Findings also support a possible association with disease severity Zuo et al. 2020; Moreira-Rosário et al. 2021) . The majority of studies were prospective cohorts from patients of Asian ethnicity, and the microbiome analysis was performed in a research environment. Here we report for the first time, the results from a prospective, observational, multicentric, controlled clinical trial (EDIFICE) using a commercial IVD test (1test1 TM ) in a French population. It is also the first report on two distinct populations: highly exposed healthcare workers and elderly hospitalised patients.

Est-III, under number 2020-A00979-30 on April 21 st , 2020. All participants consented to the study. Study participants were recruited between April 28 th , 2020, and April 29 th , 2021, in one of the following four clinical centers: Clinique Saint-Jean L'hermitage (Melun, France), Clinique du Mousseau (Evry, France), Hôpital de Fontainebleau (Fontainebleau, France) and

Centre Hospitalier Universitaire (CHU) de Nantes (Nantes, France). The study included two groups : a first group of healthcare workers working in a COVID-19 unit and representing an active population highly exposed to the virus through direct contact with COVID19 patients, and a patient group of elderly hospitalised patients treated in a COVID-19 unit. Study participants had to be between 18 and 95 years old and able to provide a stool sample non interventionally. Patients in critical care were therefore excluded. Study participants were all given at inclusion a commercial IVD home collection kit (1test1 TM , Luxia Scientific, France).

Stool samples were thus collected using the DNA/RNA Shield™ Fecal Collection Tubes (Zymo Research, Irvine, California) . Each sample contained approximately 1 gram of sample in 9 ml of stabilizer, representing a 1:10 dilution.

Samples for microbiome analysis were processed at the Integrated Biobank of Luxembourg (IBBL). All samples were processed within one month of stool collection following a validated stool DNA extraction protocol on the Chemagen MSM I instrument (PerkinElmer Chemagen, Baesweiler, Germany) (Mathay et al. 2015; Neuberger-Castillo et al. 2020 ) using the Chemagic™ DNA Blood 4k Kit special H24 with special lysis buffer for fecal samples.

In brief, each sample of 1 ml was lysed by adding 2ml of SEB lysis buffer and 30µl

Proteinase K, homogenized and incubated for 10min at 70°C, followed by 5min at 95°C.

Lysates (1.5mL) were centrifuged for 5min at 10,000 g at RT. 1ml supernatants were transferred into a 24XL deep-well plate, which were processed using the magnetic beads based Chemagic™ MSM I automated protocol. Each run included bacterial DNA extraction controls (Bacterial Pool QC) in duplicates to check for the DNA extraction efficiency and possible contamination.

Total DNA quantification and purity (A260/A280 ratio) by spectrophotometry was performed using a Synergy Mx Monochromator-Based Multi-Mode microplate reader (Biotek) with Gen 5 version 2.0 software. Double-stranded DNA quantification by spectrofluorometry was performed with Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen) using 96-well plates on the same device.

The microbial composition profiling was performed by sequencing the V3-V4 regions of the prokaryotic 16S rRNA gene on the Illumina MiSeq using 2x300 bp paired-end reads.

Amplicons were generated, cleaned, indexed and sequenced according to the Illuminademonstrated 16S Metagenomic Sequencing Library Preparation Protocol with certain modifications (Illumina). In brief, an initial PCR reaction contained at least 12.5 ng of DNA.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.06.22269945 doi: medRxiv preprint A subsequent limited-cycle amplification step was performed to add multiplexing indices and Illumina sequencing adapters. Libraries were normalized and pooled, and sequenced on the Illumina MiSeq instrument.

Each library preparation run included an internal bacterial DNA extraction control (Bacterial Pool QC) and an internal 16S mock bacterial community control (DNA QC 16S) to detect any contamination during the sequencing library preparation and/or artefacts arising from the bioinformatics analysis. DNase/RNase-free water was used as negative control to detect any contamination. QC checks on fragment size, cluster density, error rate and Q30 were performed systematically all throughout the processing. Raw sequence data generated for this study are available in the Sequence Read Archive under BioProject accession PRJNA787810. 

