Alterations of gut microbiota contribute to the progression of unruptured intracranial aneurysms - Nature
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ARTICLE
https://doi.org/10.1038/s41467-020-16990-3 OPEN
Alterations of gut microbiota contribute to the
progression of unruptured intracranial aneurysms
Hao Li 1,8, Haochen Xu1,8, Youxiang Li2,8, Yuhua Jiang2,8, Yamin Hu3, Tingting Liu1, Xueqing Tian1,
Xihai Zhao 4, Yandong Zhu 4, Shuxia Wang5, Chunrui Zhang6, Jing Ge1, Xuliang Wang1, Hongyan Wen1,
Congxia Bai1, Yingying Sun1, Li Song1, Yinhui Zhang1, Rutai Hui1, Jun Cai7 & Jingzhou Chen 1 ✉
1234567890():,;
Unruptured intracranial aneurysm (UIA) is a life-threatening cerebrovascular condition.
Whether changes in gut microbial composition participate in the development of UIAs
remains largely unknown. We perform a case-control metagenome-wide association study in
two cohorts of Chinese UIA patients and control individuals and mice that receive fecal
transplants from human donors. After fecal transplantation, the UIA microbiota is sufficient
to induce UIAs in mice. We identify UIA-associated gut microbial species link to changes in
circulating taurine. Specifically, the abundance of Hungatella hathewayi is markedly decreased
and positively correlated with the circulating taurine concentration in both humans and mice.
Consistently, gavage with H. hathewayi normalizes the taurine levels in serum and protects
mice against the formation and rupture of intracranial aneurysms. Taurine supplementation
also reverses the progression of intracranial aneurysms. Our findings provide insights into a
potential role of H. hathewayi-associated taurine depletion as a key factor in the pathogenesis
of UIAs.
1 StateKey Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking
Union Medical College, Beijing 100037, China. 2 Department of Interventional Neuroradiology, Beijing Neurosurgical Institute and Beijing Tiantan Hospital, Capital
Medical University, Beijing 100050, China. 3 Department of Cardiology, Cangzhou Central Hospital, Cangzhou 061000, China. 4 Department of Biomedical
Engineering, School of Medicine, Tsinghua University, Beijing 100084, China. 5 Chinese PLA General Hospital and Chinese PLA Medical College, Beijing 100853,
China. 6 Novogene Bioinformatics Institute, Beijing 100083, China. 7 Hypertension Center, Fuwai Hospital, State Key Laboratory of Cardiovascular Disease of
China, National Center for Cardiovascular Diseases of China, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100037, China.
8
These authors contributed equally: Hao Li, Haochen Xu, Youxiang Li, Yuhua Jiang. ✉email: chendragon1976@aliyun.com
NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications 1
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3
U
nruptured intracranial aneurysms (UIAs) are being heterogeneous community structure among UIA patients than
detected more often by cross-sectional imaging techniques among controls.
and are an important healthcare burden. Findings from It is noteworthy that 145 genera were differentially enriched in
an analysis of 68 studies reporting data from 83 study populations the UIA or control groups (Supplementary Data 4). Bacteroides,
and 1450 UIAs in 94912 patients reveal an overall prevalence of Parabacteroides, Ruminococcus, and Blautia were significantly
intracranial aneurysms of 3.2%1. Rupture of an intracranial overrepresented in UIA patients, while Faecalibacterium, Eubac-
aneurysm is the underlying cause of 80–85% of non-traumatic terium, Collinsella, and Lactobacillus were overrepresented in the
subarachnoid hemorrhages2, often leading to the loss of many controls (Fig. 1f). To further evaluate whether regional factors are
years of productive life. However, the precise mechanisms that potential confounders affecting the features of the gut microbiota,
contribute to the pathogenesis of UIAs remain to be elucidated. PCoA based on the features of the gut microbiota at the genus
Although studies of the genetic predisposition for UIAs have level was applied to discriminate control individuals from
implicated genes such as CDKN2A, SOX17, and ADAMTS153, a different areas. As expected, PCoA at the genus level showed a
large population-based heritability study estimated the heritability similar spatial distribution between the two groups (Supplemen-
of intracranial aneurysms and subarachnoid hemorrhages to be tary Fig. 2), indicating that control individuals from different
41%4, suggesting that environmental factors also contribute. areas shared similar features of gut microbiota.
Accumulating evidence indicates that the gut microbiota is a
pivotal environmental factor that influences host metabolism and
immune homeostasis. Recently, considerable interest has been Gut microbial species associated with UIAs. To further identify
focused on the role of the human gut microbiota in cardiovas- the microbial species associated with UIAs, MetaPhlAn2 (meta-
cular diseases such as hypertension5, heart failure6, and athero- genomic phylogenetic analysis)11,12 was run to identify the
sclerosis7. Studies have shown that, in these diseases, the gut taxonomic abundances at the species level. The α-diversity at the
microbiota produce numerous metabolites that are absorbed into species level did not significantly differ between the two groups
the systemic circulation, where they are further metabolized by (P = 0.824; Wilcoxon rank-sum test; Supplementary Fig. 3). In
host enzymes, leading to target organ damage8,9. A recent study the first cohort, a total of 47 species differed significantly in
reported that depletion of the gut microbiota by antibiotics abundance between the UIA and control samples (Supplementary
reduces the incidence of intracranial aneurysms in mice10. Data 5), 38 of which were more abundant in the UIA samples. A
Nevertheless, direct evidence for altered gut microbial composi- bacterial community of Bacteroides sp., such as B. thetaiotaomi-
tion in UIA patients and the precise underlying mechanisms are cron, B. massiliensis, B. nordii, B. intestinalis, and B. cellulosily-
still lacking. ticus, was significantly enriched in the UIA patients compared
Metagenome-wide association studies (MWAS) can provide with the controls. The abundance of Odoribacter splanchnicus,
valuable information about the presence of characteristic gut which has been reported to show a positive correlation with the
microbiota. In the present study, we sequence stool samples, consumption of red meat and a negative correlation with the
representative of the gut microbiota, from healthy controls and consumption of fruits and vegetables13, was higher in UIA
UIA patients and perform a MWAS. We then perform serum patients. Furthermore, we found that the abundance of Clos-
metabolomic analyses of the participants’ metabolic profiles and tridium sp., including C. bartlettii, C. nexile, and C. bolteae, was
find UIA-associated microbial species and explore their effects also significantly enriched in the UIA patients. However, C.
on host amino acid and fatty acid metabolism. Using a mouse hathewayi, which was reclassified as H. hathewayi in 201414, was
model of fecal transplantation, we uncover evidence for the the only Clostridium sp. enriched in the control group. Among
potential effects of altered Hungatella hathewayi abundance on the 47 differentially abundant species, 8 (17.0%) were also dif-
circulating taurine levels and on the increased occurrence ferentially enriched in UIA patients compared with controls in
of UIAs. the second cohort (Fig. 2 and Supplementary Data 6).
