RESEARCH ARTICLE
Dysbiosis and compositional alterations with
aging in the gut microbiota of patients with
heart failure
Takehiro Kamo
1
, Hiroshi Akazawa
1
*, Wataru Suda
2,3
, Akiko Saga-Kamo
1
, Yu Shimizu
1
,
Hiroki Yagi
1
, Qing Liu
1
, Seitaro Nomura
1
, Atsuhiko T. Naito
1
, Norifumi Takeda
1
,
Mutsuo Harada
1
, Haruhiro Toko
1
, Hidetoshi Kumagai
1,4
, Yuichi Ikeda
1
, Eiki Takimoto
1
,
Jun-ichi Suzuki
4
, Kenya Honda
3,5
, Hidetoshi Morita
6
, Masahira Hattori
2,7
, Issei Komuro
1
*
1 Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo,
Japan, 2 Graduate School of Frontier Sciences, The University of Tokyo, Kashiwa, Japan, 3 Department of
Microbiology and Immunology, Keio University School of Medicine, Tokyo, Japan, 4 Department of Advanced
Clinical Science and Therapeutics, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan,
5 RIKEN Center for Integrative Medical Sciences, Yokohama, Japan, 6 Graduate School of Environmental
and Life Science, Okayama University, Okayama, Japan, 7 Graduate School of Advanced Science and
Engineering, Waseda University, Tokyo, Japan
*
akazawah-tky@umin.ac.jp (HA); komuro-tky@umin.ac.jp (IK)
Abstract
Emerging evidence has suggested a potential impact of gut microbiota on the pathophysiol-
ogy of heart failure (HF). However, it is still unknown whether HF is associated with dysbio-
sis in gut microbiota. We investigated the composition of gut microbiota in patients with HF
to elucidate whether gut microbial dysbiosis is associated with HF. We performed 16S ribo-
somal RNA gene sequencing of fecal samples obtained from 12 HF patients and 12 age-
matched healthy control (HC) subjects, and analyzed the differences in gut microbiota. We
further compared the composition of gut microbiota of 12 HF patients younger than 60 years
of age with that of 10 HF patients 60 years of age or older. The composition of gut microbial
communities of HF patients was distinct from that of HC subjects in both unweighted and
weighted UniFrac analyses. Eubacterium rectale and Dorea longicatena were less abun-
dant in the gut microbiota of HF patients than in that of HC subjects. Compared to younger
HF patients, older HF patients had diminished proportions of Bacteroidetes and larger quan-
tities of Proteobacteria. The genus Faecalibacterium was depleted, while Lactobacillus was
enriched in the gut microbiota of older HF patients. These results suggest that patients with
HF harbor significantly altered gut microbiota, which varies further according to age. New
concept of heart-gut axis has a great potential for breakthroughs in the development of
novel diagnostic and therapeutic approach for HF.
Introduction
In the human gut, there are more than 10
14
bacterial cells, which exceed the number of human
cells in the body. Their combined genomes contain millions of genes, which are hundred
PLOS ONE | https://doi.org/10.1371/journal.pone.01740 99 March 22, 2017 1 / 14
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Kamo T, Akazawa H, Suda W, Saga-Kamo
A, Shimizu Y, Yagi H, et al. (2017) Dysbiosis and
compositional alterations with aging in the gut
microbiota of patients with heart failure. PLoS ONE
12(3): e0174099.
https://doi.org/10.1371 /journal.
pone.0174099
Editor: Yasuko Bando, Nagoya University, JAPAN
Received: November 20, 2016
Accepted: March 4, 2017
Published: March 22, 2017
Copyright: © 2017 Kamo et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
files.
Funding: This work was supported in part by
grants from Japan Society for the Promotion of
Science (KAKENHI 26670395,
https://kaken.nii.ac.
jp/ja/grant/KAKENHI-PROJECT-26670 395/
) to H.A.,
and Core Research for Evolutionary Medical
Science and Technology (CREST) from Japan
Agency for Medical Research and Development
(AMED) (M1501,
http://www.amed.go.jp/program/
list/01/07/023_02.html) to K.H., H.M. H.A. has
times the number of human genes. These large quantities of gene products complement host
metabolism and facilitate the development of host immune system [
1, 2]. In line with the cru-
cial link between gut microbiota and the maintenance of host health, there is growing evidence
that altered composition of gut microbiota, known as dysbiosis, contributes to the pathogene-
sis of host diseases [
3]. Numerous experiments with fecal microbiota transplantation to germ-
free animals have suggested that gut microbiota can initiate and influence host diseases such as
obesity-related diseases, liver diseases, inflammatory bowel diseases, and colorectal cancer [
4].
