Infection

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Lost microbes of COVID-­19:
Bifidobacterium, Faecalibacterium
depletion and decreased microbiome
diversity associated with SARS-­CoV-­2
infection severity
Sabine Hazan,1 Neil Stollman,2 Huseyin S Bozkurt,3 Sonya Dave ‍ ‍,4,5
Andreas J Papoutsis,1 Jordan Daniels,1 Brad D Barrows,1
Eamonn MM Quigley ‍ ‍,6 Thomas J Borody7

To cite: Hazan S, Stollman N, ABSTRACT

Bozkurt HS, et al. Lost microbes Objective The study objective was to compare gut
Summary box
of COVID-­19: Bifidobacterium, microbiome diversity and composition in SARS-­CoV-­2
Faecalibacterium depletion What is already known about this subject?
PCR-­positive patients whose symptoms ranged from
and decreased microbiome ► The gut microbiome is intrinsically related to host
asymptomatic to severe versus PCR-­negative exposed
diversity associated with immune response (eg, inflammation, Th1 vs Th2)
SARS-C ­ oV-­2 infection controls.
and susceptibility to infection.
severity. BMJ Open Gastro Design Using a cross-­sectional design, we performed
2022;9:e000871. doi:10.1136/ shotgun next-­generation sequencing on stool samples What are the new findings?
bmjgast-2022-000871 to evaluate gut microbiome composition and diversity in ► Patients with SARS-­
CoV-­
2 infections possess

copyright.
both patients with SARS-­CoV-­2 PCR-­confirmed infections, significantly less bacterial diversity, lower abun-
► Additional supplemental
which had presented to Ventura Clinical Trials for care dance of Bifidobacterium and Faecalibacterium
material is published online
from March 2020 through October 2021 and SARS-­CoV-­2 and increased abundance of Bacteroides at the
only. To view, please visit the
PCR-­negative exposed controls. Patients were classified genus level compared with SARS-­CoV-­2-­exposed
journal online (http://d​ x.​doi.​
org/​10.​1136/b​ mjgast-​2022-​ as being asymptomatic or having mild, moderate or controls. There are inverse associations between
000871). severe symptoms based on National Institute of Health disease severity and the Shannon and Simpson
criteria. Exposed controls were individuals with prolonged diversity indices and also with Bifidobacterium and
or repeated close contact with patients with SARS-­ Faecalibacterium abundance. There is also a di-
Received 3 January 2022
CoV-­2 infection or their samples, for example, household rect association between severity and Bacteroides
Accepted 28 March 2022
members of patients or frontline healthcare workers. abundance.
Microbiome diversity and composition were compared
between patients and exposed controls at all taxonomic How might it impact on clinical practice in the
levels. foreseeable future?
Results Compared with controls (n=20), severely ► Boosting of Bifidobacterium or Faecalibacterium
symptomatic SARS-­CoV-­2-­infected patients (n=28) through probiotic supplementation or faecal mi-
had significantly less bacterial diversity (Shannon crobiota transplant is worthy of exploration in the
Index, p=0.0499; Simpson Index, p=0.0581), and management of patients with acute severe disease
positive patients overall had lower relative abundances or protracted infection. If the changes that we doc-
of Bifidobacterium (p<0.0001), Faecalibacterium ument precede SARS-­CoV-­2 infection in those who
(p=0.0077) and Roseburium (p=0.0327), while having are most severely affected, this therapeutic ap-
increased Bacteroides (p=0.0075). Interestingly, there proach may be of particular interest. Conversely, if
was an inverse association between disease severity and the reduction follows infection, then repopulation of
abundance of the same bacteria. the gut microbiome may reduce long-­term effects
© Author(s) (or their related to gut microbiome composition changes
Conclusion We hypothesise that low bacterial diversity
employer(s)) 2022. Re-­use with SARS-­CoV-­2 infection.
permitted under CC BY-­NC. No and depletion of Bifidobacterium genera either before
commercial re-­use. See rights or after infection led to reduced proimmune function,
and permissions. Published thereby allowing SARS-­CoV-­2 infection to become
by BMJ. symptomatic. This particular dysbiosis pattern may be a INTRODUCTION
For numbered affiliations see susceptibility marker for symptomatic severity from SARS-­ The abundance of Bifidobacterium decreases
end of article. CoV-­2 infection and may be amenable to preinfection, with increasing age and body mass index
intrainfection or postinfection intervention. (BMI)1 and Bifidobacterium is the active ingre-
Correspondence to Trial registration number NCT04031469 (PCR−) and
Dr Sabine Hazan; dient of many probiotics. In vitro studies
04359836 (PCR+).
​DrHazan@​progenabiome.c​ om have demonstrated the benefits of these

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 1

Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Gram-­positive bacteria to include enhanced ATP produc- are common component of several probiotic products
tion, immune modulation and competence,2–8 mucosal and appear to be associated with SAR-­CoV-­2 infections,
barrier integrity, restriction of bacterial adherence to and one could ask if probiotics might have a role in SARS-­
invasion of the intestinal epithelium and modulation of CoV-­2 therapy or prevention.
central nervous system activity.9 10 Additionally, Bifidobac- Herein, we evaluate the relationships between gut
terium have anti-­inflammatory properties: Bifidobacterium microbiome diversity and composition compared with
animalis, B. longum and B. bifidum decrease the function clinical outcome in cross-­sectional groups of SARS-­CoV-­2
of the ‘master switch’2 proinflammatory tumour necrosis PCR-­confirmed positive patients (ranging from asymp-
factor-α (TNF-α), increase the anti-­inflammatory cyto- tomatic to severely symptomatic) versus SARS-­ CoV-­ 2
kine IL-­10 and promote the Th1 while inhibiting the Th2 PCR-­confirmed negative controls. Our controls are SARS-­
immune response.8 In a mouse model of inflammatory CoV-­2 exposed persons who remained PCR negative
bowel disease (IBD), B. bifidum and B. animalis reduced and asymptomatic. The controls likely had similar viral
proinflammatory cytokines and restored intestinal exposures, but appeared protected against infection, and
barrier integrity.8 our data suggest that some protection may reside in the
With respect to SARS-­CoV-­2 infection, there is immu- microbiome.
nologic coordination between the gut and lungs.11–13
Numerous studies have suggested that a healthy gut
microbiome may be associated with a decrease in SARS-­ METHODS
CoV-­2-­related mortality14 and that probiotics should Study design and patients
be considered for prophylaxis15 and/or treatment of Individuals who were tested for SARS-­ CoV-­ 2 infection
SARS-­ CoV-­ 2 or its associated secondary infections.15–17 either because they were symptomatic or had been
However, as of February 2022, despite the publication exposed to a ‘case’ were eligible for enrolment the week
of nearly 8000 studies on SARS-­ CoV-­
2 infection, few following testing if either they or a household member
ongoing studies (​ clinicaltrials.​
gov: NCT04443075 and was positive. Controls eligible for enrolment were PCR
NCT04486482) and only five publications to date have negative for SARS-­CoV-­2, remained antibody negative for
examined gut microbiome changes in SARS-­ CoV-­2-­ 3–6 months and asymptomatic for 6–12 months. Addition-
infected patients. Nevertheless, an association between ally, controls had to either share a household with at least

