https://orcid.org/0000-0002-7158-8687
Figures
Abstract
Oral microbiome (bacterial community) may influence systemic inflammation and vascular health, which both are critical factors in a pathogenesis of ischemic stroke. This study aimed to evaluate differences in the saliva microbiome of acute ischemic stroke (AIS) and transient ischemic attack (TIA) patients compared with matched healthy controls, hypothesizing that AIS and TIA patients are associated with oral microbiome shift. A prospective case-control study was conducted in Naresuan University Hospital, Thailand, to compare the saliva microbiome of AIS and TIA stroke patients of Thai ethnic with matched healthy controls. Microbial profiles were analyzed by metagenomics combined 16S rRNA gene sequencing to assess microbial alpha diversity, taxonomic composition, beta diversity, and microbial functional pathways.Forty-one patients (31 AIS and 10 TIA) and 20 age- and sex-matched stroke-free healthy controls were included in this study. Baseline characteristics were comparable between groups, apart from higher rates of hypertension, diabetes, and smoking in the patient group. Patients exhibited significantly higher alpha-diversity genus richness by OTUs and Chao1 index than controls (p < 0.001), highlighting an altered microbial community structure. Phylum-level analysis revealed an increased abundance of Bacillota (p = 0.0285) in the patient group, with a statistically decreasing trend for Bacteroidota, Actinomycetota and Pseudomonadota (p < 0.05). At the genus level, Streptococcus was more significantly abundant in the patients (p = 0.0171), while Prevotella was reduced. The patient and control groups were statistically separated in beta-diversity analysis (PERMANOVA, p < 0.001), with species biomarker analysis by LEfSe (Linear discriminant analysis effect size) could suggest species markers for each group. Functional pathway analysis showed the patient group the significantly higher in functional categories of, for examples, xenobiotics biodegradation and metabolism, cardiovascular diseases, signal transduction, and membrane transport (Welch’s t-test, p < 0.05). In conclusion, this study demonstrated the statistical alterations in the saliva microbiome of AIS and TIA patients, characterized by increased genus richness diversity and relatively distinct microbial shifts that may be associated with stroke-related inflammation. The findings suggest the saliva microbiome analysis as potential as a non-invasive biomarker for stroke risk and its role in stroke pathophysiology.
Citation: Roongpiboonsopit D, Wairit S, Nithisathienchai C, Pakdee A, Cheibchalard T, Sayasathid J, et al. (2025) Oral microbiome dysbiosis in acute ischemic stroke and transient ischemic attack patients. PLoS One 20(10): e0333676. https://doi.org/10.1371/journal.pone.0333676
Editor: Chun Wie Chong, Monash University Malaysia, MALAYSIA
Received: June 3, 2025; Accepted: September 16, 2025; Published: October 7, 2025
Copyright: © 2025 Roongpiboonsopit et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The 16S rRNA gene sequences in this study were deposited in the NCBI open access Sequence Read Archive database (accession number PRJNA1202695).
Funding: This research is supported by the Second Century Fund (C2F), the Multi-Omics for Functional Products in Food, Cosmetics and Animals Research Unit of Chulalongkorn University (funded to N. Somboonna), and the Health Systems Research Institute (HSRI) of Thailand (HSRI.65-059 and HSRI.66-152) (S. Tongsima). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Acute ischemic stroke (AIS) is a leading cause of mortality and long-term disability globally, impacting substantially to healthcare burdens [1,2]. Although well-established risk factors, including hypertension, diabetes, smoking, and dyslipidemia, play significant roles in stroke pathogenesis, emerging researches suggest that the dysbiosis in microbiome (bacterial community) may also involve stroke susceptibility and outcomes [3–5].
The human microbiome has increasingly been implicated important in various systemic diseases, including cardiovascular and cerebrovascular conditions. While the gut microbiome has been widely studied in stroke patients [5] and revealed potential links with immune dysregulation and systemic inflammation, the role of the oral microbiome, specifically the saliva microbiome, remains relatively unexplored. The oral cavity, home to a diverse microbial community, has been reported to influence systemic health through inflammatory pathways and bacteremia, making it as potential contributor to vascular health and stroke risk [6]. Alterations in the oral microbiome, or oral dysbiosis, may contribute to systemic inflammation, endothelial dysfunction, and atherosclerosis, which these are key factors in stroke development [7]. Since saliva can be collected easily and non-invasively, and may accurately reflect oral microbial health [8], it serves as an ideal medium for investigating potential associations between the oral microbiome and stroke pathophysiology. Dysbiosis shifts in microbial diversity and composition in the saliva of AIS and transient ischemic attack (TIA) stroke patients may reveal insights into inflammatory mechanisms relevant to stroke. Both AIS and TIA involve a blockage of blood flow to the brain, causing temporary (for TIA) or permanent (AIS) brain damage, respectively, and that the oral salivary microbiota may be associated with AIS and TIA stroke risk [9–11].
This study thereby investigated the differences in the saliva (oral) microbiome of AIS and TIA patients compared to healthy controls, focusing on microbial diversity, taxonomic composition, and microbial functional pathways. We hypothesize that stroke patients exhibit a distinct microbial profile that may contribute to systemic inflammation and immune modulation associated with stroke pathogenesis. The findings from this research supported further understanding of the oral microbiome’s diversity and role in cerebrovascular diseases, and may support future development for oral microbial biomarker and/or supportive therapeutic strategies for stroke prevention and management.