Moreover, we clustered the genes that displayed significant
differences in abundance between the two groups into metage-
Results nomic linkage groups (MLGs). A total of 220 of the MLGs
UIA-associated genes and taxonomic changes identified by differed significantly in abundance between the UIA and control
MWAS. To investigate the gut microbiota in UIA patients, we samples (adjusted P < 0.05, Wilcoxon rank-sum test; Supplemen-
performed metagenomic shotgun sequencing on a total of 280 tary Data 7), 137 of which were more abundant in the UIA
fecal samples (200 samples from 100 UIA patients and 100 con- samples.
trols in the first cohort; 80 from 40 UIA patients and 40 controls In addition to abundance differences between the UIA and
in the second cohort, Supplementary Fig. 1). For each sample, a control samples, the MLGs also showed differences in network
majority of high-quality sequencing reads were assembled de novo structure, which was constructed by SparCC (Sparse Correlations
into long contigs or scaffolds, which were used for gene prediction, for Compositional data) (Supplementary Fig. 4). We found 256
taxonomic classification, and functional annotation (Supplemen- and 171 positive significant correlations in UIA and control-
tary Data 1–3). enriched MLGs, respectively (correlation coefficient > 0.5 and
We first investigated the richness and evenness of the gut Benjamin–Hochberg corrected P < 0.01). Five negative significant
microbiota in the first cohort. Rarefaction analysis was used to correlations were also found among MLGs (correlation coeffi-
estimate the total number of genes that could be identified from cient < −0.5 and Benjamin–Hochberg corrected P < 0.01).
these samples; this showed that the gene richness approached Moreover, we assessed the divergence of gut microbiota
saturation in each group (Fig. 1a). Neither genus counts nor α- composition to explore the correlation of microbial abundance
diversity significantly differed between the two groups (P = 0.248 with host factors and found no remarkable regularity of bacterial
for counts, P = 0.276 for diversity; Wilcoxon rank-sum test; abundance based on age, BMI, sex, smoking, or alcohol consump-
Fig. 1b, c). Ordination of Bray–Curtis dissimilarity by principal tion (Supplementary Fig. 5). We further selected 61 of the 220
coordinate analysis (PCoA) revealed separation of the two groups MLGs to distinguish between individuals with and without UIAs,
(Fig. 1d). This separation was associated with differences in and the accuracy of cross-validation reached 81%, indicating that
community composition as evaluated by analysis of similarity these 61 MLGs were distinct UIA-associated features of the gut
(ANOSIM) (R = 0.061, P = 0.001; Fig. 1e), which indicated a less microbiota (Supplementary Fig. 6, Supplementary Data 8). It is
2 NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3 ARTICLE
a Rarefaction of genes d e P = 0.001
2.5 × 106 0.20 20,000
Bray-Curtis distances (β-diversity)
Control
UIA 0.15
2 × 106
15,000
Number of genes
PCoA (6.96%)
0.10
1.5 × 106
0.05
10,000
6
1 × 10 0.00
–0.05
0.5 × 106 5000
–0.10 UIA
0 Control
–0.15 0
1 10 20 30 40 50 60 70 80 90 100 –0.3 –0.2 –0.1 0.0 0.1
Control UIA
PCoA (7.94%)
b P = 0.248 c P = 0.276 f 0 Control
Relative abundance (log10)
1200 2.70 UIA
P < 0.001
P = 0.004
P < 0.001
P = 0.003
P < 0.001
Shannon index (α-diversity)
P < 0.001
2.50 –2
1000
P = 0.005
P < 0.001
P < 0.001
2.25
P = 0.002
–4
P = 0.003
Genus counts
P = 0.002
800 2.00
P = 0.002
P < 0.001
P < 0.001
P < 0.001
P = 0.002
P < 0.001
P < 0.001
–6
P < 0.001
1.75
600
1.50 –8
400
Bu coc a
so illo a
u a
no io
ys c lla
ra te era
ty ac a
Ac oco lla
pl s
oc es
ct m
M coc s
La suo us
ho my s
La yriv us
m m
as
La ollin um
um er s
C B us
1.25
le ce
i u
Ac o u
R act ide
i
Fu Ve spir
Kl sm
Bu tob ell
ro ut
ch ibr
ba riu
no riu
D ba ne
ct ke
in oid
ric ill
it c
tin cc
t c
on
c
C eri
Pa ac yv
s
oc
op la
b ro
a
Eu cte
go te
1.00
a
200 ib
c
al
B
ec
Control UIA Control UIA
Fa
Fig. 1 Gut microbial alterations in UIA patients. a Rarefaction curves for gene number in controls (n = 100) and UIA patients (n = 100) after 100 random
samplings. b, c Comparison of microbial genus counts and α-diversity (as assessed by the Shannon index) based on the genus profiles in the two groups.
Interquartile ranges (IQRs; thick bars), medians (open dots on the bars), the lowest and highest values within 1.5 times IQR from the first and third quartiles
(lines above and below the bars). d Principal coordinate analysis of samples from UIAs and controls. e β-diversity (as assessed by the Bray–Curtis
distances) based on the genus profiles in the two groups (n = 100). f Relative abundances of the most abundant genera that showed significant differences
between UIA patients and controls. For (a), (e), and (f), boxes represent the IQRs between the first and third quartiles, and the line inside the box
represents the median; whiskers represent the lowest or highest values within 1.5 times IQR from the first or third quartiles. For (b), (c), and (f), the two-
tailed Wilcoxon rank-sum test was used. For (e), ANOSIM analysis was performed. Source data are provided as a Source Data file.
worth noting that the discrimination between UIAs and control for a dysbiotic gut microbiota and highlights the potential of gut
samples for these 61 MLGs was further validated in the second microbiota biomarkers for the noninvasive detection of popula-
cohort. Seventeen MLGs (27.9%) were also differentially enriched in tions with UIAs.
the UIA patients compared with the controls (Supplementary
Fig. 7).
Notably, among the 8 MetaPhlAn2-associated species and the Functional characterization of the UIA microbiome. We next
17 MLG-associated species that differed in abundance between aimed to characterize the species-level functional profiling in UIA
the two cohorts, we found that H. hathewayi, O. splanchnicus and through the use of HUMAnN215 (the HMP Unified Metabolic
Alistipes putredinis considerably overlapped; these were the most Analysis Network 2) pipeline. This allows the assessment of
representative microbial changes in the deteriorating gut micro- molecular activities of microbial communities at the whole-
biota of the UIA patients. pathway level, which provides a more efficient and accurate
profiling of abundance in microbial pathways from a community
based on metagenomic data. We identified 46 metabolic pathways
Identification of UIAs based on gut microbiota. To illustrate that were differentially abundant between UIA patients and
the microbial signature and explore the diagnostic value of the controls (Fig. 4a, Supplementary Data 9). Within these 46 path-
composition of the gut microbiota in relation to UIA, we con- ways, those contributing to amino acid and fatty acid metabolism
structed a random forest classifier from the 200 UIA and control were adequately represented. Specifically, the microbiome of UIA
samples of the first cohort. Four-fold cross-validation and recei- patients was found to be significantly dominated by unsaturated
ver operating characteristic curves for distinguishing UIA patients fatty acid biosynthesis. Moreover, the biosynthesis of amino acids
from controls were developed. We were able to detect UIA such as threonine, isoleucine, lysine, and methionine dominated
patients accurately based on the 47 species that were differentially the metabolic landscape of the microbiome in controls. The rest
abundant (as identified by MetaPhlAn2) in the first cohort, as of the pathways were mainly represented by glucose and
indicated by an area under the receiver operating curve (AUC) of nucleotide metabolic pathways.