In patients with heart failure (HF), the structure and function of the gut are altered as a con-
sequence of microcirculatory disturbances [
5, 6]. Impaired epithelial absorption may have det-
rimental effect on nutritional status of patients with HF, and disruption of epithelial barrier
may lead to translocation of microbial products into systemic circulation, possibly aggravating
HF by inducing systemic inflammatory responses [
7–10]. Indeed, compared with healthy con-
trol (HC) subjects, patients with HF showed increases in the quantity of pathogenic bacteria in
feces and the density of bacteria adhered to colon mucosa [
5, 6], in association with an increase
in intestinal permeability [
5]. Gut microbe-derived metabolites such as indoxyl sulfate and tri-
methylamine N-oxide (TMAO) may also contribute to the pathogenesis of HF through unde-
fined mechanisms [
11–13]. Therapeutic management of HF through manipulating gut
microbiota is under investigation in animal models. For example, oral administration of anti-
biotics or probiotics to rats has been reported to reduce myocardial infarct size in ischemia-
reperfusion injury and to attenuate cardiac remodeling after myocardial infarction [
14, 15].
These observations suggest a significant impact of gut microbiota on the pathophysiological
process of HF. However, it is unclear whether dysbiosis in gut microbiota is associated with
HF.
To address this issue, we analyzed the gut microbiome of HF patients and HC subjects
using 16S ribosomal RNA (rRNA) gene sequencing. Our data revealed the presence of dys-
biosis in the gut microbiota of patients with HF. Moreover, the gut microbiota composition
of older HF patients differed from that of younger HF patients. Our studies provide new
insights into the heart-gut axis in the pathophysiology of HF, and pave the way toward
exploring the potential of manipulating gut microbiota as a future therapeutic strategy
against HF.
Materials & methods
Study population
We recruited a total of 22 patients with HF (New York Heart Association functional class II to
IV) who were hospitalized at the University of Tokyo Hospital. All patients were hospitalized
for acute decompensated HF or acute exacerbation of chronic HF. These HF patients were
classified into 2 groups according to age, those younger than 60 years of age (n = 12, aged
47.4 ± 2.8 years, 11 men and 1 woman) and those 60 years of age or older (n = 10, aged
73.8 ± 2.8 years, 7 men and 3 women). We excluded the patients with clinical signs of active
infection, chronic inflammatory diseases, malignancy, renal failure requiring renal replace-
ment therapy, or a history of gastrointestinal surgery. In addition, exclusion criteria included
receiving antibiotic, probiotic, steroid, or immunosuppressive therapy during the previous 2
months. Twelve age-matched healthy volunteers (aged 41.4 ± 2.0 years, 9 men and 3 women)
were recruited as HC subjects at Azabu University. Clinical characteristics of all subjects are
listed in
S1–S3 Tables. This study complies with the Declaration of Helsinki, and was approved
by the Research Ethics Committee, Graduate School of Medicine and Faculty of Medicine,
The University of Tokyo and the Human Research Ethics Committee of Azabu University.
The written informed consent was obtained from all of the subjects.
Gut microbial dysbiosis in heart failure
PLOS ONE | https://doi.org/10.1371/journal.pone.01740 99 March 22, 2017 2 / 14
received research funding from Takeda
Pharmaceutical Co., Ltd., Daiichi Sankyo Co., Ltd.,
Shionogi & Co., Ltd., Nippon Boehringer Ingelheim
Co., Ltd., Bristol-Myers Squibb K.K., MSD K.K.,
Sanofi K.K., and Sumitomo Dainippon Pharma Co.,
Ltd. The funders had no role in study design, data
collection and analysis, decision to publish, or
preparation of the manuscript. This does not alter
our adherence to PLOS ONE policies on sharing
data and materials.
Competing interests: The study was supported in
part by Takeda Pharmaceutical Co., Ltd., Daiichi
Sankyo Co., Ltd., Shionogi & Co., Ltd., Nippon
Boehringer Ingelheim Co., Ltd., Bristol-Myers
Squibb K.K., MSD K.K., Sanofi K.K., and Sumitomo
Dainippon Pharma Co., Ltd. There are no patents,
products in development, or marketed products to
declare.