copyright.
the status of the gut microbiome and outcome from this one symptomatic SARS-­CoV-­2-­positive individual or be a
infection has been suggested. Accordingly, increased healthcare worker who had been repeatedly exposed to
abundances of the Streptococcus, Rothia, Veilonella and symptomatic SARS-­CoV-­2-­positive patients or numerous
Actinomyces genera were associated with inflammation,18 SARS-­CoV-­2-­positive samples. All exposed controls were
whereas increased abundances of Collinsella aerofaciens, ones that, despite exposure to SARS-­CoV-­2, chose not
Collinsella tanakaei, Streptococcus infantis and Morganella to quarantine or take prophylaxis for SARS-­CoV-­2 infec-
morganii were associated with faecal samples with high tion and none had yet been vaccinated. Patients did not
SARS-­CoV-­2 infectivity,19 and increased Lachnospira- wear Personal Protective Equipment (PPE) inside their
ceae and Enterobacterioaceae abundances were associated homes and staff did not wear full PPE (ie, did not wear
with increased mortality and need for artificial ventila- masks) at the office because of its scarcity during this
tion.19 Species potentially protective against SARS-­CoV-­2 global pandemic. Patients undergoing treatment with
infection include Parabacteroides merdae, Bacteroides ster- total parenteral nutrition, or those with a history of signif-
coris, Alistipes onderdonkii, Lachnospiracea bacterium19 and icant gastrointestinal surgery (eg, bariatric surgery, total
F. prausnitzii,19 20 while vulnerability to infection and colectomy with ileorectal anastomosis, proctocolectomy,
increased severity were associated with decreased abun- postoperative stoma, ostomy or ileoanal pouch) were
dance of B.bifidum.20 21 A recent study correlated aspects excluded.
of the gut microbiome with ‘Long-­COVID’, including This study was performed between 1 March 2020 and
reduced levels of F. prausnitzii on admission.22 In short, 31 October 2021, with all but one subject recruited prior
there is still a compelling need to elucidate changes in to 1 June 2021. During that time, alpha and epsilon vari-
the human gut microbiome due to SARS-­CoV-­2 and their ants predominated in the USA.28
relationship with clinical outcomes.
The scientific community and lay public are increas- Assessments
ingly interested in the therapeutic potential of probi- A self-­administered questionnaire solicited information
otics. Bifidobacteria have potential to improve clinical on symptom severity, previous medical history, current
conditions ranging from IBD23 to Clostridioides difficile medication and probiotic use and exposure to recre-
infections.23–26 Treatment with specific strains of Bifido- ational drugs or animals. Patients with SARS-­CoV-­2 infec-
bacterium in vitro has been shown to reduce toxins from tion were further classified as being either asymptomatic
C. difficile .25 In vivo, Bifidobacterium can restore colonic carriers or having mild, moderate or severe symptoms as
integrity,27 and B. longum administered intranasally in per National Institute of Health, Clinical Spectrum of
mice prior to exposure to influenza has been associated SARS-­CoV-­2 Infection criteria.29 30 Asymptomatic PCR-­
with reduction in mortality.4 Given that Bifidobacterium confirmed SARS-­CoV-­2-­positive household members of

2 Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
SARS-­CoV-­2-­infected patients were categorised as asymp- composition and relative abundances of the top genera
tomatic carriers. Patients and controls were classified and species for all patients and controls.
as underweight, normal weight, overweight, obese or
severely obese based on BMI criteria of the Center for Data analysis
Disease Control and Prevention.31 We assessed differences in relative abundance across taxa
between the gut microbiome of SARS-­ CoV-­ 2-­
infected
Stool sample collection and processing patients and exposed controls and calculated Shannon
Patients and controls within the same household and Simpson alpha diversity indices with One Codex’s
collected stool samples within a week of the index case bioinformatics analysis pipeline using Jupyter Notebook
being positive. Patients had stool samples collected at in Python. Specifically, the One Codex Database consists
baseline, prior to any treatment, and within 48 hours of of ~114K complete microbial genomes (One Codex, San
symptom onset. No subjects had been commenced on Francisco, California). During processing, reads were first
antibiotics, SARS-­ CoV-­2 infection treatments, over the screened against the human genome and then mapped
counter (OTC) remedies (e.g., vitamins, antipyretics, to the microbial reference database using a k-­mer-­based
analgesics) or supplemental oxygen between the time of classification. Individual sequences (NGS read or contig)
symptom onset or demonstration of PCR positivity and were compared against the One Codex Database (One
stool collections. Subjects were instructed (and educated Codex) by exact alignment using k-­mers, where k=31.
on the procedure and sterile methods) to collect 1 mL Based on the relative frequency, unique k-­ mers were
filtered to control for sequencing or reference genome
of fresh stool and place it directly in a DNA/RNA Shield
artefacts.
Fecal Collection Tube (Zymo Research, Tustin, Cali-
The sequencing depth followed ProgenaBiome’s
fornia) and then mix sample thoroughly. Here,1 mL of
standard operating procedures and was 8 239 475 reads
faeces is more than sufficient to capture the microflora
on average for this study. One should note that shallow
of the gut accurately and consistently. This method is
metagenomic sequencing is typically only 0.5 million
chosen to eliminate the need for whole stool mixing and
reads but is still considered sufficient for taxonomic
aliquoting. The solution in the Fecal Collection Tube is
phyla level analysis (and even genera for the most abun-
designed to preserve samples at ambient temperature
dant bacteria).
(4°C–25°C) for >2 years, or below −20°C indefinitely.

copyright.
The relative abundance of each microbial taxonomic
Once samples reached our laboratory, they were immedi- classification was estimated based on the depth and
ately frozen at −20°C. coverage of sequencing across every available refer-
Following faecal collection, each individual sample ence genome. Beta-­diversity was calculated as weighted
DNA was extracted and purified with the Qiagen Power- UniFrac distance visualised in a distance matrix using
Fecal Pro DNA extraction kit. The isolated DNA was then the phylum-­level relative abundance obtained from One
quantitated using the Quantus Fluorometer with the Codex. Thirteen genera were selected based on our
QuantFluor ONE dsDNA kit. After DNA quantification, experience and knowledge of critical players in the gut
the DNA was normalised, that is, all samples begin library microbiome as well as similarity to other studies18 20 29
preparation (following DNA extraction and purification) To compare patients across subgroups and patients
with 100 ng of input DNA. Libraries were then prepared to exposed controls, Analysis of Variation (ANOVA),
using shotgun methodology with Illumina’s Nextera Flex Mann-­Whitney U, Kruskal-­Wallis tests and χ2 test statistics
kit. Samples then underwent the shotgun metagenomic were conducted using GraphPad V.8 with p values <0.05
processing procedure of tagmentation, amplification, considered as significant. Dunn’s post-­hoc was used for
indexing and purification. Following completion of Kruskal-­Wallis test, with correction for multiple compari-
this shotgun metagenomic standard protocol, purified sons in all situations.
libraries were again normalised to standardise sequencing All authors had access to study data and reviewed and
depth during the next-­ generation sequencing (NGS) approved the final manuscript.
run on the NextSeq 500/550. We achieved consistency
of sequencing depth (ie, number of reads) by normal-
ising the samples’ pooling concentrations (ie, molarity), RESULTS
loading the same number of samples per sequencing run, Patient characteristics
consistently using the same NextSeq High Output kits. Demographic and clinical characteristics of patients
After completion of sequencing on the Illumina (n=50) and exposed controls (n=20) are presented in
NextSeq with 500/550 High-­ Output Kits V.2.5 (300 online supplemental tables 1 and 2, and summarised in
cycles), the raw data were streamed in real time to Illumi- table 1. All patients were resident of USA, with states indi-
na’s BaseSpace cloud for FASTQ (Fast Quality, a standard cated in online supplemental table 1. Twenty-­four of 50
text file type for storing biological sequence information) (48%) patients and 7 of 20 (35%) of exposed controls
conversion. The FASTQ files were then sent through were men. The mean±SEM age in years was 50.0±2.5 for
One Codex’s bioinformatics pipeline for metagenomic patients and 44.4±3.6 for exposed controls. Fourty-­four
annotation and analyses to elucidate the microflora of 50 (88%) patients were non-­Hispanic white; 5 of 50

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 3

4
Open access

Table 1 Summary of demographics along with clinical and dietary characteristics of subjects

SARS-­CoV-­2 negative; SARS-­CoV-­2 positive

exposed control Asymptomatic Mild Moderate Severe Total
(n=20) (n=4) (n=6) (n=12) (n=28) (n=50) P value
Demographics
 USA resident 20 (100.00%) 4 (100.00%) 6 (100.00%) 12 (100.00%) 28 (100.00%) 50 (100.00%) ns
 Male 7 (35.00%) 3 (75.00%) 2 (33.33%) 5 (41.66%) 14 (50.00%) 24 (48.00%) 0.5706
 Mean age ±SEM 44.40±3.62 47.50±9.40 37.67±4.65 50.58±5.92 52.82±3.33 50.00±2.50 0.5221
 Median age 48 50.5 37 60 55.5 53
 Race
 White 17 (85.00%) 3 (75.00%) 5 (83.33%) 9 (75.00%) 27 (96.43%) 44 (88.00%) 0.3826
 Hispanic 2 (10.00%) 1 (25.00%) 0 (0.00%) 3 (25.00%) 1 (3.57%) 5 (10.00%)
 Black 1 (5.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
 Other 0 (0.00%) 0 (0.00%) 1 (16.66%) 0 (0.00%) 0 (0.00%) 1 (2.00%)
Clinical/dietary characteristics
 Severe COVID-­19 12 (60.00%) 3 (75.00%) 3 (50.00%) 10 (83.33%) 16 (57.14%) 32 (64.00%) 0.5099
comorbidities*
 Normal stool 20 (100.00%) 4 (100.00%) 6 (100.00%) 10 (83.33%) 24 (85.71%) 44 (88.00%) 0.2891
frequency
 Lack of appetite 0 (0.00%) 0 (0.00%) 0 (0.00%) 1 (8.33%) 2 (7.14%) 3 (6.00%) 0.6663
 Omnivore diet 19†(100.00%) 4 (100.00%) 6 (100.00%) 12 (100.00%) 28 (100.00%) 50 (100.00%) ns