up to 0.86 and a 95% confidence interval (CI) of 0.81–0.91 The functional characterization of the gut microbiome was
(Fig. 3a). Consistent with these results, the classification error further validated by using the KEGG database based on MLGs. We
remained relatively low in the second cohort (80 UIA and control identified 25 KEGG orthologues that differed in abundance
samples), with an AUC of 0.72 and a 95% CI of 0.65–0.79 between UIA patients and controls (adjusted P < 0.05, Wilcoxon
(Fig. 3b), indicating that information on the gut microbiota might rank-sum test; Supplementary Data 10). Notably, unsaturated fatty
be used to identify UIA patients. Overall, the presence of UIA- acid biosynthesis, amino acid (methionine, tryptophan, pyruvate;
associated features in the gut microbiota offers further evidence and isoleucine) metabolism and glycolysis were overlapping
NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications 3
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3
Control UIA The first cohort
6
(200 samples)
5
4
3
2
Relative abundance (Log2)
1
0
6 The second cohort
(80 samples)
5
4
3
2
1
0
* * * * * * * *
s_ s_A a_fo roi liva _2_
ae s_ illus 5_1 phi a
C Cl _u _c ae
ea c ore llu stin m
id ino e ov e
_b sti taio s_ ae
er o c fa le
sp Str s es s_i _ob ns
s_ Ba _m chia ale
Ba s_ s_b sw fra s
R se id as es
s_ e_b cc a_l osil alis
ra act ide s_ sis
er st te e s
ba noc ist ol is
H riu cu nv s
s_ _C ostr a_h trio s
ch ro ium he m
ic s
sp us
ce s_ s_thcoc _me tus
oi _w id tor us
de s um is
s_ ga _v gn us
_b Eu _u _6 lus
ra er ic _n ius
de ha s
s_ riba hila 1_1 rdo n
ch s ter acte las FAA
s_ s_ occ act artle A
s_ _Su s_ oco pla ss A
s_ Rum _D yla tred h8
25
e te _sa um i
um e n 57 ii
s_ s_ hia cus olte i
Ba ac no ae _n d
er e er i
s_ s_ te o lie li
ct re cte us hn d
s_ pir ost _3 m_ ed
ct _R te ide cac ei
pt e m_ 6F i
O Bil ium nd ic i
ct s_E eriu lla_ itzii
s_ es eri ect i
le d s_ ue
ct Ali ac orth gili
s_ cte oc er_i ven
s l ll n vu
op ct _t g cu
oi s_ an
c _ o n
i i
ro um roid s_ ca
y
ct tip en rdi
re ea iu _4 lli
s_ s_ ter s_o om mi
r
_ e ro
Eu i ial nis in
oi ipe ri ns
D Ba us er tt
ct ter coc ro exi
R ct _u _ nk
o
oc _b b A
in r_s cla FA
ira ep _D _c nte eu
Es op id at su
s_ s_B mi lla_ um sifie
Ba tte Ba cc nc ifie
oi ch _r op
er co _b wa
oi ide us cie
s_ ollin ostr ncl om
s_ dr
er lla ro _ ic
un m s_ is
ba um erm ate
ae Ali e cu rd
Ba s bac ero s_ dor
Pa B ro ide n
St ac rid _1 ha
s _x u p
r
Ba ct as _c
s a
_D
u fi
C te e a
C a us on yti
da a e q
ac pe ta bro
ba os ig o
ct te sn
s _p _
no _C ium ri si
er s m c
ac b nc 3
i
ba evo au
s
s
Pa na rm de
Eu Pr pr
l
i
s_ s_ m_
h
ic c
e
e r
riu
or c
d
ro ri
c
te
do op
e
d
ac
o
l
e
u
ib
t o
Ba
al
s
s_
e
ec
s_
s_
ce
Fa
c
s
no s_
La
ira
ira
s_
s_
s_
sp
sp
no
no
ch
ch
ch
La
La
La
s_
s_
s_
Fig. 2 Alterations of gut microbial species in UIA patients. Turquoise and orange, abundance of control- and UIA-enriched species; *P < 0.05, for both the
first (n = 100) and second cohorts (n = 40) (two-tailed Wilcoxon rank-sum test). In all box plots, boxes represent the interquartile ranges (IQRs) between
the first and third quartiles, and the line inside the box represents the median; whiskers represent the lowest or highest values within 1.5 × IQR from the
first or third quartiles. Source data are provided as a Source Data file.
a b
1.0 1.0
0.8 0.8
True positive rate
True positive rate
0.6 0.6
0.4 0.4
0.2 0.2
Luck Luck
Species (AUC = 0.86) Species (AUC = 0.72)
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
False positive rate False positive rate
Fig. 3 Gut microbial species differentiate UIA patients from controls. a Receiver operating curve (ROC, turquoise) according to cross-validated random
forest models on 47 species for the training set (the first cohort, controls, n = 100; UIA patients, n = 100). The area under the receiver operating curve
(AUC) is 0.86, and the 95% confidence interval (CI) is 0.81–0.91. b ROC for the test set (the second cohort, controls, n = 40; UIA patients, n = 40). The
AUC is 0.72, and the 95% CI is 0.65–0.79.
microbial pathways that were differentially abundant using both levels of these classes of amino acid in the microbiota of obese
HUMAnN2 and KEGG. In particular, isoleucine biosynthesis and humans16 and mice17. Overall, although functional annotation
methionine biosynthesis were enriched in the controls. This result analyses are predictive, these data demonstrated that impairment
demonstrated a difference in the potential of the microbiota to of the UIA microbiome may evoke a disease-linked state through
produce branched-chain and aromatic amino acids between UIA interference with physiological metabolic functions, especially fatty
patients and controls. This finding is consistent with the altered acid and amino acid metabolism.
4 NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3 ARTICLE
a b
Control enriched
Correlation
0.6
0.0
–0.6 Stearic acid
UIA enriched Palmitic acid
Lauric acid
Myristic acid
L-Cysteine
Glycolysis VI (metazoan) L-Cystine
Superpathway of pyrimidine ribonucleosides salvage Phenylalanine
L-tryptophan biosynthesis
Sarcosine
Homolactic fermentation
Superpathway of pyrimidine deoxyribonucleotides de novo biosynthesis L-2-Aminobutyric acid
Glycolysis II (from fructose 6-phosphate)
Hexitol fermentation to lactate, formate, ethanol and acetate L-Ornithine
Superpathway of L-tryptophan biosynthesis Hypotaurine
Superpathway of glucose and xylose degradation
Superpathway of GDP-mannose-derived O-antigen building blocks biosynthesis L-Histidine
Colanic acid building blocks biosynthesis Ethanolamine
Petroselinate biosynthesis Taurine
Biotin biosynthesis II
Superpathway of fatty acids biosynthesis Linoleic acid
Superpathway of unsaturated fatty acids biosynthesis Oleic acid
Superpathway of lipopolysaccharide biosynthesis
Pyruvate fermentation to acetone Docosahexaenoic acid
Seleno-amino acid biosynthesis L-Citrulline
Mannitol cycle L-Glutamine
Glycerol degradation to butanol
s__Lachnospiraceae_bacterium_2_1_58FAA
s__Lachnospiraceae_bacterium_3_1_46FAA
s__Lachnospiraceae_bacterium_1_1_57FAA
s__Lachnospiraceae_bacterium_5_1_63FAA
s__Dorea_longicatena
s__Eubacterium_ventriosum
s__Bacteroides_thetaiotaomicron
s__Eubacterium_rectale
s__Parabacteroides_merdae
s__Bacteroides_caccae
s__Ruminococcus_obeum
s__Clostridium_nexile
s__Clostridium_bolteae
s__Anaerostipes_hadrus
s__Streptococcus_salivarius
s__Dorea_formicigenerans
s__Coprococcus_comes
s__Alistipes_putredinis
s__Bacteroides_ovatus
s__Streptococcus_thermophilus
s__Bacteroides_massiliensis
s__Bacteroides_intestinalis
s__Odoribacter_splanchnicus
s__Bacteroides_cellulosilyticus
s__Bacteroides_fragilis
s__Ruminococcus_gnavus
s__Ruminococcus_torques
s__Sutterella_wadsworthensis
s__Bacteroides_xylanisolvens
s__Dialister_invisus
s__Collinsella_aerofaciens
s__Bacteroidales_bacterium_ph8
s__Parabacteroides_sp_D25
s__Ruminococcus_bromii
s__Eubacterium_hallii
s__Escherichia_coli
s__Clostridium_bartlettii
s__Bacteroides_nordii
s__Bacteroides_dorei
s__Faecalibacterium_prausnitzii
s__Alistipes_onderdonkii
s__Prevotella_copri
s__Hungatella_hathewayi
s__Elizabethkingia_unclassified
s__Coprobacillus_unclassified
s__Bilophila_unclassified
s__Escherichia_unclassified
Superpathway of heme biosynthesis from glycine
Ketogluconate metabolism
Nitrate reduction VI (assimilatory)
Allantoin degradation IV (anaerobic)
Allantoin degradation to glyoxylate II
Glutaryl-CoA degradation
L-lysine biosynthesis I
Superpathway of L-serine and glycine biosynthesis I
Inosine-5'-phosphate biosynthesis III
Inosine-5'-phosphate biosynthesis II
4-deoxy-L-threo-hex-4-enopyranuronate degradation
Superpathway of L-threonine biosynthesis
Superpathway of L-isoleucine biosynthesis I
Superpathway of hexuronide and hexuronate degradation
Superpathway of beta;-D-glucuronide and D-glucuronate degradation
D-galacturonate degradation I
D-fructuronate degradation
Superpathway of L-lysine, L-threonine and L-methionine biosynthesis II
Inosine-5'-phosphate biosynthesis I
L-lysine biosynthesis VI
Methylerythritol phosphate pathway I
5-aminoimidazole ribonucleotide biosynthesis II
Superpathway of 5-aminoimidazole ribonucleotide biosynthesis
UDP-N-acetylmuramoyl-pentapeptide biosynthesis II (lysine-containing)
Guanosine ribonucleotides de novo biosynthesis
S-adenosyl-L-methionine cycle I UIA enriched Control enriched
–10.0 –7.5 –5.0 –2.5 0.0 2.5 5.0 7.5
Log2 ratio
Fig. 4 Functional characterization of the UIA microbiome. a Pathways with biologically significant differential abundance between UIA patients and
controls (n = 100). Turquoise and orange, control- and UIA-enriched pathways. b Heat maps of Spearman’s rank correlation coefficients between 19
altered amino acids and fatty acids and 47 UIA-associated gut microbial species. +P < 0.05, *P < 0.01. Pairs with low correlations (|r| < 0.4) are not shown
(n = 30). Source data are provided as a Source Data file.
