Oral Pathobiont Streptococcus Anginosus Is Enriched in the Gut of Stroke Patients and Predicts 2-Year Cardiovascular Outcome

Abstract

Background: Several cross-sectional studies have implicated gut dysbiosis caused by an abundance of oral commensals in stroke, but the effect on long-term prognosis is still unknown. Therefore, we longitudinally investigated oral pathobionts in the gut and their clinical relevance to stroke.

Methods and Results: We analyzed the salivary and gut microbiomes collected from 189 acute stroke and 55 non-stroke subjects, and found that Streptococcus anginosus was significantly more abundant in both the saliva (median [IQR], 0.01 [0.00–0.14] vs. 0.00 [0.00–0.03], P=0.02) and gut (0.09 [0.00–0.28] vs. 0.00 [0.00–0.02], P<0.001) of the stroke patients compared with their non-stroke counterparts. Network analysis revealed S. anginosus as a central hub in gut dysbiosis. After adjusting for vascular risks, S. anginosus (odds ratio 1.20, 95% confidence interval 1.06–1.36, P<0.01), Anaerostipes hadrus (0.82, [0.73–0.93], P<0.01), and Bacteroides plebeius (0.86, [0.86–0.93], P=0.01) in the gut were independent predictors of stroke. Longitudinally, S. anginosus in the gut was significantly associated with increased rates of death and major cardiovascular events (P=0.04; log-rank test), whereas A. hadrus and B. plebeius were not (P=0.45 and P=0.19). After adjusting for vascular risks, S. anginosus in the gut was a residual risk for increased rates of death and major cardiovascular events (hazard ratio 4.78, 95% confidence interval 1.08–21.18, P=0.04)

Conclusions: S. anginosus in the gut may increase the risk of stroke and a poor prognosis.

The gut microbiome supports the intestinal mucus barrier^1,2 and is critical for human health. Disruption of gut homeostasis, known as gut dysbiosis, can trigger systemic illness, including ischemic heart disease.³ Although stroke is one of the leading causes of death and disability worldwide,⁴ the specific pattern and contribution of gut dysbiosis to stroke has not been thoroughly elucidated.^5,6

The translocation of oral commensal microbes into the gastrointestinal tract has been linked to gut dysbiosis and systemic inflammation.⁷ To date, 3 cross-sectional stroke cohort studies, but no longitudinal studies, have suggested a significant role of the oral genus Streptococcus in the gut in the pathogenesis of stroke.⁸ Therefore, we aimed to identify oral pathobiont Streptococcus species and others in gut dysbiosis in stroke subjects and investigate their influence on the development of stroke and cardiovascular events.

Methods

Study Design

The study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of the National Cerebral and Cardiovascular Center (NCVC, Japan) (approval number: R20029). Between July 2020 and July 2021, patients with acute stroke and non-stroke individuals were prospectively enrolled in the NCVC. Subjects with acute stroke were diagnosed by neurological examination and brain computed tomography or magnetic resonance imaging. The inclusion criteria were: (1) age >20 years, (2) admitted within 1 day of symptom onset, and (3) subject or relatives provided written informed consent. The exclusion criteria were: (1) antibiotics use within the past 3 months based on a systematic review⁹ and (2) defecated after admission but before recruitment. Of the 198 acute stroke patients who met the inclusion and exclusion criteria, 189 provided written informed consent (Figure 1). The Suita Study is an ongoing population-based cohort study in an urban area performed by the NCVC, the details of which have been reported.¹⁰ Briefly, 6,485 and 1,329 baseline participants in 1989–1996 and 1996–1998 respectively, as well as 546 volunteer participants aged 30–79 years, were enrolled as study participants from the community of Suita City. All patients undergo medical examination every 2 years. For among these participants, we set the baseline of the present study as a medical examination conducted between August 2020 and December 2021. The exclusion criteria were: (1) dementia or a history of stroke, ischemic heart disease, or kidney disease and (2) antibiotics use within the past 3 months. In total, 60 met the inclusion criteria, and 55 provided written informed consent (Figure 1).

Figure 1.

Flow chart of the study. The study participants were drawn from 2 cohorts: patients with acute stroke from a single-center stroke cohort and non-stroke participants from a population-based cohort study. All participants underwent clinical profiling of saliva and fecal samples for microbiome analysis. All available samples were used to investigate the gut and salivary microbiomes associated with stroke.