. Numbers in cells indicate number of subjects with percentage in categories, except for age and BMI, which indicate value. P values calculated via one-­way ANOVA (age and BMI) or χ2
(others). Normal stool frequency is defined as absence of diarrhoea. Total refers to sum of SARS-­CoV-­2-­positive subjects, and it is not used in statistics.
*Comorbid conditions indicative of severe SARS-­CoV-­2 infection (not including hypertension), based on Center for Disease Control (CDC) definitions.44
†Nineteen total subjects had data available for diet within exposed control.

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

copyright.
BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
(10%) were Hispanic and 1 of 50 (2%), Native Amer-
ican and 17 of 20 (85.0%) of exposed controls were non-­
Hispanic white; 2 of 20 (10.0%), Hispanic and 1 of 20
(5.0%), Black. Of patients, 28 of 50 (56%) had severe, 12
of 50 (24%) had moderate and 6 of 50 (12%) had mild
disease and 4 of 50 (8%) were asymptomatic. Thirty-­two of
50 (64%) patients and 12 of 20 (60.0%) exposed controls
had underlying comorbidities considered risk factors for
increased severity of SARS-­CoV-­2 infection by the Center
for Disease Control (CDC).1 The mean±SEM BMI of the
46 patients with available data was 27.1±0.98 compared
with 25.1±0.96 for the 20 exposed controls. There was no
significant difference (p>0.2) in gender, age, racial demo-
graphics, loss of appetite, change in stool frequency, diet
or presence of underlying comorbidities. Figure 2 Diversity of gut microbiome composition of
Of the exposed controls, 16 were household contacts of SARS-­CoV-­2 positive patients (severely symptomatic: n=28;
SARS-­CoV-­2-­positive patients in the study, 2 were health- moderately symptomatic: n=12; mildly symptomatic: n =
care workers with extensive, non-­ protected, exposure 6; asymptomatic: n=4) versus exposed controls (n=20). (A)
to SARS-­CoV-­2-­positive patients and 2 were laboratory Shannon index (p=0.0499), (B) Simpson index (p=0.0581).
personnel exposed to thousands of SARS-­CoV-­2 samples Differences between severely symptomatic positive and
exposed negative controls were analysed via Kruskal-­Wallis
(healthcare workers and laboratory personnel did not
test Dunn’s post-­hoc, correcting for multiple comparisons,
wear full PPE, that is, did not wear a face mask, due its showing significant for Shannon index at p=0.0201.
scarcity; see the Methods section). During the timeframe
of the study, none of the patients or controls was on SARS-­
CoV-­2 prophylaxis or treatment, and none had yet been and 7.2% (SARS-­CoV-­2 positive) and Bifidobacterium 7.6%
vaccinated. No patients or exposed control were positive (exposed control) and 1.5% (SARS-­Cov-­2 positive).
for SARS-­CoV-­2 prior to the study. Figure 2 shows two diversity indices for all subgroups

copyright.
Gut microbiome diversity and composition studied, namely, Shannon diversity (figure 2A) and
Figure 1 depicts pie charts of the composition of the gut Simpson diversity index (figure 2B). The overall p value
microbiome for the exposed control at the phylum level for Shannon index (richness of bacterial composition)
(figure 1A) and genus level (figure 1B). At phylum level, demonstrated a significant (p=0.0499) decrease in diver-
Firmicutes and Bacteroides dominated, comprising 59.6% sity with increased severity, and significance was seen for
(exposed control) and 54.7% (SARS-­CoV-­2 positive) and exposed control versus severely symptomatic (p=0.0201),
29.1% (exposed control) and 40.4% (SARS-­CoV-­2 posi- analysed via Kruskal-­ Wallis test. The Simpson (even-
tive) of all phyla, respectively. At the level of genus, Bacte- ness of bacterial composition) indexes showed a trend
roides contributed 12.4% (exposed control) and 21.8% (p=0.0581) of a decrease in diversity with increased
(SARS-­CoV-­2 positive), Alistipes 6.4% (exposed control) SARS-­CoV-­2 severity.
Further metagenomic analysis comparing SARS-­CoV-­2
patients and controls revealed significant differences
in relative abundance of specific bacteria. The relative
abundance of SARS-­ CoV-­2 positive (exposed control)
versus negative subjects is presented in table 2, along
with comparative p values via Mann-­ Whitney U test.
Patients with SARS-­CoV-­2 infection showed a significantly
decreased relative abundance of Bifidobacterium and
Faecalibacterium, and significantly increased relative abun-
dance of Bacteroides (table 2).
Table 3 lists the genera/species relative abundances
(mean±SEM) for various levels of severity of SARS-­CoV-­
2-­positive patients versus exposed control. Analysed via
Kruskal-­Wallis test, the main effect (ie, overall p value)
of these changes are shown in the left column. Table 3
proceeds to compare, correcting for multiple compar-
ison, the three levels of severity in infected patients versus
Figure 1 Distribution of bacterial relative abundance in exposed control and asymptomatic groups. Specifically,
various (A) phlya and (B) genera for exposed control (n=20, increased disease severity was associated with decreased
left) and SARS-­CoV-­2 positive subjects (n=50, right). relative abundance of Bifidobacterium, Faecalibacterium, F.

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 5

Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Table 2 Relative abundances of Bacteroides increase and of Bifidobacterium, Faecalibacterium and Roseburium decrease in
SARS-­CoV-­2 positive subjects versus SARS-­CoV-­2 negative exposed controls
Relative abundance (mean±SEM)
Genus (±species) Exposed controls SARS-­CoV-­2 positive P value
Alistipes 0.0639±0.0095 0.0721±0.0100 0.8709
Bacteroides 0.1235±0.0178 0.2183±0.0191 0.0025
Bifidobacterium 0.0755±0.0219 0.0147±0.0051 <0.0001
Blautia 0.0261±0.0040 0.0524±0.0088 0.1349
Clostridium 0.0431±0.0075 0.0309±0.0039 0.9948
Collinsella 0.0146±0.0045 0.0158±0.0029 0.9948
Dorea 0.0137±0.0024 0.0185±0.0022 0.2777
Eubacterium 0.0441±0.0063 0.0402±0.0043 0.4786
Faecalibacterium 0.0550±0.0086 0.0310±0.0039 0.0137
F. prausnitzii 0.0542±0.0085 0.0313±0.0039 0.0153
Prevotella 0.0110±0.0086 0.0091±0.0066 0.6538
Roseburium 0.0329±0.0056 0.0195±0.0032 0.0097
Ruminococcus 0.0376±0.0079 0.0391±0.0056 0.9844
Mean±SEM relative abundances, as well as p value via Mann-­Whitney U test are indicated, with bold marking p<0.05.