Fecal sample collection and bacterial DNA extraction
Fecal samples were freshly collected and transported to the laboratory under anaerobic condi-
tion in AnaeroPack (Mitsubishi Gas Chemical Company, Inc., Tokyo, Japan) at 4˚C. The fecal
samples were frozen by liquid nitrogen in phosphate-buffered saline containing 20% glycerol,
and stored at -80˚C until use. Bacterial DNA was extracted from the fecal samples by enzy-
matic lysis method using lysozyme (Sigma-Aldrich Co., St. Louis, Missouri) and achromopep-
tidase (Wako Pure Chemical Industries, Ltd., Osaka, Japan), as previously described [
16, 17].
Sequencing of 16S ribosomal RNA gene amplicons
Bacterial DNA from the fecal samples was amplified by PCR, as previously described [
17].
Primers 27Fmod and 338R with adaptor sequences for 454 pyrosequencing were used to
amplify the bacterial 16S rRNA gene V1-V2 region. PCR was run for 25 cycles, using Ex Taq
polymerase (Takara Bio Inc., Kusatsu, Japan). PCR amplicons were purified by AMPure XP
magnetic purification beads (Beckman Coulter, Inc, Brea, California) and quantified using the
Quant-iT PicoGreen dsDNA Assay Kit (Thermo Fisher Scientific Inc., Waltham, Massachu-
setts). Equal amount of amplicons from each sample were sequenced with 454 GS FLX Tita-
nium or 454 GS JUNIOR (Roche Applied Science, Indianapolis, Indiana) according to the
manufacturer’s instructions.
Data analysis
The previously established analysis pipeline was utilized for data analysis [
17]. Filter-passed
3,000 reads, with an average quality score of 25 or higher, were randomly selected from the
reads for each sample. The number of operational taxonomic units (OTUs) in each sample
was calculated by clustering the 3,000 reads at a 96% identity threshold. The richness and
diversity of microbial communities in each sample were evaluated by Chao1-estimated OTU
number and Shannon index respectively. For taxonomic assignment, the read sequences were
aligned against the 16S rRNA gene database constructed from RDP, CORE, and NCBI genome
databases, and were assigned to taxonomic groups at a 96% identity threshold. Taxonomic
groups with relative abundance in any subject above 0.1% were included in the analysis. Uni-
Frac analysis was used to calculate phylogenetic tree-based distances between microbial com-
munities of the individuals [
18].
Statistical analysis
Data are presented as mean ± SEM. The unpaired Student t test was used to evaluate the
between-group differences. Values of p < 0.05 were considered statistically significant.
Results
Gut microbiota in patients with heart failure and healthy control subjects
We performed 16S rRNA gene sequencing of fecal samples from 12 younger HF patients
(younger than 60 years of age) and 12 age-matched HC subjects. Gut microbial richness in the
given individual was measured by Chao1-estimated OTU number, and gut microbial diversity
in the individual was evaluated by Shannon index. The richness and diversity of gut micro-
biota were similar between the samples from HF patients and HC subjects (Chao1-estimated
OTU number: 191 ± 20 vs. 195 ± 12, Shannon index: 3.38 ± 0.19 vs. 3.48 ± 0.06) (
Fig 1A and
1B
). We next estimated the distances between fecal samples obtained from the individuals
using UniFrac analysis [
18, 19]. UniFrac distances between gut microbial communities of the
individuals were visualized by a scatter plot created by Principal Coordinate Analysis (PCoA).
Gut microbial dysbiosis in heart failure
PLOS ONE | https://doi.org/10.1371/journal.pone.01740 99 March 22, 2017 3 / 14
Fig 1. Richness, diversity, and UniFrac distances of gut microbiota in heart failure patients and healthy control
subjects. Chao1-estimated operational taxonomic unit (OTU) number (A) and Shannon index (B) of gut microbiota
samples obtained from younger heart failure (HF-Y) patients and healthy control (HC) subjects. Unweighted UniFrac
analysis (C, D) and weighted UniFrac analysis (E, F) of gut microbiota samples obtained from HF-Y patients and HC
subjects. Principal Coordinate Analysis (PCoA) of UniFrac distances between gut microbial communities of the individuals
Gut microbial dysbiosis in heart failure
PLOS ONE | https://doi.org/10.1371/journal.pone.01740 99 March 22, 2017 4 / 14
Unweighted UniFrac is a qualitative measure that reflects inter-individual differences in the
presence or absence of each taxon. Weighted UniFrac is a quantitative measure that reflects
inter-individual differences in the relative abundance of each taxon. Mean unweighted and
weighted UniFrac distances between gut microbiota of HF patients and HC subjects were
0.750 and 0.432, respectively, which were greater than inter-individual UniFrac distances in
the gut microbiota of HC subjects (0.722 and 0.383, respectively, both p < 0.00001) (
Fig 1C–
1F). Accordingly, the composition of gut microbial communities of HF patients was distinct
from that of HC subjects in both unweighted and weighted UniFrac analyses. These data also
showed greater inter-individual diversity in the gut microbiota of HF patients compared to
HC subjects (Fig 1C–1F).