Clinical Characteristics

The following variables were obtained from electronic medical chart or screening questionnaires: age, sex, smoking habits, alcohol consumption, past vascular events (ischemic heart disease, stroke, and kidney disease), and medications. Comorbidities were diagnosed according to the following criteria: hypertension, defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or a history of antihypertensive medication use; diabetes mellitus, defined as a fasting plasma glucose level ≥126 mg/dL, hemoglobin A1c level of ≥6.5%, or history of antidiabetic drugs or insulin use; hyperlipidemia, defined as low-density lipoprotein cholesterol ≥140 mg/dL, high-density lipoprotein cholesterol ≤40 mg/dL, triglyceride ≥150 mg/dL, or a history of lipid-lowering drug use; atrial fibrillation, defined on ECG, a history of anticoagulant use, or diagnosis.

Major Cardiovascular Events

In the longitudinal observation study, we collected data on major cardiovascular events, including all-cause death, admission for stroke, myocardial infarction, and heart failure, using electronic medical chart data collected between July 2020 and July 2023.

Sample Collection and DNA Extraction

In the acute stroke cohort, we collected fecal samples after admission within 3 days of onset. Samples were collected using a fecal container (SARSTEDT AG & Co., KG, Germany) and a guanidine thiocyanate solution kit (TechnoSurga Laboratory Co., Ltd, Shizuoka, Japan) immediately after defecation. Saliva samples were collected within 3 days of admission. Patients were instructed to rinse their mouths with water to eliminate oral cavity residues before brushing or eating in the morning. Following this preparatory step, the patients spat out 1.0 mL of saliva using a self-collection kit (OMNIgene-ORAL, OM-501, DNA Geneotek, Canada). All samples were immediately frozen and stored at −80℃ until further processing. In the population-based cohort, participants collected fecal samples using a guanidine thiocyanate solution kit and saliva samples at home within one day in the same manner as by the stroke patients. Their samples were transported within 3 days, frozen immediately, and stored at −80℃ until further processing.

DNA was extracted using a Qiagen QIAamp Power Fecal Pro DNA Stool Kit (Qiagen, Hilden, Germany). For 16S rRNA metagenomic analyses, we used DNA extracted from the fecal samples in guanidine thiocyanate solution and the saliva samples in OMNIgene-Oral. The extracts were treated with DNase-free RNases to eliminate RNA contamination. All the extracted DNA samples were quantified using a spectrophotometer (Q5000; Quawell Technology, USA).

16S Ribosomal RNA Metagenomic Analyses

DNA libraries were generated using Illumina 16S metagenomic sequencing protocols and a primer set (27Fmod: 5′-GTT TGA TYM TGG CTC AG-3′ and 338R: 5′-TGC CTC CCG TAG GAG T-3′) targeting the V1–V2 regions of the 16S rRNA gene. Amplicons were subjected to 251-bp paired-end sequencing on a MiSeq system (Illumina, San Diego, CA, USA) with 500 cycles using the MiSeq Reagent Kit v2 (Illumina). Microbiome bioinformatics analysis was performed using QIIME 2 2021.2 and 2022.2.¹¹ A median of 61,541 sequence reads with 250-bp paired-ends was demultiplexed and quality-filtered using the q2-demux plugin, followed by denoising using DADA2. After quality filtering, 32,953,796 sequences were retrieved, with an average of 72,267 sequences per sample (minimum: 33,481, maximum: 191,299). All amplicon sequences were aligned using Mafft via q2-alignment, and fasttree2 was used to construct a phylogeny via q2-phylogeny. Amplicon sequences were taxonomically classified using the q2-feature-classifier with a similarity criterion of 99% against the curated Silva 138-reference database. The pathway abundance based on Meta Cyc ontology was predicted using the amplicon sequence and abundance in PICRUSt2 pipeline.¹²

Alpha Diversity Analyses

Alpha diversity metrics (observed taxonomic features) were calculated using the q2-diversity plugin.