prausnitizii and Roseburium, along with an increased rela- with axis representing directions of divergence). Herein,
tive abundance of Bacteroides. the PC1 accounts for 43.16% of the variation, whereas
Depicted in figure 3 are the 12 most abundant fami- PC2 accounts for 12.78%. This analysis reveals a clear
lies and the 12 most abundant genera for patients, strat- divergence of a subset of SARS-­CoV-­2-­positive patients

copyright.
ified by disease severity and in comparison to exposed clustering on the right side tracking along the x-­ axis
controls. Distinguished by colour, the bars represent (PC1), highlighting microbiota divergence as a function
the relative per cent bacterial families and genera abun- of disease. Thus, figure 6 shows that exposed controls
dance. Note the reduced diversity of the microbiome of cluster similar separately from SARS-­CoV-­2 patients; that
SARS-­CoV-­2-­positive patients shown in column B. is, patients are more similar in terms of their microbiome
Figure 4 summarises the microbiome changes according to each other than to controls.
to SARS-­CoV-­2 positivity and severity, with green boxes
depicting significant elevation and red boxes indicating
significant depletion in genera/species abundance asso- DISCUSSION
ciated with SARS-­CoV-­2. Immune function and health could be enhanced by bacterial
Figure 5A,B exhibit the relative abundance of Bifidobac- abundance
terium for each subject, grouped by SARS-­CoV-­2 infection Interactions between the host and gut microbiota are
severity. This diagram, with subjects groups ordered by complex, numerous and bidirectional. Gut microbiota
severity (from severe on left, to exposed controls on the regulate the development and function of the innate and
right), depicts how Bifidobacterium abundance increases adaptive immune systems,32 potentially allowing them
as severity decreases. to protect against infections and infection severity. The
Analysis of the beta-­diversity of subjects demonstrated primary findings of our study are that SARS-­CoV-­2 posi-
that the diversity of control subjects cluster separately tivity and infection severity are associated with decreased
from that of patients. Figure 6A shows the beta-­diversity-­ levels of the protective Bifidobacterium and Faecalibacterium
weighted (quantitative) UniFrac analysis featuring phyla genera and with decreased bacterial diversity, as exempli-
bacterial profiles for all individuals in the study (n=70). fied by the Shannon and Simpson indices. This accords
Figure 6A reveals that, although there is a range of with studies showing bacterial diversity inversely relates to
dissimilarity, the SARS-­ CoV-­ 2-­
negative individuals are the presence of various common disorders.33 Uniquely,
more similar to one another than they are to SARS-­CoV-­ our study compared SARS-­CoV-­2-­exposed SARS-­CoV-­
2-­positive patients. The matrix also highlights clusters 2-­negative persons (ie, controls) with symptomatic and
of similarity among SARS-­ CoV-­2-­
positive patients, and asymptomatic SARS-­ CoV-­ 2-­
positive patients. Thus, we
darker quadrants of dissimilarity where positive and nega- controlled for SARS-­CoV-­2 exposure.
tive patients intersect. At a more granular level, figure 6B The genus Bifidobacterium has important immune func-
used principal component (PC) analysis of genera, where tions,8 is a major component of the microbiome and is
the axes depict the per cent of variance. In PC analysis, frequently used in probiotics.34 Bifidobacterum increase
points closer together are more similar (less divergence Treg responses and reduce cell damage by inhibiting

6 Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

 Table 3 Relative abundance of Bacteroides increases and those of Bifidobacterium, Faecalibacterium and Roseburium decrease with increasing severity of disease
Relative abundance (mean±SEM)
Genus
(±species) Main effect Severe Moderate Mild Asymptomatic Exposed control
A. Relative abundance of genera/species for various severities of SARS-­CoV-­2 positivity, as well as exposed controls (SARS-­CoV-­2 negative). Overall p value of Kruskal-­Wallis test is indicated in
‘Main Effect’ column.
Alistipes 0.8119 0.0793±0.0166 0.0520±0.0084 0.0664±0.0196 0.0901±0.0272 0.0639±0.0095
Bacteroides 0.0075 0.2187±0.0272 0.2849±0.0305 0.0912±0.0226 0.2058±0.0606 0.1235±0.0178
Bifidobacterium <0.0001 0.0018±0.0006 0.0037±0.0015 0.0507±0.0244 0.0840±0.0308 0.0755±0.0219
Blautia 0.2098 0.0495±0.0095 0.0426±0.0132 0.1037±0.0510 0.0260±0.0067 0.0261±0.0040
Clostridium 0.2721 0.0331±0.0062 0.0334±0.0055 0.0265±0.0083 0.0145±0.0025 0.0431±0.0075
Collinsella 0.7476 0.0145±0.0035 0.0107±0.0038 0.0203±0.0126 0.0334±0.0174 0.0146±0.0045
Dorea 0.4820 0.0218±0.0035 0.0126±0.0032 0.0174±0.0046 0.0152±0.0015 0.0137±0.0024
Eubacterium 0.9619 0.0436±0.0070 0.0367±0.0064 0.0366±0.0085 0.0332±0.0058 0.0441±0.0063
Faecalibacterium 0.0077 0.0209+0.0037 0.0428+0.0093 0.0359±0.0139 0.0597+0.0098 0.0550±0.0086
F. prausnitzii 0.0054 0.0220±0.0041 0.0417±0.0092 0.0356±0.0137 0.0589±0.0095 0.0542±0.0085
Prevotella 0.1687 0.0149±0.0118 0.0000±0.0000 0.0050±0.0046 0.0023±0.0020 0.0110±0.0086
Roseburium 0.0327 0.0146±0.0037 0.0245±0.0075 0.0268±0.0138 0.0274±0.0081 0.0329±0.0056

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

Ruminococcus 0.8033 0.0384±0.0076 0.0293±0.0073 0.0620±0.0250 0.0394±0.0139 0.0376±0.0079
P value vs exposed control P value vs asymptomatic
Genus
(±species) Mild Moderate Severe Asymptomatic Mild Moderate Severe
B. P value for various levels of severity of symptomatic infection, compared to negative exposed controls and positive asymptomatic subjects, using Dunn’s post-­hoc.
Alistipes >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999
Bacteroides >0.9999 0.0078 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999
Bifidobacterium >0.9999 0.0006 0.0026 >0.9999 >0.9999 0.0006 <0.0001
Blautia 0.4116 >0.9999 0.6373 >0.9999 >0.9999 >0.9999 >0.9999
Clostridium >0.9999 >0.9999 >0.9999 0.4944 >0.9999 0.9841 >0.9999
Collinsella >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999
Dorea >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999
Eubacterium >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999
Faecalibacterium >0.9999 >0.9999 0.0082 >0.9999 >0.9999 >0.9999 0.1422
F. prausnitzii >0.9999 >0.9999 0.0109 >0.9999 >0.9999 >0.9999 0.1568
Prevotella >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 0.8916 >0.9999
Roseburium >0.9999 >0.9999 0.0196 >0.9999 >0.9999 >0.9999 >0.9999
Ruminococcus >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999 >0.9999

For A and B, bold values indicate p<0.05. Note, no apparently significant (p<0.05) p values were observed in post-­hocs, with main effects non-­significant.
Open access

7
copyright.
BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
copyright.

Figure 3 Graphic of relative abundance of the 12 most common (A) families and (B) genera. The top group represents the
SARS-­CoV-­2 positive samples (n=50), stratified by severity. The bottom group represents the exposed control samples (n=20).
The coloured boxes represent the fraction of the entire rectangle composed of the given family/genera of bacteria.

TNF-α and macrophages.35 Bifidobacterium also protects numerous disease states.37 The numbers of commensal
against intestinal epithelial cell damage independently Bifidobacterium have been shown to decrease with age
from their effects on TNF-α production. The exopolysac- and obesity, major SARS-­CoV-­2 infection risk factors. We
charide coat, which is a feature of some Bifidobacterium, demonstrate that patients with a more severe course of
plays a significant role in this protective effect.36 These viral infection had decreased abundance of Bifidobac-
immune functions of Bifidobacterium could be critical in terium. However, it should be noted that there are no
relation to its SARS-­CoV-­2 infection-­prevention effects. definitive studies concerning what constitutes a normal
Evidence has accumulated to support a beneficial baseline abundance of Bifidobacterium in a ‘healthy’
effect from supplementation with Bifidobacterium in individual.

8 Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Figure 4 Diagram of taxa comparing the gut microbiome of SARS-­CoV-­2 patients and exposed controls. Red or green
background indicates a significant depletion or increase (due to positivity or severity), respectively, of the genus or species in
SARS-­CoV-­2 positive subjects.