Fastq files were analysed with 1TEST1SOFT, an internal bioinformatics pipeline that is based on Qiime2 (Bolyen et al. 2019) (2018.6) and RDP (Cole et al. 2014 ) (trainset n°16) and that has been extensively validated as part of the IVD status of 1test1 TM . The pipeline first removes expected 5' and 3' primers, CCTACGGGNGGCWGCAG and GACTACHVGGGTATCTAATCC, respectively, from the paired-end reads in the Fastq files, using the Cutadapt tool. Joined sequences are then quality-filtered using the Quality-filter plugin of Qiime2 with default parameters. A denoising step follows using the Denoise-16S method of the Deblur plugin with a trimming option set to 400bp, which produces amplicon sequence variants (ASVs) from which diversity is estimated. Taxonomic quantification of the denoised samples is finally obtained through the Classify-sklearn method of the Feature-classifier plugin, using the default confidence value (0.7) and a classifier model trained on the V3-V4 region (extracted using the above primers) of the reference sequences from the RDP database. Alpha-diversity has been considered through the Shannon index calculated in the Core-metrics-phylogenetic pipeline of the Diversity plugin of Qiime2 with a cut-off value of 5.57 for diagnosing the low bacterial diversity at a rarefaction of 9 000 reads.

Statistical analysis was performed internally using ("R: The R Project for Statistical Computing" n.d 

The cutoff of 5.57 for the Shannon index was previously established based on the first quartile of alpha-diversity in a healthy group of 29 west European residents. The cutoff was further validated in a real life cohort of 100 participants who self reported their health status as one of five categories ("Excellent", "Very good", "Good", "Not so good", " Poor") . The frequency in the Excellent and Very good groups was 25% versus 87% in the self reported poor health group ( Figure 1 ). The technical repeatability of the diversity measure was estimated to be 10%.

In this trial we enrolled 143 participants: 64 elderly hospitalized COVID-19 patients and 79

healthcare workers working in close contact with COVID-19 patients. Amongst the patient group 62 had the follow up visit at 4 weeks. 55 patients had their microbiome successfully analysed by 16S rRNA gene sequencing and 61 were successfully analysed by RT-PCR for SARS-CoV-2 detection. 52 patients had results from both analyses (microbiome and viral detection). 24 of the 79 exposed staff self-reported having had COVID-19 like symptoms in the past but as those had not been confirmed by viral PCR tests, this variable was not analysed. None reported having had severe symptoms needing hospital care.

The two groups were unmatched and differed for most variables ( compared to a minority, respectively 9% and 5%, amongst the healthcare workers.

In the COVID-19 population, the three most commonly reported symptoms at diagnosis were tiredness (86%), fever (54%) and cough (52%). Diarrhea was reported by 15% of patients.

Loss of bacterial diversity was assessed in 76 healthcare workers and 55 elderly COVID-19

patients. 41% of the healthcare workers was diagnosed with a loss of bacterial diversity (Table 2a) , which is slightly higher than the standard of 25% from the healthy reference European cohort used for calibration of 1test1 TM (Figure 1 ). Amongst the elderly COVID-19

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022.

patients (Table 2a) 65% had lost their bacterial diversity which is an expected significant increase compared to the healthcare workers (p= 0.005, Chi-squared test).

Amongst the healthcare workers only one person tested positive in stool, representing 2% of that population (Table 2b) . Amongst the elderly COVID-19 hospitalised patients the majority tested positive (61%) which is an expected significant increase compared to the healthcare workers (p<0.001, Chi-squared test). Samples tested positive even 30 days after initial diagnosis (Supplementary Figure 1) .

We also tested whether loss of bacterial diversity or presence of SARS-CoV-2 in stool was associated with the patient status 4 weeks after the stool collection. At 4 weeks, 34% of hospitalized SARS-CoV-2 patients had left the hospital without sequels, 53% had left the hospital with sequels, and 13% died (Table 1 ). Sequels at 4 weeks included loss of autonomy, undernutrition, oxygen supply and asthenia. None of those conditions were significantly associated with either loss of bacterial diversity (p-value=1 for death and p-value= 0.375 for without sequels) nor viral positivity in stool (p-value=0.138 for death and p-value= 0.404 for without sequels). 7 out of the 8 patients who died were positive for SARS-CoV-2 in stool (87.5%) however this was not statistically significant but should be confirmed in a larger cohort.