Identification of Distinctive Features of Microbial Communities

We analyzed the composition of the microbiomes (ANCOM) to identify the taxonomic characteristics that discriminate the microbial composition. For the ANCOM analyses, the taxonomic feature table was filtered with an average prevalence of >20% and a proportion >0.01 and >0.001 at the genus and species levels, respectively. Using q2-composition and data from the core microbiome, we identified specific microbes associated with an acute stroke at the genus and species levels.¹³ A volcano plot was constructed to describe the ANCOM differential features based on the additive log-ratio along the X-axis and W-value, representing the number of additive log-ratio transformed taxa that were differently abundant regarding the variable of interest along the Y-axis.

Microbial Co-Occurrence Network Analysis

The relative proportions of taxonomic features at the species level were normalized using a centered log₁₀ scale. Among the significantly different taxonomic features identified using the Mann-Whitney U test, key microbes with a prevalence >20% and a proportion >0.001 at the species level were selected through filtration. Intermicrobial interactions were assessed using Spearman’s rank correlation test. A microbial co-occurrence network was created using the networkD3 R package.

Statistical Analysis

Continuous variables are presented as median with interquartile range (IQR), except for age as mean with standard deviation, and categorical variables are presented as numbers and proportions. Two-group differences in continuous variables were examined with the Mann-Whitney U test, except for age, which was analyzed with Student’s t-test, and categorical variables were examined with the Chi-square test. Multigroup differences in continuous variables and categorical variables were analyzed by Bonferroni corrected Dunn test and Benjamini-Hochberg correction test. Multivariable logistic regression analysis was performed to assess associations between stroke and presence of key microbes after adjustment for age, sex, smoking, hypertension, diabetes mellitus, and dyslipidemia. The Kaplan-Meier method was used to evaluate the cumulative incidence of major cardiovascular events. The estimated differences were compared using a log-rank test. Cox proportional hazards modelling was performed to identify independent risk factors for major cardiovascular events in stroke patients after adjustment for age, sex, smoking, hypertension, diabetes mellitus, and dyslipidemia. The likelihood ratio test was conducted to compare their predictive performances. Statistical analyses were performed using the GraphPad Prism software (version 9.0; San Diego, CA, USA) and R versions 4.2.1, 4.2.2, and 4.2.3 (Vienna, Austria). The unadjusted and adjusted odds ratios (ORs) or hazard ratio (HR) with 95% confidence intervals (CIs) were calculated. Statistical significance was set at two-sided P<0.05 and 95% CI.

Results

Baseline Characteristics of Study Participants

A total of 20 patients with acute stroke and 2 non-stroke participants were excluded because of missing samples. We used all of the collected 204 saliva samples (151 patients with acute stroke, 53 from non-stroke participants) and 160 fecal samples (109 patients with acute stroke, 51 non-stroke participants) (Figure 1). The prevalence of vascular risk factors, including smoking, hypertension, diabetes mellitus, dyslipidemia, and atrial fibrillation, was significantly higher in patients with acute stroke than in non-stroke subjects (Table 1).

Table 1.

Baseline Characteristics of the Study Population

	Gut microbiome			Salivary microbiome
	Non-stroke	Stroke	P value	Non-stroke	Stroke	P value
N	51	109		53	151
Age, years (SD)	74.3 (7.0)	71.8 (12.8)	0.19	74.4 (6.9)	72.4 (12.7)	0.19
Female, n (%)	32 (63)	44 (40)	<0.01	28 (53)	61 (41)	<0.01
Vascular risk factors
Smoking, n (%)	6 (12)	55 (51)	<0.01	7 (13)	79 (52)	<0.01
Drinking, n (%)	15 (29)	49 (45)	0.06	16 (30)	70 (45)	<0.01
Type 2 DM, n (%)	4 (8)	36 (33)	<0.01	5 (9)	43 (29)	<0.01
Hypertension, n (%)	20 (39)	94 (86)	<0.01	21 (40)	124 (82)	0.01
Dyslipidemia, n (%)	15 (29)	55 (51)	0.01	15 (28)	72 (48)	0.01
AF, n (%)	0 (0)	24 (22)	<0.01	0 (0)	40 (26)	<0.01
Stroke subtypes
ICH, n (%)		29 (27)			32 (21)
NIHSS, median (IQR)		3 (1–9)			3 (1–6)
Past medical history
Stroke, n (%)	0 (0)	25 (23)	<0.01	0 (0)	35 (23)	<0.01
IHD, n (%)	0 (0)	9 (8)	0.03	0 (0)	18 (12)	<0.01
CKD, n (%)	0 (0)	10 (9)	0.03	0 (0)	11 (7)	0.04

P values based on Student’s t-test for age and Chi-square test for other variables. AF, atrial fibrillation; CKD, chronic kidney disease; DM, diabetes mellitus; IHD, ischemic heart disease; IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale; SD, standard deviation.