The abundance of Faecalibacterium genus and F. praus- analysis. Age and diabetes are risk factors for SARS-­CoV-­2
nitzii species was also inversely related to SARS-­CoV-­
2 infection, and F. prausnitzii levels decline markedly in
positivity and SARS-­ CoV-­
2 infection severity in this elder and diabetic populations.37 In fact, Faecalibacterium
levels have been considered an indirect ‘indicator’ of
overall human health.38 The abundance of F. prausnitzii
is reduced by the ‘Western’ diet (consumption of more
meat, animal fat, sugar, processed foods and low fibre),
while it is enhanced by the high-­fibre containing ‘Medi-
terranean’ diet of vegetables and fruits with low meat
intake.39 Preliminary studies showed that reduced inges-
tion of a Mediterranean diet within the same country
is associated with increased SARS-­ CoV-­2-­
related death

copyright.
rates.40 In short, we show that F. prausnitzii levels nega-
tively correlated to SARS-­ CoV-­2 infection severity and
prior studies show that reduced F. prausnitzii is associated
with SARS-­CoV-­2 infection vulnerabilities such as age,
diabetes, obesity and possibly diet.
SARS-­CoV-­2 positivity and severity were also associated
with decreased abundance of Roseburium and increased
abundance of Bacteroides. The implications of these
changes remain unclear.

Innate immunity could be enhanced by increased level of

beneficial bacteria
The pathological impact of SARS-­ CoV-­2 infection
includes both direct effects from viral invasion and
complex immunological responses including, in its most
severe form, the ‘cytokine storm’. The cytokine storm
is the result of a sudden increase in circulating levels
of proinflammatory cytokines produced by activated
Figure 5 Relative abundance of Bifidobacterium in SARS-­ macrophages, mast cells, endothelial cells and epithelial
CoV-­2 positive patients (n=50) versus SARS-­CoV-­2 negative cells during innate immune responses, which appear to
exposed controls (n=20). Data are plotted as (A) mean with be modulated by the abundance of Bifidobacterium and
error bars for 95% CI and (B) individual points of relative Faecalibacterium and bacterial diversity (5, 23, 25). Steroid
abundance for varying SARS-­CoV-­2 infection severity. treatment has situational success in SARS-­CoV-­2 infec-
Analysed via Kruskal-­Wallis test, there were significant tion, based on suppressing this over activation of the
reductions in Bifidobacterium relative abundance for innate immune system, reviewed by Tang et al41
severely (p<0.0001) and moderately (p=0.0002) symptomatic
Zhao reported that elevated serum levels of proin-
patients. Subjects 1–28 = severely symptomatic; subjects
29–40 = moderately symptomatic; subjects 41–46 = mildly flammatory cytokines such as IL-­16 and IL-­17 predict
symptomatic subjects; subjects 47–50 = asymptomatic; poor prognoses in patients with SARS-­CoV-­2 infection.42
subjects 51–70 = exposed control. Figure A,B depicts same Also, Tao et al showed that changes in gut microbiota
data. composition might contribute to SARS-­ CoV-­2-­
induced

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 9

Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
copyright.
Figure 6 SARS-­CoV-­2 positive patients’ microbiome is more similar to each other than to that of exposed controls. (A)
Weighted UniFrac distance matrix of phylum level SARS-­CoV-­2 positive (n=50) and exposed negative control samples (n=20).
Distance of microbiome differences increases with increasing blue colour intensity (see legend top right). The centre of the
diagram consists of negative subjects on both axis and is yellow indicative of less distance (ie, lessdifference in microbiome).
The central area of the left as well as central-­top side of diagram, consists of negative subjects on one axis and positive on
the other, and are darker blue, indicative of more distance (more difference in microbiome). (B) Principal component analysis of
microbiota from SARS-­CoV-­2 positive (n=50) and exposed negative controls (n=20). Dots closer in distance are more similar in
microbiome composition. Axes depict the per cent of variance explained by principal component (PC) 1 and 2. Plots are based
on bacterial genera relative abundance profiles.

production of inflammatory cytokines in the intestine, enzyme 1 pathway.16 Additionally, some species of Bifido-
which may lead to cytokine storm onset.29 Both authors bacterium are likely to prevent the replication of coro-
report significantly reduced gut microbiota diversity naviruses by reducing endoplasmic reticulum stress,
and increased opportunistic pathogens in patients with also through the inositol-­requiring enzyme 1 pathway.
SARS-­ CoV-­
2. Interestingly, the bloom of opportunistic Reduced Bifidobacterium abundance has been observed
pathogens positively correlated with the number of in the gut microbiome of patients with IBD, which has
Th17 cells. Bozkurt and Quigley reported that IL-­6 and mechanisms involving IL-­17.31 Furthermore, the direct
IL-­17 promote viral persistence by immune interactions endoscopic delivery of Bifidobacterium has been shown
through cellular autophagy via the inositol-­ requiring to be effective in promoting symptom resolution and

10 Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
Figure 7 Potential mechanism for cytokine storm and immune hyper-­response in SARS-­CoV-­2 positive patients. In individuals
infected with SARS-­CoV-­2, the macrophages become activated; these in turn activate T-­cells, additional macrophages, and
neutrophils―all of which release cytokines, including TNF-α. Bifidobacterium, when present in sufficient numbers, can bind to
TNF-α and prevent the subsequent cytokine storm. Therefore, patients with a bifidobacterial dysbiosis characterised by low
levels of Bifidobacterium lack this line of defense, which may lead to a cytokine storm.

copyright.
mucosal healing in IBD—an effect likely to be associated With the lack of data on the gut microbiome prior to
with the anti-­Th17 effect of Bifidobacterium.(8) Figure 7 onset of SARS-­CoV-­2 infection, we cannot completely rule
demonstrates how Bifidobacterium might hypothetically out the confounding effect of illness on the microbiome.
quell a heightened immune response by dampening the Nonetheless, we eliminated effects of treatment on the
effect of the master switch TNF-α. gut microbiome by sampling prior to administration of
SARS-­CoV-­2 infection therapeutics of any kind and within
48 hours of symptom onset. Specifically, no subjects were
CONCLUSIONS given antibiotics, antivirals, anti-­inflammatory medicines,
Given our cross-­sectional study design, it is not possible oxygen or any other therapeutic agent between symptom
to determine whether the differences in Bifidobacterium onset or PCR positivity and stool sampling. We also note
levels observed between patients and exposed controls that the prevalence of appetite changes, alterations of
preceded or followed infection. If preceding infection, stool frequency and gastrointestinal (GI) symptoms, in
they could be a marker of susceptibility, and boosting general, were not significantly different between any of
Bifidobacterium levels might decrease the risk or severity the severity groups or controls (table 1), although the
of SARS-­CoV-­2 infection. If these changes followed infec- small sample sizes for some groups should be considered
tion, alteration of the gut microbiome (such as through in evaluating these statistics.
faecal microbiota transplantation or possibly probi- SARS-­ CoV-­ 2 infection presentation variability
otic supplementation) to increase Bifidobacterium could correlates with colon microbiome bacterial composition
be an area worth exploring for improved outcomes. If and overall diversity. The same changes we observe due
future studies can demonstrate improved outcomes, such to SARS-­CoV-­2 infection, namely reduced Bifidobacterium
therapy can be considered for complex cases of SARS-­ and/or Faecalibacterium abundance, are associated with
CoV-­2 infection, such as ‘long-­haulers’, and those with SARS-­ CoV-­2 infection risk factors, including old age,
severe disease. Developing outbreaks within tightly closed obesity and diabetes.9 37 39 43 Thus, colon microbiome
communities such as nursing homes might be a good diversity and relative abundance of Bifidobacterium and
setting in which to assess susceptibility: faecal samples Faecalibacterium should be explored as potential markers
could be collected during the outbreak and run post hoc for predicting SARS-­CoV-­2 infection severity.
on ‘cases’ and ‘controls’. Future studies of individuals In summary, we demonstrate in a study of PCR-­positive
with baseline prepandemic microbiome data would be and PCR-­negative SARS-­CoV-­2-­exposed subjects, reduced
highly valuable, although acquiring such baseline prein- bacterial diversity and reduced levels of various genus/
fection microbiome data is still costly. species are highly associated with both SARS-­ CoV-­ 2