To investigate whether HF patients had significant changes in specific taxonomic groups of
gut microbial communities, we analyzed the relative abundances of 16S rRNA reads assigned
to each phylum, genus, or species. The majority of gut microbiota was dominated by the four
phyla, Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria. Significant differences
were not observed between the samples from HC subjects and HF patients in terms of relative
abundances of respective phyla (Firmicutes: 55.8 ± 3.2% vs. 59.4 ± 3.4%, Bacteroidetes:
27.0 ± 3.9% vs. 21.7 ± 4.1%, Actinobacteria: 14.7 ± 2.7% vs. 16.1 ± 4.1%, Proteobacteria:
1.3 ± 0.4% vs. 1.6 ± 0.5%) (
Fig 2A). Taxonomic assignment performed at the genus level dem-
onstrated that Clostridium and Dorea were less abundant in the gut microbiota of HF patients
than in that of HC subjects (5.1 ± 1.1% vs. 10.1 ± 2.0%, p = 0.040, and 0.9 ± 0.2% vs.
1.8 ± 0.3%, p = 0.039, respectively) (
Fig 2B). At the species level, Eubacterium rectale and
Dorea longicatena were significantly reduced in the samples from HF patients compared to
HC subjects (1.2 ± 0.7% vs. 3.8 ± 0.9%, p = 0.032, and 0.6 ± 0.2% vs. 1.4 ± 0.3%, p = 0.031,
respectively) (Fig 2C).
Gut microbiota in younger and older patients with heart failure
Given that clinical characteristics and outcomes of HF are influenced by aging process [
20], we
next examined whether gut microbial communities of patients with HF varied according to
age. We sequenced 16S rRNA gene amplicons from additional fecal samples obtained from 10
HF patients who were 60 years of age or older. We then compared the composition of gut
microbiota of younger HF patients (younger than 60 years of age; n = 12) with that of older
HF patients (60 years of age or older; n = 10). The richness and diversity of gut microbial com-
munities within the individual, as evaluated by Chao1-estimated OTU number and Shannon
index respectively, were not significantly different between younger and older patients with
HF (191 ± 20 vs. 178 ± 13, and 3.38 ± 0.19 vs. 3.21 ± 0.11, respectively) (
Fig 3A and 3B). How-
ever, both unweighted and weighted UniFrac analyses demonstrated that the differences in gut
microbiota composition between the two groups were larger than inter-individual differences
in the gut microbiota of older HF group (unweighted UniFrac distance: 0.743 ± 0.004 vs.
0.716 ± 0.007, weighted UniFrac distance: 0.490 ± 0.006 vs. 0.443 ± 0.011; both p < 0.01) (
Fig
3C–3F
). Interestingly, unweighted but not weighted UniFrac analysis revealed greater inter-
individual diversity in the gut microbiota of younger HF patients compared to older patients
(
Fig 3C–3F). The phylum Bacteroidetes was less abundant (11.7 ± 2.3% vs. 21.7 ± 4.1%,
p = 0.047) whereas Proteobacteria was more abundant (8.4 ± 2.9% vs. 1.6 ± 0.5%, p = 0.046) in
the gut microbiota of older HF patients than in that of younger patients (
Fig 4A). At the genus
and species level, the genus Faecalibacterium, F. prausnitzii, and Clostridium clostridioforme
(C, E), and UniFrac distances between gut microbial communities of the individuals within each group and between the two
groups (D, F). Data are presented as mean ± SEM. NS, not significant. * p < 0.05, ** p < 0.00001.
https://doi.org/10.1371 /journal.pone.0174099.g001
Gut microbial dysbiosis in heart failure
PLOS ONE | https://doi.org/10.1371/journal.pone.01740 99 March 22, 2017 5 / 14