Reduced Microbial Diversity of the Gut But Not of Saliva in Acute Stroke Patients

A comparison of the identified taxonomic features revealed no significant difference in the salivary microbiome (median [IQR], 250.0 [189.0–316.0] vs. 231.0 [205.5–292.5], P=0.71); however, patients with acute stroke had a significantly less diverse gut microbiome than non-stroke participants (median [IQR], 245.0 [135.5–339.0] vs. 279.0 [221.0–344.0], P=0.04) (Figure 2A). The status of alcohol consumption, smoking, hypertension, diabetes mellitus, dyslipidemia, and atrial fibrillation did not affect the diversity of taxonomic features in the salivary or gut microbiome (Supplementary Figure 1A–F).

Figure 2.

Acute stroke-related features of the salivary and gut microbiomes. (A) Dot plots of the taxonomic features of the salivary microbiome in non-stroke participants (n=53, green) and acute stroke patients (n=151, blue) and those of the gut microbiome in non-stroke participants (n=51, yellow) and acute stroke patients (n=109, red) (*P<0.05; Bonferroni corrected Dunn test). Volcano plots representing the proportion of taxonomic features based on the analysis of the composition of the microbiome in patients with acute stroke (red) and non-stroke participants (blue) at the species level of the (B) salivary and (C) gut microbiomes. Unknown species of microbes are not labeled. (D) Boxplots showing the proportion of Streptococcus anginosus in the salivary and gut microbiomes of patients with and without acute stroke (*P<0.01; ***P<0.001; Bonferroni corrected Dunn test).

Enrichment of Streptococcus Anginosus in Both the Saliva and Gut Microbiomes of Acute Stroke Patients

Next, we identified prominent salivary and gut microbes in acute stroke using microbiome composition analysis. In the salivary microbiome, Treponema denticola, Streptococcus cristatus, Prevotella loescheii, and S. anginosus were abundant in patients with acute stroke, whereas Eubacterium sulci, Streptococcus infantis, Haemophilus parainfluenzae, Prevotella nanceiensis, and Lancefieldella parvula were abundant in non-stroke participants at the species level (Figure 2B). In the gut microbiome, S. anginosus was prominently abundant in patients with acute stroke, whereas Anaerostipes hadrus and Bacteroides plebeius were abundant in non-stroke participants at the species level (Figure 2C). The proportion of S. anginosus in the salivary and gut microbiomes was significantly higher in patients with acute stroke than in non-stroke participants (median [IQR], 0.01 [0.00–0.14] vs. 0.00 [0.00–0.03], P=0.02; 0.09 [0.00–0.28] vs. 0.00 [0.00–0.02], P<0.001) (Figure 2D). Other species were comparable in the 2 groups for the salivary and gut microbiomes.

S. Anginosus as a Central Hub in Analysis of Gut Dysbiosis in Acute Stroke Patients

We used microbial co-occurrence network analyses to investigate the role of S. anginosus in the salivary-gut microbial communities of acute stroke patients (Figure 3).

Figure 3.

Microbial community structure based on microbial co-occurrence network analysis. The microbial co-occurrence network analysis represents the intermicrobial interaction in the salivary and gut microbiomes of acute-stroke subjects and non-stroke participants. The nodes represent key microbes in the salivary and gut microbiomes of acute stroke patients and non-stroke participants: 11 key microbes were detected in the salivary microbiomes of acute stroke patients (blue) and 4 in their gut microbiomes (red); 4 key microbes were detected in the salivary microbiomes of non-stroke participants (green) and 6 in their gut microbiomes (yellow). The diameter of the node represents the number of connections. The solid or dotted lines represent positive or negative correlations between nodes determined using Spearman correlation tests.