Associations of gut microbial species with circulating meta- distributed near the center of gravity of 100 controls, and the UIA
bolites. We functionally characterized the UIA microbiome based donors were distributed near the center of gravity of the 100 UIA
on the metabolism of fatty acids and amino acids by further patients (Supplementary Fig. 8a). Furthermore, to demonstrate
exploring their metabolic profiles in the sera of a subset of 60 the extent to which the mouse recipients had microbial profiles
participants (30 UIA patients and 30 controls) from the first similar to the human donors, we used SourceTracker analyses to
cohort using targeted metabolomics analysis. In total, the serum determine the extent of donor engraftment using a Bayesian
concentrations of 7 of 9 fatty acids and 12 of 32 amino acids algorithm18. We found that 61.5% of the fecal bacterial com-
differed substantially between UIA patients and controls (Sup- munities in mice treated with control feces were attributable to
plementary Data 11). We then investigated whether the abun- the control donors, and 53.8% of the fecal bacterial communities
dance of the 19 altered circulating metabolites correlated with the in mice treated with UIA feces were attributable to the UIA
47 altered gut microbiota at the species level. Spearman’s corre- donors (Supplementary Fig. 8b; Supplementary Data 12 and 13).
lation of differentially enriched species and metabolites was cal- To further identify the microbial species after fecal trans-
culated, and a heat map was hierarchically clustered to represent plantation, MetaPhlAn2 was run to identify the taxonomic
the species-metabolite-associated patterns. Notably, consistent abundances at the species level in mice. Among the 14
with the proposed differential capacities for producing aromatic differentially abundant species between mice treated with UIA
amino acids in the microbiota of UIA patients, the serum con- patient feces and those treated with control feces, 5 species
centrations of phenylalanine were considerably higher in UIA (35.7%), namely, H. hathewayi, O. splanchnicus, A. putredinis, B.
patients than in controls (Fig. 4b). Furthermore, we found that intestinalis, and B. nordii, were also differentially enriched in
circulating amino acids such as taurine, hypotaurine, L-histidine MetaPhlAn2-associated species in UIA patients compared with
and L-citrulline, which were significantly decreased in UIA controls in the two cohorts (Supplementary Data 14). The
patients, had a strong association with altered microbial species. relative abundances of the top five most abundant species that
Together, although it remains to be explored whether these showed significant differences between the two groups are
metabolic products are directly metabolized by the gut micro- shown in Supplementary Fig. 9.
biota, our data indicate that gut microbial species can modulate In all, 20 mice treated with UIA patient feces and 20 treated
the levels of circulating amino acids and fatty acids that are with control feces underwent aneurysm-induction surgery. There
further correlated with UIA. was no mortality during the surgical procedure. Compared with
mice treated with control feces (Con-FMT), treatment with feces
UIA gut microbiota contributes to the intracranial aneurysms. from UIA patients (UIA-FMT) significantly increased the overall
To further determine whether changes in the gut microbiota are a incidence of aneurysms (Fig. 5a, b and Supplementary Fig. 10a;
direct factor in the progression of UIAs in vivo, fecal bacteria UIA-FMT vs Con-FMT: 85% vs 45%; n = 20; P = 0.019) and the
from UIA patients and controls were transplanted into mice. rupture rate (Fig. 5b and Supplementary Fig. 10b; UIA-FMT vs
Depletion of the gut microbiota was achieved by administering a Con-FMT: 82% vs 22%; n = 17 vs 9; P = 0.009). Moreover, a
mouse broad-spectrum antibiotic cocktail. For fecal transplanta- symptom-free curve (Kaplan–Meier analysis) was plotted after
tion, mice were administered 100 μL of the stool supernatant excluding mice that did not have intracranial aneurysms (Fig. 5c).
from two individuals with UIAs and two control donors twice at a The log-rank test revealed a significant increase in aneurysmal
1-day interval (Supplementary Table 1). The fecal samples of rupture with UIA fecal treatment (P = 0.0056). Together, these
human donors and recipient mice post-transplantation were findings provide direct evidence that the gut microbiota can
investigated by metagenomic shotgun sequencing. As expected, influence the formation and rupture of intracranial aneurysms in
PCoA at the species level showed that the control donors were the host.
NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications 5
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3
a b c d UIA-FMT model vs sham
Con-FMT
Ruptured
80
Incidence of aneurysm (%)
UIA-FMT
Con-FMT
Unruptured 100
Cumulative symptom-free
70% 1039
No aneurysm
N=9
60 55% 80
survivival (%)
60 173
40 35% P = 0.0056
40
UIA-FMT
20 15%15% N = 17 7
10% 20
0 0
Con-FMT UIA-FMT 0 5 10 15 20 25 Con-FMT model vs sham
Days
e 3
f
2 Extracellular region Cytokine-cytokine receptor interaction
Expression
1
0 Extracellular space Chemokine signaling pathway
–1 Inflammatory response
–2 Basal cell carcinoma
–3 Extracellular matrix
Lysosome
Chemotaxis 16
Fc gamma R-mediated phagocytosis 2.5
–log10 (q-value)
Intracellular 14
–log10 (q-value)
12 2.0
Integral component of plasma membrane 10 Cell cycle
8 1.5
Cell adhesion 6 Thyroid hormone synthesis 1.0
4
Angiogenesis 2 PPAR signaling pathway 0.5
0 0.0
Heparin binding Natural killer cell mediated cytotoxicity
Size
RAGE receptor binding 5 Phagosome
15 Size
Cytokine activity 5
25 IL-17 signaling pathway
Cell cycle 15
40 Apoptosis 25
Protein kinase binding
50 B cell receptor signaling pathway 40
Smooth muscle cell migration
Response to oxidative stress 150
Glycosaminoglycan degradation 50
Receptor binding Pathways in cancer
C -FM -sh -1
U -FM -mo -4
U MT od l-3
U FM -mo el-1
C -FM -sh -3
U -FM -sha -1
C -FM -sh -2
U -FM -mo el-2
od 4
on T o 2
on T od 1
U -FM -sh -3
U -FM -sh -2
M o -3
C -FM -m -4
U -FM od l-4
-5
U FM -sha -5
m el-
C -FM -m el-
C -FM -m el-
m
IA T am
m
IA T m
m
IA T am
IA T m
-F -m el
m
el
l
-F -m de
IA -m e
IA T e
IA T d
IA T d
T- d
on T a
on T od
IA T d
on T a
on T a
on T a
2.5 5.0 7.5 10.0 12.5 15.0 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00
C -FM -sh
on T
–log10 (q-value) –log10 (q-value)
C -FM
-
on
-
Gene ontology KEGG pathway
C
Fig. 5 UIA gut microbiota contributes to the intracranial aneurysms. a Representative images of intracranial aneurysms (left; arrows, aneurysms; scale
bar, 1 mm) and cerebral arteries stained with hematoxylin-eosin (right; scale bar, 100 μm) in mice that received fecal microbiota transplantation (FMT)
from UIA patients or controls, followed by aneurysm induction with angiotensin II and elastase. Independent experiments were repeated at least five times.