Major differences were observed in the taxonomical composition between the hospitalised patients group and the healthcare workers. A total of 33 out of 56 predominant genera were found to be significantly altered in the gut microbiome (Table 3) On the other hand, no significant difference was observed in any of the genera when comparing the microbiome of patients that were SARS-CoV-2 positive (n=29) in stool to those with a SARS-CoV-2 negative result (n=23). Supplementary Table 1. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.06.22269945 doi: medRxiv preprint

We report for the first time on the clinical performance of a diagnostic test of loss of bacterial diversity, gut microbiome bacterial composition and stool positivity to SARS-CoV-2 in two distinct populations : a cohort of elderly hospitalized COVID-19 patients and a highly exposed active healthcare workers population.

Using the stool based self-collection kit 1test1 TM (IVD, Luxia Scientific), participants were classified as either having a normal diversity or having a reduced diversity (also termed loss of diversity) based on a predefined cutoff value of the Shannon alpha diversity index. The initial motivation for a binary classification as "normal" and "lost diversity" in the general population is to identify people who would the most benefit from a dietary and life style change to improve their gut diversity. Using this binary classification of diversity, we found that a large proportion (65%) of elderly hospitalized COVID-19 patients had lost their bacterial diversity and that this value was significantly higher than our healthy reference cohort (25%) as well as the population of exposed healthcare workers (41%). (Badal et al. 2020) . Host genetics also impact gut microbial composition (Kurilshikov et al. 2021 ). The significant decrease in alpha diversity observed in elderly COVID-19 patients with significant co-morbidities was therefore expected and cannot, here, be attributed to COVID-19 status given the unmatched design of the trial. Previous reports indicate, however, that the decrease in alpha diversity is unlikely to be attributed to age or to antibiotics alone given that similar conclusions were reached in trials with matched designs or excluding previous antibiotics use Reinold et al. 2021; Li et al. 2021 ) and that alpha diversity correlates to disease severity . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. COVID19 patients and showed that this reduction is associated with increased plasma levels of inflammatory proteins. The observed decrease in SCFA and L-isoleucine biosynthesis in gut microbiome persisted even after recovery in patients. Bifidobacterium are also beneficial bacteria that are significantly decreased in COVID19 patients consistent with reports from others Zuo et al. 2020; Yeoh et al. 2021; Reinold et al. 2021; Moreira-Rosário et al. 2021) . Bifidobacterium have documented health benefits and many strains are used as probiotics.

Genera that are the most significantly increased in COVID19 patients, include Erysipelotrichaceae_Other, Eisenbergiella, Escherichia and Parabacteroides. Those genera have also been reported by others Tang et al. 2020; Zuo et al. 2020; Mazzarelli et al. 2021; Yeoh et al. 2021; Reinold et al. 2021; Newsome et al. 2021; Li et al. 2021; Tao et al. 2020) . The family Erysipelotrichaceae appears to be highly immunogenic, can potentially flourish post-treatment with broad spectrum antibiotics, and has been reported to be increased in colorectal cancer and metabolic disorders (Kaakoush 2015) . This taxon does not in fact correspond to a genera, but rather a family level classification that could not be assigned at the genus level. Indeed, 16S based analysis does not allow accurate species level classification. Two other reports that used a shotgun metagenomic approach also reported Erysipelotrichaceae_noname as enriched in COVID-19 patients or associated with disease severity (Zuo et al. 2020) . It could thus be interesting to perform de . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 novo assembly of this bacteria in COVID19 patients to better characterize it. The genus Eisenbergiella has been reported to be correlated to cognitive impairment (Stadlbauer et al. 2020 ). Pro-inflammatory bacteria such as Escherichia are well known pathogens and are often reported in patients with diarrhea and in IBD cases.