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 11

 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
positivity and SARS-­CoV-­2 infection severity. These find- Supplemental material This content has been supplied by the author(s). It
ings suggest that probiotic supplementation or faecal has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have
been peer-­reviewed. Any opinions or recommendations discussed are solely
microbiota transplantation should be explored as a those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability
potential therapeutic avenue for patients with SARS-­ and responsibility arising from any reliance placed on the content. Where the
CoV-­2. Additionally, individual colon microbiome eval- content includes any translated material, BMJ does not warrant the accuracy and
reliability of the translations (including but not limited to local regulations, clinical
uation may predict vulnerability to the development of guidelines, terminology, drug names and drug dosages), and is not responsible
severe SARS-­CoV-­2 infection. Finally, our data suggest a for any error and/or omissions arising from translation and adaptation or
new area for exploration: if SARS-­CoV-­2 severity is found otherwise.
to be dependent on the microbiome, then accounting Open access This is an open access article distributed in accordance with
for microbiome differences could reduce variability in the Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license,
which permits others to distribute, remix, adapt, build upon this work non-­
outcomes for SARS-­CoV-­2. commercially, and license their derivative works on different terms, provided the
original work is properly cited, appropriate credit is given, any changes made
Author affiliations indicated, and the use is non-­commercial. See: http://creativecommons.org/​
1 licenses/by-nc/4.0/.
N/A, ProgenaBiome LLC, Ventura, California, USA
2
Division of Gastroenterology, Alta Bates Summit Medical Center, Berkeley,
ORCID iDs
California, USA
3 Sonya Dave http://orcid.org/0000-0002-7653-0388
Clinic of Gastroenterology, Istanbul Maltepe University, Istanbul, Turkey
4 Eamonn MM Quigley http://orcid.org/0000-0003-4151-7180
N/A, Microbiome Research, Inc, Ventura, California, USA
5
Medical Writing and Biostatistics, North End Advisory, Smyrna, Georgia, USA
6
Division of Gastroenterology and Hepatology, The Methodist Hospital, Weill Cornell
Medical College, Houston, Texas, USA REFERENCES
7 1 De Vuyst L, Moens F, Selak M, et al. Summer meeting 2013: growth
N/A, Centre for Digestive Diseases, Five Dock, New South Wales, Australia
and physiology of bifidobacteria. J Appl Microbiol 2014;116:477–91.
2 Parameswaran N, Patial S. Tumor necrosis factor-α signaling in
Acknowledgements Medical writing assistance was provided by Sonya Dave, macrophages. Crit Rev Eukaryot Gene Expr 2010;20:87–103.
PhD (an author on the publication) and was funded by ProgenaBiome. The authors 3 Martins AKS, Martins FS, Gomes DA, et al. Evaluation of in vitro
thank all clinicians for their involvement and contribution to the study. The authors antagonism and of in vivo immune modulation and protection
thank Kate Hendricks, MD, MPH and TM for many helpful editorial suggestions. against pathogenic experimental challenge of two probiotic
Finally, the authors owe a depth of gratitude to the late Sydney M Finegold, MD strains of Bifidobacterium animalis var. lactis. Arch Microbiol
for mentorship that sparked the interest in the microbiome to many scientists, 2010;192:995–1003.
4 Groeger D, Schiavi E, Grant R, et al. Intranasal Bifidobacterium

copyright.
including authors of this paper.
longum protects against viral-­induced lung inflammation and
Contributors All authors (SH, NS, HSB, SD, AJP, JD, BDB, EMMQ and TJB) injury in a murine model of lethal influenza infection. EBioMedicine
participated in the drafting, critical revision, and approval of the final version of 2020;60:102981.
the manuscript. SH led study design. SH and AJP conducted the bioinformatic 5 Konieczna P, Akdis CA, Quigley EMM, et al. Portrait of an
analysis. SD conducted the statistical analysis and was a major contributor to immunoregulatory Bifidobacterium. Gut Microbes 2012;3:261–6.
6 Konieczna P, Ferstl R, Ziegler M, et al. Immunomodulation by
writing the paper. SH was primarily responsible for interpretation of the study Bifidobacterium infantis 35624 in the murine lamina propria requires
results, with contributions from all authors. EMMQ and TJB are senior authors who retinoic acid-­dependent and independent mechanisms. PLoS One
provided overall direction and advice. SH acts as the guarantor and accepts full 2013;8:e62617.
responsibility for the work, has access to all data, and controlled the decision to 7 Schiavi E, Plattner S, Rodriguez-­Perez N, et al. Exopolysaccharide
publish. from Bifidobacterium longum subsp. longum 35624™ modulates
murine allergic airway responses. Benef Microbes 2018;9:761–73.
Funding The authors have not declared a specific grant for this research from any 8 Ruiz L, Delgado S, Ruas-­Madiedo P, et al. Bifidobacteria and their
funding agency in the public, commercial or not-­for-­profit sectors. molecular communication with the immune system. Front Microbiol
Competing interests SH declares that she has pecuniary interest in Topelia 2017;8:2345.
9 Marras L, Caputo M, Bisicchia S, et al. The role of bifidobacteria
Pty Ltd in Australia, and Topelia Pty Ltd in USA where development of COVID-­19
in predictive and preventive medicine: a focus on eczema and
preventative/treatment options is being pursued. She has also filed patents hypercholesterolemia. Microorganisms 2021;9. doi:10.3390/
relevant to Coronavirus treatments. She is the founder and owner of Microbiome microorganisms9040836. [Epub ahead of print: 14 04 2021].
research foundation, Progenabiome and Ventura Clinical Trials. TJB declares that he 10 Stavropoulou E, Bezirtzoglou E. Probiotics in medicine: a long
has pecuniary interest in Topelia Pty Ltd in Australia, and Topelia Therapeutics Inc. debate. Front Immunol 2020;11:2192.
in USA developing COVID-­19 preventative/treatment medications. He has also filed 11 Ahlawat S, Sharma KK. Immunological co-­ordination between gut
patents relevant to COVID-­19 treatments. SD declares she has corporate affiliation and lungs in SARS-­CoV-­2 infection. Virus Res 2020;286:198103.
to McKesson Specialty Health/Ontada and North End Advisory, LLC. SD is unaware 12 Follmer C. Viral infection-­induced gut dysbiosis, neuroinflammation,
of SARS-­CoV-­2 and microbiome projects and not directly involved in COVID-­19 and α-synuclein aggregation: updates and perspectives on
COVID-­19 and neurodegenerative disorders. ACS Chem Neurosci
relevant projects at McKesson, but they may exist. AJP and BDB have corporate
2020;11:4012–6.
affiliations to Progenabiome. EMMQ serves as a consultant to Precisionbiotics, 13 Marsland BJ, Trompette A, Gollwitzer ES. The Gut-­Lung axis in
Novazymes, Salix, Biocodex and Axon Pharma and has received research support respiratory disease. Ann Am Thorac Soc 2015;12 Suppl 2:S150–6.
from 4D Pharma. 14 Janda L, Mihalčin M, Šťastná M. Is a healthy microbiome responsible
for lower mortality in COVID-­19? Biologia 2021;76:819–29.
Patient consent for publication Not applicable. 15 Tiwari SK, Dicks LMT, Popov IV, et al. Probiotics at war against
Ethics approval The study was conducted in accordance with ethical principles of viruses: what is missing from the picture? Front Microbiol
the Declaration of Helsinki, the International Council for Harmonisation Harmonised 2020;11:11.
Tripartite Guideline for Good Clinical Practice and the Ethical and Independent Review 16 Bozkurt HS, Quigley EM. The probiotic Bifidobacterium in the
management of Coronavirus: A theoretical basis. Int J Immunopathol
Board. This study involves human participants and was approved by the 'Ethical and
Pharmacol 2020;34:2058738420961304.
Independent Review Board' (IRB, IRB00007807). Participants gave informed consent 17 Jin X, Lian J-­S, Hu J-­H, et al. Epidemiological, clinical and virological
to participate in the study before taking part. characteristics of 74 cases of coronavirus-­infected disease 2019
Provenance and peer review Not commissioned; externally peer reviewed. (COVID-­19) with gastrointestinal symptoms. Gut 2020;69:1002–9.
18 Gu S, Chen Y, Wu Z, et al. Alterations of the gut microbiota in
Data availability statement Data are available upon reasonable request. Data patients with coronavirus disease 2019 or H1N1 influenza. Clin Infect
available upon reasonable request from corresponding author, Dr. Sabine Hazan. Dis 2020;71:2669–78.