b Incidence of unruptured and ruptured aneurysms 21 days after induction (n = 20). c Cumulative symptom-free curves for mice with aneurysms showing
the time course of the onset of symptoms (n = 9 for control-FMT group, n = 17 for UIA-FMT group; log-rank (Mantel–Cox) test). d Venn diagram of the
number of DEGs. Number in cyan circle, DEGs between the aneurysm induction group and the sham group after UIA-FMT. Number in pink circle, DEGs
between the aneurysm induction group and the sham group after Control-FMT. e Right, heat maps of cluster analysis showing the transcript expression
levels of DEGs in the cerebral vessels of the control-FMT and UIA-FMT groups on day 5 after aneurysm induction. The color scale illustrates the relative
expression levels across all samples: orange, above the mean; blue, below the mean. Left, dendrogram showing the clustering of transcripts (n = 4 for sham
groups, n = 5 for aneurysm induction groups). f Top terms showing enrichment based on −log10 (q-value) from GO enrichment analysis (left) and KEGG
pathway analysis (right) of the 1039 UIA-FMT-induced DEGs. Source data are provided as a Source Data file.
UIA gut microbiota aggravates gene transcriptional altera- multiple categories associated with inflammatory processes, cell
tions. To gain mechanistic insights into the effects of the UIA gut adhesion, response to oxidative stress, and the extracellular
microbiota on the formation and rupture of intracranial aneur- matrix, which are known pivotal mechanisms of intracranial
ysms, we performed whole-transcriptomic analysis of cerebral aneurysm formation and rupture (Fig. 5f)19,20. KEGG pathway
vessels using RNA-seq. In mice transplanted with feces from UIA analysis suggested that the DEGs were enriched in several
patients, a total of 1212 genes (850 upregulated and 362 down- important intracranial aneurysm-related signaling pathways,
regulated) were differentially expressed between the sham group including apoptosis and several inflammation-related pathways,
and the angiotensin II- and elastase-treated group (Fig. 5d, e, such as cytokine–cytokine receptor interactions and chemokine
Supplementary Data 15). In contrast, in mice transplanted with signaling (Fig. 5f). Taken together, the RNA-seq data revealed
control feces, only 180 genes were differentially expressed that the UIA gut microbiota significantly aggravated the profound
between these two groups. Among these 180 genes, 173 over- changes in the gene transcriptional profile of cerebral vessels and
lapped between the transplantation of feces from UIA patients that the mechanism by which intracranial aneurysms are induced
and controls. To more specifically elucidate the underlying by UIA gut microbiota may involve inflammation, modulation of
mechanism by which the UIA gut microbiota affected intracranial the extracellular matrix in cerebral vessels, and apoptosis.
aneurysms, we excluded these 173 genes from the 1212 differ-
entially expressed genes (DEGs) induced by UIA gut microbiota, Taurine reduces the incidence of intracranial aneurysms. To
i.e., 1039 genes were used for further functional analysis. We validate the altered circulating metabolites in UIA patients and to
investigated the intersection between these 1039 genes and further assess the relationships between gut microbiota and cir-
another human transcriptome-wide characterization of gene culating metabolites, we performed targeted metabolomics pro-
expression associated with UIA (GSE26969) and found con- filing of fatty acids and amino acids in mice after fecal
siderable overlap between DEGs across species (Supplementary transplantation. In total, the serum concentrations of 2 of 8 fatty
Fig. 11, Supplementary Data 16). This suggested that the pro- acids and 8 of 38 amino acids differed substantially between mice
found alterations of the transcriptional profile evoked by the UIA transplanted with feces from UIA patients and controls (Sup-
gut microbiota are a shared feature of intracranial aneurysms. plementary Data 17). Specifically, among these altered metabo-
Furthermore, this list of 1039 genes was subjected to Gene lites, taurine, L-histidine, and linoleic acid also corroborated the
Ontology enrichment analysis to identify overrepresented biolo- lower abundance in UIA patients compared with controls (Sup-
gical functions and canonical pathways. These genes belonged to plementary Fig. 12). On this basis, we next determined whether
6 NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3 ARTICLE
a Con-FMT UIA-FMT UIA-FMT b c
Ruptured Unruptured No aneurysm
+ Vehicle + Vehicle + Taurine
Incidence of aneurysm (%)
80 100
Cumulative symptom-free
70%
60% 80 N=8
60 55%
survival (%)
60
P = 0.0158
N=9
P = 0.0290
40
30% 30%
40
Con-FMT + Vehicle
20%
20 15% UIA-FMT + Vehicle
10% 10% 20 N = 18
UIA-FMT + Taurine
0 0
Con-FMT UIA-FMT UIA-FMT 0 5 10 15 20 25
+ Vehicle + Vehicle + Taurine
Days
d e g
(fold of UIA-FMT + Vehicle)
Con-FMT UIA-FMT UIA-FMT P = 0.0038 P = 0.0024 P = 0.0126 P = 0.0014 8
macrophage cells/field
200 3
00 2
mRNA expression
50 0.0 34
+ Vehicle + Vehicle + Taurine P= P=
0.0
0.0
02
3
IL-6 (pg / mL)
40 150 6 3 17 P=
Number of
LY6G α-SMA DAPI F4/80 CD31 DAPI
0.0 1
P= 42 8
30 0.1 00
4 2 P= 0.4
100 63 P=
0.6
20 P=
50 2
10
0 0 0
C 1
C 1
C 1
C 1
1
C 3
C 2
hi MT
hi MT
hi MT
hi MT
ur MT
ur MT
2a
4a
1a
3a
5a
4a
4a
e
e
U e
e
e
e
Ve F
Ve -F
Ve -F
Ve -F
Ta -F
Ta -F
in
in
ol
ol
ol
ol
ol
cl
cl
ol
ol
cl
cl
-
C
+ IA
+ A
+ on
+ on
+ IA
+ IA
I
U
U
U
C
C
(foldof UIA-FMT + Vehicle)
9
400 P = 0.0014 P = 0.0003 8 07
0.0
Number of neutrophil
P=
mRNA expression
40 P = 0.0068 P = 0.0030
TNF-α (pg / mL)
8
300 6 0.0
24
06
8
1
30 07
9 P= 0.1 71
cells/field
0.0 P= 0.1 95
2
P= P= 0.0
200 83 P=
20 4 0.9
6
P=
10 100 2
0 0 0
ur MT
hi MT
hi MT
La 1
La 1
La 1
La 2
b2
La 3
c2
ur MT
hi MT
hi MT
a
b
c
a
a
e
U e
e
e
U e
m
m
m
m
m
m
e
m
Ta -F
in
Ve -F
Ve -F
cl
Ta -F
in
Ve -F
Ve -F
cl
cl
cl
La
La
+ IA
+ IA
+ on
+ IA
+ A
+ on
I
U
U
C
C
f Vehicle Vehicle Taurine h
Arbitrary densitometric units
(fold of Con-FMT + Vehicle)
Con-FMT UIA-FMT UIA-FMT
Number of TUNEL+ cells
+ on α-SMA per arterial
P = 1.7E-6 P = 1.1E-6 60 P = 0.0018 P = 0.0009
5 + Vehicle + Vehicle + Taurine
Con-FMT
Con-FMT
Standard
UIA-FMT
UIA-FMT
UIA-FMT
UIA-FMT
4
40
Tunel α-SMA
3
2
20
Pro-MMP9 95 kDa 1
0
Pro-MMP2 72 kDa 0
hi T
hi MT
ur MT
Ve FM
U le
e
U le
hi MT
hi T
Ve F
ur MT
Ta F
in
c
c
Ve FM
+ IA-
+ on-
+ IA-
U le
e
U le
Ve F
Ta F
in
c
c
+ on-
+ IA-
C
+ IA-
C
Fig. 6 Taurine reduces the incidence of intracranial aneurysms. a Representative images of intracranial aneurysms (upper; arrows, intracranial
aneurysms; scale bar, 1 mm) and cerebral arteries stained with hematoxylin-eosin (lower; scale bar, 100 μm) in mice that received fecal microbiota
transplantation (FMT) from UIA patients or controls, followed by taurine supplementation and aneurysm induction with angiotensin II and elastase.