The mechanism by which the virus enters the human host is becoming increasingly clear (Jackson et al. 2022) . SARS-CoV-2 enters the host cells through binding top angiotensinconverting enzyme 2 (ACE2) receptors (P. Zhou et al. 2020; Q. Wang et al. 2020, 2) and the serine protease TMPRSS2 for S protein priming (Hoffmann et al. 2020) . The ACE2 receptor is particularly highly expressed in colonic enterocytes (H. ) (Y. Wang et al. 2020, 2) . In mice, mutations in Ace2 leads to decreased expression of antimicrobial peptides and gut microbiome dysbiosis (Hashimoto et al. 2012) . ACE2 is linked to amino acid transport in the gut (Cole-Jeffrey et al. 2015) and to tryptophan metabolism which is altered in COVID-19 patients (Gardinassi et al. 2020 ).

One of the main findings from our study is the large prevalence of SARS-CoV-2 observed in stool samples from hospitalized patients several weeks after initial infection (overall 61% reaching 87.5% in the subgroup of patients who died). For the exposed healthcare workers, the virus was only detected in one person that had been tested positive by nasopharyngal swab 35 days prior to sample collection and had not had any severe symptoms. Viral detection in fecal samples in 61% of COVID-19 cases is consistent with other reports that vary from 13% to 73% (Zuo) with an estimated average of 40.3% from a metaanalysis conducted in 2020 (Parasa et al. 2020 ) and of 43% in 2021 (Yawen Zhang et al. 2021 ). SARS-CoV-2 is detectable in stool for a longer period than in nasophyringeal swabs, ) estimate that on average samples stay positive for 27.9 days after the first symptoms later confirmed by a meta-analysis indicating 21.8 days (Yawen Zhang et al. 2021) . Given the high rates of positivity in fecal samples, understanding the causes and consequences of SARS-CoV-2 presence in the digestive track are important issues that need to be addressed. Interestingly, persistence of SARS-CoV-2 nucleic acids detected in intestinal biopsies from subgroups of COVID19 patients 4 months post infection favorably impacts memory B cell response (Gaebler et al. 2021) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.06.22269945 doi: medRxiv preprint As pointed out by (Venzon and Cadwell 2022) "an important research direction would be to determine whether the gut microbiota dysbiosis observed in patients with severe COVID-19 affects the progression and recovery from the disease". Here we investigated clinical outcome by looking at disease progression at 4 weeks, neither loss of diversity nor fecal viral RNA were associated with whether the patient had been put under oxygen during the 4 weeks follow up, whether he had been discharged with or without aggravation from initial symptoms or whether the patient had died. 7 out of the 8 COVID19 patients who died tested positive, although these results lacked statistical power, they are consistent with a previous report of association between death and fecal viral RNA (Britton et al. 2021) . Analysis of clinical progression is important to assess the diagnostic value of biomarkers in a clinical setting. Although the four clinical outcomes we investigated here were not associated with loss of diversity or viral positivity, other clinical outcomes might be.

The EDIFICE trial suffers from various limitations, additionally to the lack of matching between groups, such as the lack of systematic testing of healthcare workers due to the noninterventional nature of the trial and it would have been interesting to systematically perform a follow up fecal RT-PCR test at 4 weeks in both groups. Additional exploration of inflammatory and antibody markers in blood and intestinal biopsies would also have been of great interest.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 T A B L E S : ( . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.06.22269945 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.06.22269945 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted February 8, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted February 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 

The Gut Microbiome, Aging, and Longevity: A Systematic Review

Reproducible, Interactive, Scalable and Extensible Microbiome Data Science Using QIIME 2

Limited Intestinal Inflammation despite Diarrhea, Fecal Viral RNA and SARS-CoV-2-Specific IgA in Patients with Acute COVID-19

Six-Month Follow-up of Gut Microbiota Richness in Patients with COVID-19

Gastrointestinal Manifestations of SARS-CoV-2 Infection and Virus Load in Fecal Samples From a Hong Kong Cohort: Systematic Review and Meta-Analysis

Ribosomal Database Project: Data and Tools for High Throughput RRNA Analysis

ACE2 and Microbiota: Emerging Targets for Cardiopulmonary Disease Therapy

Features of 20 133 UK Patients in Hospital with Covid-19 Using the ISARIC WHO Clinical Characterisation Protocol: Prospective Observational Cohort Study

Impact of Diet in Shaping Gut Microbiota Revealed by a Comparative Study in Children from Europe and Rural Africa