12 Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871

 Open access

BMJ Open Gastroenterol: first published as 10.1136/bmjgast-2022-000871 on 28 April 2022. Downloaded from http://bmjopengastro.bmj.com/ on May 15, 2023 by guest. Protected by
19 Zuo T, Liu Q, Zhang F, et al. Depicting SARS-­CoV-­2 faecal viral 33 Lloyd-­Price J, Abu-­Ali G, Huttenhower C. The healthy human
activity in association with gut microbiota composition in patients microbiome. Genome Med 2016;8:51.
with COVID-­19. Gut 2021;70:276–84. 34 Ghouri YA, Richards DM, Rahimi EF, et al. Systematic review
20 Yeoh YK, Zuo T, Lui GC-­Y, et al. Gut microbiota composition reflects of randomized controlled trials of probiotics, prebiotics, and
disease severity and dysfunctional immune responses in patients synbiotics in inflammatory bowel disease. Clin Exp Gastroenterol
with COVID-­19. Gut 2021;70:698–706. 2014;7:473–87.
21 Xu K, Cai H, Shen Y. Management of COVID-­19: the Zhejiang 35 Hughes KR, Harnisch LC, Alcon-­Giner C, et al. Bifidobacterium
experience. Zhejiang Da Xue Xue Bao Yi Xue Ban 2020;49:147–57. breve reduces apoptotic epithelial cell shedding in an
22 Liu Q, Mak JWY, Su Q, et al. Gut microbiota dynamics in a exopolysaccharide and MyD88-­dependent manner. Open Biol
prospective cohort of patients with post-­acute COVID-­19 syndrome.
2017;7:160155.
Gut 2022;71:544–52.
36 Fanning S, Hall LJ, Cronin M, et al. Bifidobacterial surface-­
23 Din AU, Hassan A, Zhu Y, et al. Inhibitory effect of Bifidobacterium
exopolysaccharide facilitates commensal-­host interaction through
bifidum ATCC 29521 on colitis and its mechanism. J Nutr Biochem
2020;79:108353. immune modulation and pathogen protection. Proc Natl Acad Sci U
24 Nitzan O, Elias M, Peretz A, et al. Role of antibiotics for S A 2012;109:2108–13.
treatment of inflammatory bowel disease. World J Gastroenterol 37 Arboleya S, Watkins C, Stanton C, et al. Gut bifidobacteria
2016;22:1078–87. populations in human health and aging. Front Microbiol
25 Valdés-­Varela L, Hernández-­Barranco AM, Ruas-­Madiedo P, et al. 2016;7:1204.
Effect of Bifidobacterium upon Clostridium difficile growth and 38 Ferreira-­Halder CV, Faria AVdeS, Andrade SS. Action and function
toxicity when co-­cultured in different prebiotic substrates. Front of Faecalibacterium prausnitzii in health and disease. Best Pract Res
Microbiol 2016;7:738. Clin Gastroenterol 2017;31:643–8.
26 Wei Y, Yang F, Wu Q, et al. Protective Effects of Bifidobacterial 39 Ganesan K, Chung SK, Vanamala J, et al. Causal relationship
Strains Against Toxigenic Clostridium difficile. Front Microbiol between diet-­induced gut microbiota changes and diabetes: a novel
2018;9:888. strategy to transplant Faecalibacterium prausnitzii in preventing
27 Philippe D, Heupel E, Blum-­Sperisen S, et al. Treatment with diabetes. Int J Mol Sci 2018;19. doi:10.3390/ijms19123720. [Epub
Bifidobacterium bifidum 17 partially protects mice from Th1-­driven ahead of print: 22 Nov 2018].
inflammation in a chemically induced model of colitis. Int J Food 40 Greene MW, Roberts AP, Frugé AD. Negative association between
Microbiol 2011;149:45–9. Mediterranean diet adherence and COVID-­19 cases and related
28 Overview of variants in countries. Available: https://covariants.org/​ deaths in Spain and 23 OECD countries: an ecological study. Front
per-country Nutr 2021;8:591964.
29 Tao W, Zhang G, Wang X, et al. Analysis of the intestinal microbiota
41 Tang L, Yin Z, Hu Y, et al. Controlling cytokine storm is vital in
in COVID-­19 patients and its correlation with the inflammatory factor
COVID-­19. Front Immunol 2020;11:570993.
IL-­18. Med Microecol 2020;5:100023.
30 Yagisawa MFP, Hanaki H, Ōmura S. Global trends in clinical 42 Zhao M. Cytokine storm and immunomodulatory therapy in
studies of ivermectin in COVID-­19. Japanese Journal of Antibiotics COVID-­19: role of chloroquine and anti-­IL-­6 monoclonal antibodies.
2021;74:44–94. Int J Antimicrob Agents 2020;55:105982.
31 Kostic AD, Xavier RJ, Gevers D. The microbiome in inflammatory 43 Biagi E, Nylund L, Candela M, et al. Through ageing, and beyond:
bowel disease: current status and the future ahead. gut microbiota and inflammatory status in seniors and centenarians.
PLoS One 2010;5:e10667.

copyright.
Gastroenterology 2014;146:1489–99.
32 Negi S, Das DK, Pahari S, et al. Potential role of gut microbiota in 44 Center for Disease Control. People with certain medical conditions.
induction and regulation of innate immune memory. Front Immunol Available: https://www.cdc.gov/coronavirus/2019-ncov/need-extra-​
2019;10:2441. precautions/people-with-medical-conditions.html

Hazan S, et al. BMJ Open Gastro 2022;9:e000871. doi:10.1136/bmjgast-2022-000871 13

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

Supplementary Table S1. Demographic and SARS-CoV-2 overall clinical characteristics of

SARS-CoV-2-positive patients along with SARS-CoV-2-negative exposed controls and their

exposure relation to SARS-CoV-2 positive individual.