Independent experiments were repeated at least five times. b Incidence of unruptured and ruptured aneurysms 21 days after aneurysm induction (n = 20).
c Cumulative symptom-free curves for mice with aneurysms to show the time course of symptom onset (n = 9 for control-FMT + Vehicle group, n = 18 for
UIA-FMT + Vehicle group, n = 8 for UIA-FMT + Taurine group; log-rank (Mantel–Cox) test). d Upper panels, representative images and quantification of
F4/80+ macrophage immunostaining surrounding CD31+ cerebral vessels. Lower panels; representative images and quantification of Ly6G+ neutrophil
immunostaining surrounding α-SMA+ cerebral vessels (scale bars, 50 μm). e Levels of IL-6 and TNF-α in serum measured by ELISA. f Representative
gelatin zymography and quantitative analysis of the MMP-9 activity in cerebral vessels from each group (n = 6; one-way ANOVA with the Bonferroni post
hoc test). g Quantitative PCR showing the expression of pivotal components of arterial structure (n = 5; Student’s unpaired two-tailed t-test or
Mann–Whitney U test with the exact method). h Representative images and quantification of α-SMA+ vascular smooth muscle cell apoptosis in each
group (scale bar, 50 μm). Data are the mean ± SD. For (d), (e), and (h), n = 5; one-way ANOVA with the Bonferroni post hoc test. Source data are
provided as a Source Data file.
supplementation with taurine, L-histidine, or linoleic acid has Intracranial aneurysms are increasingly recognized as a
potential therapeutic value for reducing the formation and rup- condition driven by chronic inflammation21,22. We observed
ture of intracranial aneurysms in mice. cumulative infiltration of macrophages and neutrophils in
Compared with UIA fecal treatment, taurine supplementation perivascular spaces in mice transplanted with feces from UIA
normalized the serum levels of taurine (Supplementary Fig. 13) and patients (Fig. 6d). Together with increased levels of the pro-
significantly reduced the overall incidence of aneurysms (Fig. 6a, b inflammatory cytokines interleukin (IL)-6 and tumor necrosis
and Supplementary Fig. 14a; UIA-FMT + Taurine vs UIA-FMT + factor alpha (TNF-α) (Fig. 6e), these data confirmed that stronger
Vehicle: 40% vs 90%; n = 20; P = 0.002) and the rupture rate vascular inflammation was evoked by the UIA gut microbiota.
(Fig. 6b and Supplementary Fig. 14b; UIA-FMT + Taurine vs UIA- Notably, inflammatory processes were dramatically ameliorated
FMT + Vehicle: 25% vs 78%; n = 8 vs 18; P = 0.026). Kaplan–Meier by supplementation with taurine. It has been reported that the
analysis further revealed a significant decrease in aneurysmal inflammatory process in cerebral arteries leads to matrix
rupture with taurine supplementation (P = 0.0158, Fig. 6c). Notably, metalloproteinase (MMP)-mediated degradation of the extra-
taurine supplementation did not affect blood pressure, indicating cellular matrix and apoptosis of smooth muscle cells19,23. Using
that an antihypertension-independent mechanism is involved in the gelatin zymography, we confirmed that the MMP-9 activity in
protective effects of taurine (Supplementary Fig. 15). cerebral arteries was significantly induced by treatment with feces
NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications 7
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3
from UIA patients compared with treatment with control feces rate (Supplementary Fig. 17c–e). Kaplan–Meier analysis further
(Fig. 6f). Likewise, the mRNA levels of collagen IV and laminin, revealed a significant increase in aneurysmal rupture with β-
which play substantial roles in arterial structure, were signifi- alanine administration (Supplementary Fig. 17f). These data
cantly decreased after UIA fecal treatment (Fig. 6g). Importantly, demonstrated that the diminished level of taurine increases the
both MMP activation and extracellular matrix remodeling were incidence of the formation and rupture of intracranial aneurysms
dramatically ameliorated by taurine supplementation. Moreover, in mice.
both significantly reduced TUNEL-positive vascular smooth
muscle cell apoptosis and compensatory hypertrophy with wall
thickening of cerebral arteries were found after taurine supple- H. hathewayi protects against the intracranial aneurysms. To
mentation (Fig. 6h). However, we did not find a protective role of further investigate the extent to which the altered microbiome in
L-histidine or linoleic acid in reducing the incidence of the UIAs is associated with the changed circulating taurine level in the
formation and rupture of intracranial aneurysms (Supplementary host, we calculated Spearman’s rank correlation between altered
Fig. 16). Taken together, these data clearly demonstrated the metabolites and altered species in mice after fecal transplantation.
essential roles of taurine in blunting inflammatory processes, We found that five species were positively correlated and 1 species
reducing extracellular matrix remodeling, and maintaining the was negatively correlated with taurine levels (Fig. 7a). Among these
structural integrity of cerebral arteries, all of which ultimately 6, H. hathewayi was the only overlapping species that also posi-
reduced the formation and rupture of intracranial aneurysms. tively correlated with taurine levels in the human study. Further-
To further corroborate the critical role of taurine reduction in more, lower relative abundances of H. hathewayi all occurred in
intracranial aneurysm pathogenesis, mice that transplanted with UIA patients in the two cohorts and mice treated with UIA patient
feces from control donors were supplemented with 3% β-alanine feces compared with their respective controls (Fig. 7b). On this
in the drinking water, which significantly decreased the plasma basis, to explore the possible causal role of H. hathewayi in the
concentration of taurine (Supplementary Fig. 17a). Compared regulation of the level of circulating taurine and the formation and
with Vehicle treatment, β-alanine treatment significantly rupture of intracranial aneurysms, mice were gavaged with live H.