In the salivary microbiome, S. anginosus was part of a complex community associated with acute stroke. Multiple negative correlations were observed between microbes associated with the acute stroke and non-stroke groups (Figure 3, Upper panel). S. anginosus, Streptococcus mutans, and Lactobacillus fermentum were consistently detected, forming a community in the gut microbiome in acute stroke patients, while Anaerostipes hadrus, Bacteroides plebeius, Dialister invisus, Eubacterium rectale, Prevotella copri, and Sutterella wadsworthensis formed a community in the non-stroke participants. In the gut microbiome, we identified S. anginosus as a central hub, with a larger number of connections than other microbes, and positively correlated with Lactobacillus fermentum, and Streptococcus mutans colonization, and negatively correlated with Anaerostipes hadrus, Bacteroides plebeius, Eubacterium rectale, and Sutterella wadsworthensis colonization (Figure 3, Lower panel). Functional prediction showed microbial enrichment of lactic fermentation and glycolysis (Supplementary Figure 2).

S. Anginosus in the Gut as an Independent Biomarker of Acute Stroke

We comprehensively investigated the clinical significance of the 4 stroke-associated and 6 non-stroke-associated gut microbes in the microbial co-occurrence network analyses (Figure 3, Lower panel). Multivariable logistic regression analysis, adjusted for conventional vascular risk factors, including age, sex, smoking, hypertension, diabetes mellitus, and hyperlipidemia, and key gut microbes revealed that the presence of S. anginosus in the gut microbiome (adjusted OR [aOR], 1.18, 95% CI 1.05–1.32; P<0.01) was independently associated with stroke. In contrast, the presence of Anaerostipes hadrus and Bacteroides plebeius was inversely associated with stroke (0.82, [0.73–0.93], P<0.01; 0.86, [0.86–0.93], P=0.01) (Figure 4, Table 2).

Figure 4.

Association between acute clinical profiles and key gut microbes. Multivariable logistic regression analysis predicting the development of stroke associated with conventional vascular risk factors, age (per year), sex (male), smoking, hypertension, diabetes mellitus, dyslipidemia, and key microbes.

Table 2.

Multivariable Logistic Regression Analysis of Key Gut Microbes Associated With Stroke

	Model 1				Model 2		Model 3
	OR (95% CI)	BH-adj P value	aOR (95% CI)	P value	aOR (95% CI)	P value	aOR (95% CI)	P value
Detection of gut microbes (positive vs. negative)
Anaerostipes hadrus	0.74 (0.64–0.85)	<0.001	0.75 (0.64–0.86)	<0.001	0.79 (0.70–0.90)	<0.001	0.82 (0.73–0.93)	<0.01
Bacteroides plebeius	0.70 (0.61–0.80)	<0.001	0.69 (0.61–0.79)	<0.001	0.80 (0.71–0.90)	<0.001	0.86 (0.76–0.97)	0.01
Streptococcus anginosus	1.34 (1.15–1.55)	0.02	1.35 (1.17–1.56)	<0.001	1.20 (1.06–1.36)	<0.01	1.18 (1.05–1.32)	<0.01

BH-adj P value based on a t-test using the Benjamini-Hochberg method. P values based on multiple logistic regression analyses. aOR, adjusted odds ratio; CI, confidence interval. Model 1: adjusted for age (per year), sex (male); Model 2: adjusted for age (per year), sex (male), smoking, hypertension, diabetes mellitus, and dyslipidemia; Model 3: adjusted for age (per year), sex (male), smoking, hypertension, diabetes mellitus, dyslipidemia, and key microbes.

S. Anginosus as Potential Biomarker of Death and Major Cardiovascular Events

Finally, we longitudinally investigated the influence of gut dysbiosis on the prognosis of acute stroke patients. We analyzed the association between the presence of S. anginosus, Anaerostipes hadrus, and Bacteroides plebeius in the gut and death and major cardiovascular events following the reference stroke event. The log-rank test showed that S. anginosus in the gut indicated a significantly worse prognosis than for those without (P=0.04) (Figure 5A). In contrast, Bacteroides plebeius or Anaerostipes hadrus in the gut did not result in a better prognosis than in those without (P=0.19, P=0.45) (Figure 5B,C). S. anginosus in the gut had an increased risk of major cardiovascular events (adjusted HR, 4.78, 95% CI 1.08–21.18; P=0.04), after adjustment for vascular risk factors (Table 3). Furthermore, S. anginosus in the gut had superior prognostic performance compared with combined vascular risk factors (likelihood ratio test 5.19 vs. 3.65).