No. Age Sex Ethnicity State Risk Factors BMI PCR Severity

of USA * Result

1 21 F NHW PA none 24.8 positive severe

2 71 M H CA none 18.4 positive severe

3 86 M NHW CA Alzheimer's 21.1 positive severe

4 18 F NHW AZ none 22.0 positive severe

5 31 M NHW CO overweight 26.9 positive severe

6 43 F NHW TX overweight 25.8 positive severe

7 70 F NHW LA overweight 25.8 positive severe

8 56 M NHW CA overweight 27.3 positive severe

9 44 F NHW GA overweight 27.8 positive severe

10 53 M NHW GA obesity 30.7 positive severe

11 37 M NHW FL hypertension 23.8 positive severe

12 60 M NHW CO none 22.7 positive severe

13 69 F NHW CO hypertension 22.9 positive severe

14 55 M NHW NY obesity, HIV 37.2 positive severe

15 67 F NHW IL hypertension, 48.5 positive severe

diabetes mellitus,

obesity, cerebral

aneurysm

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

16 66 M NHW IL hypertension, 33.2 positive severe

obesity, pulmonary

embolus

17 49 F NHW AL overweight 27.4 positive severe

18 60 M NHW CA hypertension, 26.2 positive severe

overweight

19 63 F NHW GA hypertension, 28.1 positive severe

overweight

20 59 F NHW GA hypertension NA positive severe

21 53 F NHW CA none 24.5 positive severe

22 36 M NHW OR obesity 30.7 positive severe

23 10 M NHW HI none 14.5 positive severe

24 52 M NHW HI none 24.1 positive severe

25 55 F NHW HI hypertension, 25.0 positive severe

overweight

26 61 F NHW WA none 23.8 positive severe

27 72 F NHW CA none 19.5 positive severe

28 62 M NHW TN obesity, 31.5 positive severe

hypertension

29 23 F NHW CA obesity 30.9 positive moderat

30 25 F NHW CA Obesity, asthma 30.6 positive moderat

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

31 24 F NHW AZ asthma NA positive moderat

32 36 F H AZ hypertension, 25.4 positive moderat

overweight e

33 33 M H CA none 18.4 positive moderat

34 63 F H CA overweight NA positive moderat

35 65 M NHW TX COPD NA positive moderat

36 75 M NHW CA hypertension, 48.5 positive moderat

severe obesity e

37 67 M NHW CA obesity 33.1 positive moderat

38 70 M NHW GA obesity 30.7 positive moderat

39 57 F NHW TX obesity 32.0 positive moderat

40 69 F NHW NE hypertension 20.9 positive moderat

41 19 M NHW GA none 23.5 positive mild

42 36 F NHW NY none 18.6 positive mild

43 35 F NHW CA overweight 28.5 positive mild

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

44 53 M NHW NE overweight 29.0 positive mild

45 38 F NA TN obesity 31.3 positive mild

46 45 F NHW PA hypertension 23.0 positive mild

47 22 M NHW GA overweight 25.8 positive asympto

matic

48 48 F H CA none 20.4 positive asympto

matic

49 67 M NHW WI overweight 28.5 positive asympto

matic

50 53 M NHW IL obesity, diabetes 35.3 positive asympto

matic

51 67 F NHW WI overweight, 26.6 negative exposed

` husband

52 58 F H PA obesity, + daughter 30.0 negative exposed

53 51 F NHW CA hypertension, 25.0 negative exposed

prediabetes, + son,

+ husband, +

daughter

54 63 M NHW CA overweight, + 26.4 negative exposed

mother

55 29 M NHW CA overweight, 26.4 negative exposed

+ brother

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

56 23 F NHW CA overweight, 26.4 negative exposed

+ brother

57 62 F NHW CA none, (kissed) 17.4 negative exposed

+ boyfriend

58 51 F NHW CA none, + son, 19.9 negative exposed

+ husband

+ daughter

59 16 F NHW CA none, + mom 21.2 negative exposed

60 26 M B CA none, + patients 19.5 negative exposed

61 56 F NHW CA none, + patient 25.5 negative exposed

samples

62 53 F NHW CA none, + workers, + 24.0 negative exposed

patients, + patient

samples

63 35 M NHW KS none, + wife 23.7 negative exposed

64 35 M NHW CA none, + roommate 21.7 negative exposed

65 64 F H TN prediabetic, 26.0 negative exposed

overweight,

+ physician,

+ daughter,

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

+ son-in-law

66 43 F NHW CA overweight, 26.6 negative exposed

+ patient samples

67 45 F NHW CA obesity, + patients 35.1 negative exposed

68 44 M NHW CA overweight, + 27.2 negative exposed

patients, + patient

samples

69 52 M NHW CA overweight, 27.1 negative exposed

+ patients

70 15 F NHW AZ overweight, + 27.1 negative exposed

brother

*BMI, body mass index; + NHW = non-Hispanic White, H = Hispanic, B=Black, NA = Native

American

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

Supplementary Table S2: Detailed SARS-CoV-2 clinical characteristics and typical dietary habits of

subjects.

No. Severity GI Loss of SARS-CoV-2 Symptoms Diet

Symptoms appetite?

1 severe None N Fever, cough , hypoxemia 88%, anosmia Omnivore

2 severe None Y Fever, cough, hypoxemia of 92%, Mexican diet,

anosmia, anorexia omnivore

3 severe None N Fever, cough, hypoxemia of 88%, Omnivore

anosmia

4 severe None N Fever, Tourette's-like symptoms, Omnivore

anosmia

5 severe Diarrhea N Fever, cough, no anorexia, anosmia, Omnivore

hematochezia, dizziness, hypoxemia

92%

6 severe None N Fever, cough, anosmia, hypoxemia 91% Omnivore

7 severe None N Fever, cough, anosmia Omnivore

8 severe Diarrhea, N Pain while eating, dehydration, fever, Omnivore

abdominal anosmia, cough, short of breath,

pain hypoxemia 88%

9 severe None N Fever, cough, short of breath, anosmia, Omnivore

dizziness, anosmia, hypoxemia 89%,

neurological symptoms

10 severe None N Fever, cough, hypoxemia 88%, anosmia Omnivore

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

11 severe Diarrhea N Fever, headache, chills, body pain Omnivore

12 severe Heartburn N Fever, cough, short of breath, hypoxemia Omnivore

88%, Anosmia

13 severe Heartburn N Fever, cough, hypoxemia 88%, anosmia, Omnivore

full body rash

14 severe None Y Fever, chills, chest pain, short of breath, Omnivore

hypoxemia 94%, anorexia, anosmia

15 severe None N Cough, fever, short of breath, hypoxemia Omnivore

88%, rash

16 severe None N Cough, fever, short of breath, hypoxemia Omnivore

88%

17 severe None N Fever, cough, hypoxemia of 89-90%, Omnivore

anosmia

18 severe None N Fever, cough, hypoxemia 88%, anosmia Omnivore

19 severe None N Fever, cough, hypoxemia 88%, anosmia, Omnivore

chest pain

20 severe Diarrhea N Fever, cough, hypoxemia 88%, anosmia Omnivore

21 severe None N Fever, cough, anosmia, oxygen Omnivore

saturation not recorded, short of breath

22 severe None N Fever, cough, anosmia, short of breath Omnivore

23 severe None N Anosmia, high fever, cough, short of Omnivore

breath, hypoxemia 93%

24 severe None N Anosmia, high fever, cough, short of Omnivore

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

breath, hypoxemia 90%

25 severe None N Anosmia, high fever, cough, short of Omnivore

breath, hypoxemia 90%

26 severe None N Cough, anosmia, short of breath, Omnivore

headache, sinus issues, oxygen saturation

not recorded

27 severe None N Anosmia, high fever, cough, short of Omnivore

breath, hypoxemia 92%-93%

28 Severe None N Anorexia, anosmia, cough, fever, short of Omnivore

breath, dizziness, sore throat, fatigue,

hypoxemia 91%

29 moderate Diarrhea N Cough, fever, short of breath, oxygen Omnivore

saturation 98%, went to ER not admitted

30 moderate N Cough, anosmia, oxygen saturation Omnivore

100%, fever

31 moderate Diarrhea N Cold symptoms, sinusitis, rhinorrhea, Omnivore

fever, cough, anosmia, stable oxygen

saturation

32 moderate None N Chest pain, fever, severe Mexican diet,

headaches/migraines, body aches, chest omnivore

pressure, anosmia, oxygen saturation

100%

33 moderate Nausea N Rhinorrhea, congestion, body ache, fever Mexican diet,

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

of 101°, anosmia, oxygen saturation 99% omnivore

34 moderate None N Anosmia, anorexia, fatigue, cough, Omnivore

oxygen saturation 98%

35 moderate Abdominal N Chest tightness, headache, head Omnivore

pain congestion, cough, oxygen saturation

100%

36 moderate None N Sluggish, cough, nasal congestion, Omnivore

oxygen saturation 98%

37 moderate None N Myalgia, backache, oxygen saturation Omnivore

99%

38 moderate None N Fever, cough, anosmia, oxygen Omnivore

saturation 95%

39 moderate None N rhinorrhea, anosmia, mild cough oxygen Omnivore

saturation 98%

40 moderate None N Cough, headache, oxygen saturation 96% Omnivore

41 mild None N Nasal congestion, oxygen saturation Omnivore

100%

42 mild Bloating N Fever, cough, anosmia, rhinorrhea, Omnivore

oxygen saturation 100%

43 mild None N Congestion, sinusitis, anosmia, oxygen Omnivore

saturation 100%

44 mild None N Fatigue, short of breath, anosmia, oxygen Omnivore

saturation 100%

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

45 mild None N Cough, fatigue, dizziness, sore throat, Omnivore

myalgia, oxygen saturation 100%

46 mild None N Cough, head congestion, oxygen Omnivore

saturation 98%

47 asymptomatic None N None Omnivore

48 asymptomatic None N None Mexican diet,

omnivore

49 asymptomatic None N None Omnivore

50 asymptomatic None N None Armenian diet,

omnivore

51 Ctrl None N None Omnivore

52 Ctrl None N None Omnivore

53 Ctrl None N None Omnivore

54 Ctrl None N None Kosher,

omnivore

55 Ctrl None N None Kosher,

omnivore

56 Ctrl None N None Kosher,

omnivore

57 Ctrl None N None Omnivore

58 Ctrl None N None Kosher,

omnivore

59 Ctrl None N None Kosher,

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871

 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance
Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Open Gastro

omnivore

60 Ctrl None N None Omnivore

61 Ctrl None N None Omnivore

62 Ctrl None N None Omnivore

63 Ctrl None N None Omnivore

64 Ctrl None N None Omnivore

65 Ctrl None N None Omnivore

66 Ctrl None N None Omnivore

67 Ctrl None N None Omnivore

68 Ctrl None N None Omnivore

69 Ctrl None N None Kosher,

omnivore

70 Ctrl None N None Not available

Hazan S, et al. BMJ Open Gastro 2022; 9:e000871. doi: 10.1136/bmjgast-2022-000871