increased the overall incidence of aneurysms and the rupture hathewayi or vehicle [sterile phosphate-buffered saline (PBS)] after
a UIA-FMT enriched Control-FMT enriched
Coefficient
0.8
–0.0
–0.8
L-Kynurenine
Allantoin
2-Aminoisobutyric acid
L-Histidine
Taurine
DL-methionine sulfoxide
Methylamine
L-Aspartic acid
Arachidonic acid
Linoleic acid
ce s_ clo a
_b ns ae
s
tip nc ns
riu tus
lis
s
ut us
82
s_ ter gar par
st 5
bs rce ayi
rib as a_ ri
es sp i
ac rep tero ella cen
ni
oc
_H oid s_ rdi
s_ nga _in _D2
do on ell op
ra _h ina
lis pla ra
_p ic
ae co ac
_0
di
la
s_ _Kl ma ew
ct xyt
ac _a dis
er de no
es hn
_A _s ivo
_O m lon _c
re
s
ac tel
m
_
_ th
ba _o
s_ tero eil tella
te
ac ero es
_
a
cc er
te
id
V o
s_ aba tero
er lla
v
i
i
re
s_ ia
u
_S te
t
_P
e
a c
_B ct
co
_P _B
s_
_
n
t
s_
u
to
_E
s_ s_
ra
ar
pi
l
s_
oa
os
_L _St
ud
hn
_
se
s
_P
s_
s_
b 0.6
Relative abundance of
Control
H.hathewayi (Log2)
P = 0.003 P = 0.037
0.4 UIA
P = 0.0003
0.2
0.0
e
rs
rt)
rt)
ic
st an
no
ho
ho
tm
nd n
fir m
do
co
co ma
co
n
e Hu
ie
an
se Hu
ip
um
ec
R
H
(th
e
(th
Fig. 7 Associations of gut microbial H. Hathewayi with circulating taurine in mice. a Heat maps of Spearman’s rank correlation coefficients between 10
altered metabolites and 14 differentially abundant gut microbial species. +P < 0.05, *P < 0.01. Pairs with low correlations (|r| < 0.4) are not shown (n = 10).
b Relative abundances of H. hathewayi in two cohorts of UIA patients (n = 100; n = 40), Human donors (n = 2), and mice that received fecal microbiota
transplantation (FMT) from UIA patients or controls (n = 10). Two-tailed Wilcoxon rank-sum test. In all box plots, boxes represent the interquartile ranges
(IQRs) between the first and third quartiles, and the line inside the box represents the median; whiskers represent the lowest or highest values within 1.5 ×
IQR from the first or third quartiles.
8 NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3 ARTICLE
fecal transplantation. Real-time PCR analysis showed a sig- and Supplementary Fig. 18a; UIA-FMT + hathewayi vs UIA-FMT
nificantly reduced amount of H. hathewayi in the feces of mice + Vehicle: 45% vs 90%; n = 20; P = 0.006) and the rupture rate
treated with UIA patient feces compared with that in the feces of (Fig. 8d and Supplementary Fig. 18b; UIA-FMT + hathewayi vs
mice treated with control feces. Oral gavage with 1 × 109 colony- UIA-FMT + Vehicle: 44% vs 83%; n = 9 vs 18; P = 0.026).
forming units of live H. hathewayi was sufficient to restore the Kaplan–Meier analysis further revealed a significant decrease in
diminished level caused by the transplantation of UIA patient feces aneurysmal rupture with H. hathewayi gavage (P = 0.0134,
(Fig. 8a). Importantly, the plasma concentration of taurine also Fig. 8e). Furthermore, inflammatory processes, extracellular matrix
significantly increased after H. hathewayi gavage (Fig. 8b). Then, 3 remodeling, and MMP-9 activity in cerebral arteries were all
groups of mice underwent aneurysm-induction surgery (repre- dramatically ameliorated by H. hathewayi gavage (Fig. 8f–i).
sentative intracranial aneurysms in the 3 groups are shown in Likewise, H. hathewayi gavage significantly reduced TUNEL-
Fig. 8c). Compared with UIA fecal treatment, H. hathewayi gavage positive vascular smooth muscle cell apoptosis in cerebral arteries
significantly reduced the overall incidence of aneurysms (Fig. 8d (Fig. 8j). Taken together, these findings provide direct evidence
a b f Con-FMT UIA-FMT UIA-FMT
(Log10 bacteria/g feces content)
P = 0.0036 P = 0.0007 P = 0.0004 P = 0.0015
P = 0.0258 P = 0.0020
+ Vehicle + Vehicle + H.hathewayi
macrophage cells /field
50
Concentration level of
8 1000
SMC F4/80 DAPI 40
taurine (μmol/L)
Number of
800
H. hathewayi
6
30
600
4 20
400
2 10
200
0
0 0
hi T
hi T
i
th T
ay
Ve FM
Ve FM
ha -FM
hi T
i
hi T
hi T
hi T
i
th T
th T
cle
ay
cle
ay
ew
Ve FM
Ve FM
Ve FM
Ve FM
ha -FM
ha -FM
+ UI le
cle
cle
cle
+ on-
+ IA-
ew
ew
H. IA
c
+ on-
+ IA-
+ on-
+ IA-
H. IA
U
H. A
C
U
U
U
U
C
C
+
+
c Con-FMT UIA-FMT UIA-FMT
g P = 0.0001 P = 0.0006
h
(fold of UIA-FMT + Vehicle)
250
+ Vehicle + Vehicle + H.hathewayi 15
09
mRNA expression
200
04
IL-6 (pg / mL)
0.
33
=
01
P
60
150 10
72
79
47
0.
00
00
01
00
=
0.
74
0.
P
0.
0.
=
01
100
=
=
=
P
0.
P
P
P
=
5
P
50
0
0
hi T
i
hi T
th T
ay
Ve FM
Ve FM
ha -FM
1
a1
1
b1
1
a3
2
cle
cle
ew
4a
3a
1a
4a
+ on-
+ IA-
m
m
m
H. A
ol
ol
ol
ol
+ UI
La
La
La
U
C
C
C
C
C
d Ruptured Unruptured No aneurysm i
Incidence of aneurysm (%)
80 75%
P = 7.3E–7 P = 5.4E–7
(fold of Con-FMT + Vehicle )
Arbitrary densitometric units
4
60 55% Vehicle Vehicle H.hathewayi
50% 3
40 35% 2
Pro-MMP9 95 kDa
25%
20% 72 kDa
20 15% 15% Pro-MMP2 1
10%
0
rd
T
T
T
T
T
T
0
M
M
M
M
M
M
da
hi T
hi T
i
th T
ay
Ve FM
Ve FM
-F
-F
-F
-F
-F
-F
ha -FM
cle
cle
Con-FMT UIA-FMT UIA-FMT
ew
an
IA
+ IA-
IA
+ on-
on
on
IA
IA
H. IA
St
U
U
U
U
U
U
C
+ Vehicle + Vehicle + H.hathewayi
C
C
+
e j
Cumulative symptom-free
100
Con-FMT UIA-FMT UIA-FMT P = 0.0017 P = 0.0167
Number of TUNEL cells
80
on α-SMA per arterial
80 + Vehicle + Vehicle + H.hathewayi
N = 10
+
survival (%)
SMC TUNEL 60
60 N=9
P = 0.0054
P = 0.0044
40
40
Con-FMT + Vehicle 20
20 UIA-FMT + Vehicle N = 18
0
UIA-FMT + H.hathewayi
hi T
i
hi T
th T
ay
Ve FM
Ve FM
ha -FM
+ UI le
0
cle
ew
c
+ on-
+ IA-
H. A
U
C
0 5 10 15 20 25
Days
Fig. 8 H. hathewayi protects against the intracranial aneurysms. a Quantitative PCR detection of H. hathewayi in fecal DNA in each group (n = 5).
b Serum concentrations of taurine in each group measured by UHPLC-MS/MS analyses (n = 10; one-way ANOVA with the Bonferroni post hoc test).
c Representative images of intracranial aneurysms in UIA patients and controls that received fecal microbiota transplantation (FMT), followed by
oral gavage with H. hathewayi and aneurysm induction with angiotensin II and elastase (arrows, intracranial aneurysms; scale bar, 1 mm). d Incidence
of unruptured and ruptured aneurysms 21 days after aneurysm induction (n = 20). e Cumulative symptom-free curves for mice with aneurysms to
show the time course of symptom onset (n = 10 for control-FMT + Vehicle group, n = 18 for UIA-FMT + Vehicle group, n = 9 for UIA-FMT + H.
hathewayi group; log-rank (Mantel-Cox) test). f Representative images and quantification of F4/80+ macrophage immunostaining surrounding
CD31+ cerebral vessels (scale bar, 50 μm). g Concentration of IL-6 in serum measured by ELISA. h Quantitative PCR showing the expression
of pivotal components of arterial structure (n = 5; Student’s unpaired two-tailed t-test or Mann–Whitney U test with the exact method).
i Representative gelatin zymography and quantitative analysis of the MMP-9 activity in cerebral vessels from each group (n = 6; one-way ANOVA
with the Bonferroni post hoc test). j Representative images and quantification of α-SMA+ vascular smooth muscle cell apoptosis in each group (scale
bar, 50 μm). Data are the mean ± SD. For (a), (f), (g), and (j), n = 5; one-way ANOVA with the Bonferroni post hoc test. Source data are provided as
a Source Data file.
NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunications 9
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16990-3
that H. hathewayi affects circulating taurine concentrations and remodeling, and maintenance of the structural integrity of cerebral
protects mice against the formation and rupture of intracranial arteries, all of which are pivotal mechanisms of intracranial
aneurysms. aneurysm formation and rupture19,20,23. Interestingly, our results
are analogous to those of a previous study that found that the
circulating concentration of taurine is reduced in germ-free mice
Discussion receiving fecal transplantation from individuals with autism spec-
Despite extensive research in recent years, relatively little is trum disorder, and taurine supplementation improves repetitive
known about the precise mechanisms that contribute to the and social behaviors in mice33. Nevertheless, the microbiome fea-
pathogenesis of UIAs. In addition to several untreatable risk tures of autism spectrum disorder depicted in this study are dis-
factors, including genetic factors, old age, and female sex, there tinct from those of UIAs in the present study.
are also treatable risk factors that increase the occurrence of UIA, Possible UIA is increasingly identified in clinical practice as the
such as hypertension, cigarette smoking, and heavy alcohol use of MRI and CT becomes more common; magnetic resonance,
use24,25. Accumulating evidence has shown that the gut micro- digital subtraction, or CT angiography is then necessary to clarify
biota is a pivotal risk factor in cardiovascular diseases by influ- the presence and specify the location of the aneurysm34. How-
encing host metabolism and immune homeostasis. However, no ever, these techniques usually cannot achieve the high sensitivity
direct evidence has established a direct and causal relationship required to detect aneurysms smaller than 4 mm34. In addition,
between an altered gut microbiota and UIAs. In the present study, these techniques are relatively expensive and are not available to
we identified UIA-associated microbial species and explored their patients for whom contrast administration is contraindicated. In
effects on host amino acid and fatty acid profiles. Using a mouse the current study, we were able to identify UIA patients based on
model of fecal transplantation and gavage with H. hathewayi, we the differentially abundant species. These results demonstrate the
uncovered evidence for the potential effects of altered H. hathe- presence of UIA-associated features in the gut microbiota that
wayi abundance on circulating taurine levels and on the increased may be further developed into noninvasive and sensitive bio-
occurrence of UIAs. markers for the early detection of UIAs, especially in the high-risk
The current understanding characterizes chronic inflammation population without aneurysmal mass-effect symptoms.
as the leading factor in the pathogenesis of UIAs10,19. In the Although endovascular treatment and surgical clipping serve as
present study, increased levels of the pro-inflammatory cytokines interventional options for UIAs, the optimal management strat-
IL-6 and TNF-α were measured in the serum of mice transplanted egy remains controversial because of the risk associated with
with feces from UIA patients, corroborating a possible causal these treatments34. A recent study reported that depletion of the
relationship between the altered gut microbiota of UIA patients gut microbiota with an antibiotic cocktail reduces the incidence of
and stronger systemic inflammation. Researchers have previously intracranial aneurysms in mice, representing a potential strategy
highlighted that Bacteroides are potentially pro-inflammatory and for therapeutic intervention for UIAs10. Although the elimination
associated with inflammatory bowel disorder26. Interestingly, as of disease-causing microbiota by antibiotics is an obvious
shown by our data, a bacterial community of Bacteroides species approach, the general consensus is that beneficial microbiota may
was overrepresented in UIA individuals and mice treated with have protective effects, and such a nonspecific antimicrobial
feces from UIA patients. Moreover, a recent study has reported approach may do more harm than good35. In the present study,
that mice treated with H. hathewayi exhibit reduced release of we found that H. hathewayi was the only overlapped species that
cytokines and lower NF-κB activation in dendritic cells27. positively correlated with taurine levels in both human and
The gut microbiota is known to influence host intestinal mouse studies. Gavage with live H. hathewayi after fecal trans-
homeostasis, in part via metabolic pathways such as short-chain plantation normalized the serum levels of taurine and protected
fatty acid metabolism, bile acid metabolism, and amino acid mice against the formation and rupture of intracranial aneur-
metabolism. Concomitant with the significant change in gut ysms, which indicated a causal role of H. hathewayi in the reg-
microbial composition, we found changes in the gene functions of ulation of taurine and intracranial aneurysms. H. hathewayi is a
bacteria from UIA patients, especially in fatty acid and amino acid strictly anaerobic bacterium that was described as a new species
metabolism. We further performed metabolic profiling and found in 200114. This bacterium is known to be responsible for the
the altered circulating metabolites exhibited varying degrees of production of acetate, ethanol, carbon dioxide, and hydrogen, and
correlation with the gut microbial species that differed in abun- there is no direct evidence that it is associated with taurine
dance between UIA patients and controls and may exert pivotal synthesis. Thus, we speculate that the effects of H. hathewayi on
influences on the pathophysiology of UIA development. Among modulating the taurine levels may depend on interactions with
the significantly altered metabolites, stearic acid was found to certain additional microbial species. Further studies are clearly
accumulate to high levels in the thickened aneurysmal wall and warranted to delineate how H. hathewayi, in concert with other
may be involved in the phenotypic switching of medial vascular microbial species, modulates taurine metabolism and whether H.
smooth muscle cells of the aneurysmal wall28. In addition, taurine hathewayi affects UIAs only via taurine. However, our data may
showed a considerable decrease in UIA patients compared with add a dimension to our understanding of the beneficial effects of
controls. Taurine has been shown to reduce inflammatory injury in H. hathewayi on human health. Supplementation with H.
various diseases and to play a protective role in acute ischemic hathewayi, or replacement with taurine, may serve as a viable and
stroke29, traumatic brain injury30, subarachnoid hemorrhage31, convenient approach for future investigations of UIA prevention
and aortic aneurysm formation32. Depletion of the taurine trans- and intervention.
port system has been reported in the microbiome in atherosclerotic Our study population consisted only of middle-aged to elderly
cardiovascular disease7. In the present study, taurine was also Chinese individuals in a northern Chinese population. Thus, our
decreased in the serum of mice after transplantation of feces from current results potentially face limitations in generalizability. In
UIA patients, implying that the altered gut microbiota of UIA future studies, the scope could be broadened to include other
patients may be a direct factor in the low level of taurine, not ethnicities within Asia and other groups such as Caucasians.
merely an adverse complication. Importantly, taurine supple- In summary, we describe the disordered profiles of the gut
mentation significantly alleviated the formation and rupture of microbiota in UIA patients. Our findings extend our insights to
intracranial aneurysms in mice. The mechanistic findings involved a potential role of aberrant gut microbiota in the pathogenesis
blunted inflammatory processes, reduced extracellular matrix of UIAs and provide important evidence for a role of H.
10 NATURE COMMUNICATIONS | (2020)11:3218 | https://doi.org/10.1038/s41467-020-16990-3 | www.nature.com/naturecommunicationsYou can also read