Figure 5.

Association between key gut microbes and death and major cardiovascular events. After initial gut microbiome assessments, a composite outcome of all-cause death or cardiovascular events was observed for approximately 2 years. Kaplan-Meier analyses show influences of positive (red) and negative (blue) detection of Streptococcus anginosus (A), positive (blue) and negative (red) detection of Bacteroides plebeius (B), and Anaerostipes hadrus (C) on the composite outcome of mortality and major cardiovascular events.

Table 3.

Cox Proportional Hazards Models for Univariate and Multivariate Factors for Post-Stroke Major Cardiovascular Events

	Univariate		Multivariate
	HR (95% CI)	P value	aHR (95% CI)	P value
Streptococcus anginosus	4.02 (0.94–17.1)	0.06	4.78 (1.08–21.2)	0.04
Age >75 years	1.23 (0.56–2.77)	0.60	1.53 (0.67–3.51)	0.32
Sex (male)	1.57 (0.66–3.72)	0.31	1.46 (0.40–5.37)	0.57
Smoking	1.56 (0.68–3.60)	0.30	1.42 (0.42–4.86)	0.57
Hypertension	0.55 (0.19–1.64)	0.29	0.60 (0.20–1.86)	0.38
Diabetes mellitus	1.58 (0.70–3.53)	0.27	1.28 (0.56–2.93)	0.55
Dyslipidemia	0.98 (0.44–2.19)	0.96	1.02 (0.45–2.34)	0.96

P values based on Cox proportional hazard model. aHR, adjusted HR; CI, confidence interval; HR, hazard ratio.

Discussion

S. anginosus plays a key role in gut dysbiosis in acute stroke, serving as a central hub in network analysis. We demonstrated that S. anginosus in the gut was associated with (a) stroke in the cross-sectional study, and (b) death and major cardiovascular events after stroke in the longitudinal study.

Gut microbial diversity, but not salivary, was reduced in acute stroke patients in this study. A reduction in microbial diversity is generally a characteristic of dysbiosis (i.e., imbalance in the microbiome). Previous studies have shown that microbial diversity continuously reduces as cardiometabolic disease progress.¹⁴ Therefore, gut microbial diversity in cardiovascular disease may be similar to that in cerebrovascular disease.

Our current research demonstrated that S. anginosus was abundant in both the salivary and gut microbiomes of the acute stroke patients. In particular, S. anginosus in the gut was identified as an independent predictor of stroke after adjustment for vascular risk factors. Previous research has reported that S. anginosus is abundant in the oral microbiome of individuals with hypertension,¹⁵ as well as in the gut microbiome of patients with asymptomatic atherosclerosis¹⁶ or coronary artery calcification.¹⁷ S. anginosus in the gut modulates metabolic byproducts (e.g., decrease in short-chain fatty acids or increase in trimethylamine N-oxide) during the development of atherosclerosis.¹⁶ A specific link between S. anginosus enrichment in the gut and subclinical coronary artery calcification correlated with elevated high-sensitive C-reactive protein.¹⁷ Among the diverse commensal Streptococcus species in the oral cavity, S. anginosus is particularly noted for its microaerophilic and acid-tolerant traits, which enables it to migrate and ectopically proliferate in the gut.¹⁸ Other research has identified S. anginosus as a pathogen that directly induces chronic mucosal inflammation in the gastrointestinal tract.¹⁹ Our findings highlighted that S. anginosus, in concert with S. mutans and Lactobacillus fermentum, may produce excess lactic acid through anaerobic respiration in the gut and indirectly cause systemic inflammation through the regulation of gene expression in T cells,²⁰ whereas protective microbes, such as Anaerostipes hadrus and Bacteroides plebeius, may produce short fatty acids, which presumably have anti-inflammatory effects.^21–23

Longitudinally, our research demonstrated that S. anginosus in the gut is predictive of death and cardiovascular events after stroke. Our findings further suggested that S. anginosus is a pathobiont in the gut, representing a previously unrecognized residual risk for cardiovascular disease and its poor prognosis.²⁴ To easily detect S. anginosus in the gut, utilization of 16rRNA gene sequencing tools, including specific microbe detection (e.g., SYMGRAM^®, Symbiosis Solutions Co., Ltd., Tokyo, Japan), would be a promising option for a future best clinical examination.