Evolution of Antibody Immunity to SARS-CoV-2

Immune and Metabolic Signatures of COVID-19 Revealed by Transcriptomics Data Reuse

Alterations of the Gut Microbiota in Patients With Coronavirus Disease 2019 or H1N1 Influenza

Potential Intestinal Infection and Faecal-Oral Transmission of SARS-CoV-2

ACE2 Links Amino Acid Malnutrition to Microbial Ecology and Intestinal Inflammation

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Mechanisms of SARS-CoV-2 Entry into Cells

Insights into the Role of Erysipelotrichaceae in the Human Host

Large-Scale Association Analyses Identify Host Factors Influencing Human Gut Microbiome Composition

Richness of Human Gut Microbiome Correlates with Metabolic Markers

Butyrate: Implications for Intestinal Function

Microbiome Profiling Using Shotgun Metagenomic Sequencing Identified Unique Microorganisms in COVID-19 Patients With Altered Gut Microbiota

Method Optimization for Fecal Sample Collection and Fecal DNA Extraction

16S RRNA Gene Sequencing of Rectal Swab in Patients Affected by COVID-19

COVID-19-Associated Diarrhea

Gut Microbiota Diversity and C-Reactive Protein Are Predictors of Disease Severity in COVID-19 Patients

Recovered Patients Returns to Uninfected Status in a Minority-Dominated United States Cohort

Rapid SARS-CoV-2 Whole-Genome Sequencing and Analysis for Informed Public Health Decision-Making in the Netherlands

Prevalence of Gastrointestinal Symptoms and Fecal Viral Shedding in Patients With Coronavirus Disease 2019

A Microbial Signature for Crohn's Disease

R: The R Project for Statistical Computing

A Pro-Inflammatory Gut Microbiome Characterizes SARS-CoV-2 Infected Patients and a Reduction in the Connectivity of an Anti-Inflammatory Bacterial Network Associates With Severe COVID-19

RStudio | Open Source & Professional Software for Data Science Teams

Dysbiosis, Gut Barrier Dysfunction and Inflammation in Dementia: A Pilot Study

Clinical Significance of the Correlation between Changes in the Major Intestinal Bacteria Species and COVID-19 Severity

Analysis of the Intestinal Microbiota in COVID-19 Patients and Its Correlation with the Inflammatory Factor IL-18

Investigation of SARS-CoV-2 Faecal Shedding in the Community: A Prospective Household Cohort Study (COVID-LIV) in the UK

COVID-19 and the Forgotten Organ: Prolonged Changes to the Metabolic Output of the Gut Microbiome

Impact of Commonly Used Drugs on the Composition and Metabolic Function of the Gut Microbiota

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Structural and Functional Basis of SARS-CoV-2 Entry by Using Human ACE2

A Comprehensive Investigation of the MRNA and Protein Level of ACE2, the Putative Receptor of SARS-CoV-2, in Human Tissues and Blood Cells

Virological Assessment of Hospitalized Patients with COVID-2019

Covid-19 and the Digestive System

Prolonged Presence of SARS-CoV-2 Viral RNA in Faecal Samples

Infectious SARS-CoV-2 in Feces of Patient with Severe COVID-19

Gut Microbiota Composition Reflects Disease Severity and Dysfunctional Immune Responses in Patients with COVID-19

Prolonged Impairment of Short-Chain Fatty Acid and L-Isoleucine Biosynthesis in Gut Microbiome in Patients With COVID-19

Specific ACE2 Expression in Small Intestinal Enterocytes May Cause Gastrointestinal Symptoms and Injury after 2019-NCoV Infection

Prevalence and Persistent Shedding of Fecal SARS-CoV-2 RNA in Patients With COVID-19 Infection: A Systematic Review and Meta-Analysis

Isolation of 2019-NCoV from a Stool Specimen of a Laboratory-Confirmed Case of the Coronavirus Disease 2019 (COVID-19)

A Pneumonia Outbreak Associated with a New Coronavirus of Probable Bat Origin

Gut Microbiota Dysbiosis Correlates with Abnormal Immune Response in Moderate COVID-19 Patients with Fever

Alterations in Gut Microbiota of Patients With COVID-19 During Time of Hospitalization