Study Limitations

First, in our acute stroke cohort, we were not able to collect enough information about lifestyle, including diet and socioeconomic status, due to consciousness disturbance, aphasia, or cognitive impairment in some patients. Second, the small sample size; however, in previous reports on patients with acute stroke, 1 included only 41 ischemic stroke patients and 40 control subjects,²⁵ and the other included only 79 patients with ischemic stroke and 98 healthy controls.²⁶ Therefore, to the best of our knowledge our cohort would be considered one of the largest human microbiome studies in the acute stroke clinical setting. Notwithstanding, a larger sample size is warranted to enhance statistical power. Third, our study included selection bias. The prevalence of vascular risk factors differed between the stroke and non-stroke participants. Therefore, we used logistic regression and Cox proportional hazards models to adjust for these confounding factors. Nonetheless, future studies should include non-stroke participants matched for vascular risk factors to reduce selection bias. Fourth, mechanistic data supporting S. anginosus as a causative agent were limited. An in vivo experimental study using gnotobiotic rodents with S. anginosus inoculation should be performed to elucidate the underlying mechanisms. Fifth, we did not have enough data about cardiovascular and diabetic patients in this study because we aimed to collect patients with acute stroke. We plan a collaboration with cardiologists to provide broader insights into the field of cardiology.

Acknowledgments

The graphical abstract was created using Biorender.com. We thank Editage (www.editage.jp) for editing the English language. We also thank Kokubo for his valuable comments, all the neurology, cerebrovascular medicine, and preventive cardiology staff for performing medical examinations and all cohort members.

Disclosures

None of the authors have conflicts of interest.

This research was supported by the Japan Agency for Medical Research and Development [grant number: JP21ek0210143 (Y.N.)]. This research was also supported by the Japan Society for the Promotion of Science [grant number: 20J15262 (S.T.)], Bristol Myers Squibb (M.I.), EA Pharma AS2021A000148143 (M.I.), Danone Institute of Japan Foundation for financial support with the 2022 FDIJF Research Grant (Y.H.), Ohyama Health Foundation (S.T., Y.H.), Takeda Science Foundation (Y.H.), and Senshin Medical Research Foundation (S.T.). Yakult Bio-Science Foundation (M.I.), the Intramural Research Fund (23-B-9, 20-4-9) for the cardiovascular diseases of the National Cerebral and Cardiovascular Center, and Japan Health Research Promotion Bureau (JH1-1).

IRB Information

The Research Ethics Committee of the National Cerebral and Cardiovascular Center (NCVC, Japan) (R20029) approved this study.

Data Availability

The deidentified participant data will not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-24-0872

References

• 1.   Ding T, Schloss PD. Dynamics and associations of microbial community types across the human body. Nature 2014; 509: 357–360.
• 2.   Lee JY, Tsolis RM, BäumLer AJ. The microbiome and gut homeostasis. Science 2022; 377: eabp9960.
• 3.   Fromentin S, Forslund SK, Chechi K, Aron-Wisnewsky J, Chakaroun R, Nielsen T, et al. Microbiome and metabolome features of the cardiometabolic disease spectrum. Nat Med 2022; 28: 303–314.
• 4.   GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke, 1990–2019: A systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol 2021; 20: 795–820.
• 5.   Tonomura S, Ihara M, Friedland RP. Microbiota in cerebrovascular disease: A key player and future therapeutic target. J Cereb Blood Flow Metab 2020; 40: 1368–1380.
• 6.   Honarpisheh P, Bryan RM, McCullough LD. Aging microbiota-gut-brain axis in stroke risk and outcome. Circ Res 2022; 130: 1112–1144.
• 7.   Atarashi K, Suda W, Luo C, Kawaguchi T, Motoo I, Narushima S, et al. Ectopic colonization of oral bacteria in the intestine drives TH1 cell induction and inflammation. Science 2017; 358: 359–365.
• 8.   Peh A, O’Donnell JA, Broughton BRS, Marques FZ. Gut microbiota and their metabolites in stroke: A double-edged sword. Stroke 2022; 53: 1788–1801.
• 9.   Nel Van Zyl K, Matukane SR, Hamman BL, Whitelaw AC, Newton-Foot M. Effect of antibiotics on the human microbiome: A systematic review. Int J Antimicrob Agents 2022; 59: 106502.
• 10.   Hirayama A, Konta T, Kamei K, Suzuki K, Ichikawa K, Fujimoto S, et al. Blood pressure, proteinuria, and renal function decline: Association in a large community-based population. Am J Hypertens 2015; 28: 1150–1156.
• 11.   Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable, and extensible microbiome data science using QIIME 2. Nat Biotechnol 2019; 37: 852–857.
• 12.   Douglas GM, Maffei VJ, Zaneveld JR, Yurgel SN, Brown JR, Taylor CM, et al. PICRUSt2 for prediction of metagenome functions. Nat Biotechnol 2020; 38: 685–688.
• 13.   Mandal S, Van Treuren W, White RA, Eggesbø M, Knight R, Peddada SD. Analysis of composition of microbiomes: A novel method for studying microbial composition. Microb Ecol Health Dis 2015; 26: 27663.
• 14.   Chakaroun RM, Olsson LM, Bäckhed F. The potential of tailoring the gut microbiome to prevent and treat cardiometabolic disease. Nat Rev Cardiol 2023; 20: 217–235.
• 15.   LaMonte MJ, Gordon JH, Diaz-Moreno P, Andrews CA, Shimbo D, Hovey KM, et al. Oral microbiome is associated with incident hypertension among postmenopausal women. J Am Heart Assoc 2022; 11: e021930.
• 16.   Jie Z, Xia H, Zhong SL, Feng Q, Li S, Liang S, et al. The gut microbiome in atherosclerotic cardiovascular disease. Nat Commun 2017; 8: 845.
• 17.   Sayols-Baixeras S, Dekkers KF, Baldanzi G, Jönsson D, Hammar U, Lin YT, et al. Streptococcus species abundance in the gut is linked to subclinical coronary atherosclerosis in 8973 participants from the SCAPIS cohort. Circulationl 2023; 148: 459–472.
• 18.   Poole PM, Wilson G. Infection with minute-colony-forming beta-haemolytic streptococci. J Clin Pathol 1976; 29: 740–745.
• 19.   Fu K, Cheung AHK, Wong CC, Liu W, Zhou Y, Wang F, et al. Streptococcus anginosus promotes gastric inflammation, atrophy, and tumorigenesis in mice. Cell 2024; 187: 882–896.
• 20.   Zhang D, Tang Z, Huang H, Zhou G, Cui C, Weng Y, et al. Metabolic regulation of gene expression by histone lactylation. Nature 2019; 574: 575–580.
• 21.   Shetty SA, Boeren S, Bui TPN, Smidt H, de Vos WM. Unraveling lactate-acetate and sugar conversion into butyrate by intestinal Anaerobutyricum and Anaerostipes species by comparative proteogenomics. Environ Microbiol 2020; 22: 4863–4875.
• 22.   Nishijima S, Suda W, Oshima K, Kim SW, Hirose Y, Morita H, et al. The gut microbiome of healthy Japanese and its microbial and functional uniqueness. DNA Res 2016; 23: 125–133.
• 23.   Hehemann JH, Correc G, Barbeyron T, Helbert W, Czjzek M, Michel G. Transfer of carbohydrate-active enzymes from marine bacteria to Japanese gut microbiota. Nature 2010; 464: 908–912.
• 24.   Fujiyoshi A, Kohsaka S, Hata J, Hara M, Kai H, Masuda D, et al. JCS 2023 Guideline on the primary prevention of coronary artery disease. Circ J 2024; 88: 763–842.
• 25.   Yamashiro K, Tanaka R, Urabe T, Ueno Y, Yamashiro Y, Nomoto K, et al. Gut dysbiosis is associated with metabolism and systemic inflammation in patients with ischemic stroke. PLoS One 2017; 12: e0171521.
• 26.   Li H, Zhang X, Pan D, Liu Y, Yan X, Tang Y, et al. Dysbiosis characteristics of gut microbiota in cerebral infarction patients. Transl Neurosci 2020; 11: 124–133.