Materials and methods
Standard protocol and ethical approvals
This study was approved by the institutional review board (IRB) of the Naresuan University Hospital (IRB No. P3-0020/2565), following the Declaration of Helsinki. Written informed consents were obtained from all participants before collecting all data.
Study design and population
This prospective, case-control study was conducted at Naresuan University Hospital, Phitsanulok, Thailand, from August 2022 to October 2023. Sixty-one total subjects were included: 41 patients were consecutively recruited from the Acute Stroke Unit, while 20 controls were recruited from the Check-up Unit. Acute ischemic stroke (AIS) patients were defined as those with acute focal neurological deficits evidenced by brain infarction confirmed via computed tomography (CT) or magnetic resonance imaging (MRI). Transient ischemic attack (TIA) was defined acute focal neurological deficits lasting less than 24 hours without brain infarction on MRI scans. Inclusion criteria included: 1) age ≥ 45 years, 2) onset of stroke symptoms < 72 hours, 3) admission to the stroke unit, 4) available CT or MRI, and 5) written informed consent. Control participants were consecutive healthy individuals aged ≥ 45 years who attended annual check-ups and were matched to patients by age (± 5 years) and gender. Controls were recruited in a 2:1 ratio to stroke patients.
Exclusion criteria for both groups included: 1) altered consciousness or inability to provide consent, 2) history of stroke (controls only), 3) recent antibiotic use, 4) active infection, 5) temperature > 38°C, 6) white blood cell count > 15,000 cells/mm3, 7) cancer history, 8) abnormal kidney function (serum creatinine > 2 mg/dL), 9) autoimmune diseases, and 10) pregnancy or lactation. Demographic data and clinical data were collected from medical records and patient interviews. Data included gender, age, height, weight, history of hypertension, history of diabetes mellitus, history of dyslipidemia, smoking, history of previous stroke (for stroke patients). Laboratory investigation including fasting blood sugar (FBS), total cholesterol (TC), triglyceride (TG), low density lipoprotein (LDL), high density lipoprotein (HDL), blood urea nitrogen (BUN), creatinine (Cr), were measured. For stroke patients, etiology of stroke was classified according to the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria [12]. Stroke severity was assessed using the National Institute of Health Stroke Scale (NIHSS), and functional outcome was evaluated using the modified Ranking Scale (mRS) on admission and at discharge [13].
Saliva sampling collections
Saliva samples were collected in the morning before breakfast from the control group and within 72 hours of stroke onset (before any meal) in the patient group. Participants were instructed to sit comfortably, rinse their mouths with water, and spit 3 mL of saliva into a 50 mL sterile Falcon tube. Samples were then frozen at −20°C and transferred under continuous cold storage to the Microbiome Research Unit for Probiotics in Food and Cosmetics at Chulalongkorn University, Bangkok, Thailand.
Microbial analysis by metagenomic extraction and 16S rRNA gene sequencing
Metagenomic DNA was extracted from 1 mL of each sample and centrifuged (5,000 × g) for 5 minutes [14]. The cell pellet was extracted with a DNeasy PowerSoil kit according to the manufacturer’s instructions (Qiagen, Maryland, USA). DNA quantity and quality in the extracted metagenomes were determined by nanodrop spectrophotometry (A260 and A260/A280) and agarose gel electrophoresis. The metagenomic DNA was stored at −20°C.
Metagenomic DNA was used as a template for 16S rRNA gene V3–V4 region library preparation. Universal prokaryotic primers 341F (5′-CCTACGGGNGGCWGCAG-3′) and 805R (5′-GACTACHVGGGTATCTAATCC-3′) were used [15,16]. In brief, the 341F–805R PCR mixture comprised 2 × SparQ HiFi PCR Master Mix (QuantaBio, Massachesetts, USA), 0.2 μM 341F primer, 0.2 μM 805R primer, and 30 ng of the template. The PCR conditions were as follows: 98°C for 2 minutes, followed by 30 cycles of 98°C for 20 seconds, 60°C 30 seconds, 72°C for 60 seconds, and 72 °C for 1 minute. Then, each 16S rRNA gene V3–V4 amplicons were purified using SparQ Puremag Beads (QuantaBio), and appended index and barcode sequences via 8–10 cycles of PCR using 0.5 μM of Nextera TX index primers with the same PCR conditions. The final products were purified and pooled for the 2 × 250 bp next generation sequencing performed according to the Illumina MiSeq protocols (Illumina, San Diego, CA, USA) [17] at the Omics Sciences and Bioinformatics Center, Chulalongkorn University (Bangkok, Thailand).
Bioinformatic and statistical analyses
Bioinformatic analyses for bacterial microbiota diversity and potential metabolisms. Raw sequences were processed using Mothur version 1.46.1 according to the standard operating procedures for MiSeq [18,19]. Quality sequences were obtained by removing sequences with <100 base pairs or ≤ 8 ambiguous bases and chimera sequences, and were aligned to reference genomes in SILVA version 138.1 (updated taxonomic classification to SILVA version 138.2) and Greengenes database version 13.8 for removal of non-prokaryote sequences (mitochondria, chloroplast, eukaryote, and unclassified sequences). The quality sequences were phylotype-classified operational taxonomic units (OTUs) into phylum (abbreviated p_), class (c_), order (o_), family (f_), genus (g_) and species (s_) levels; and since the number of quality reads differed across samples, we normalized (rarefied) all samples to an equal sequencing depth (37,483 quality reads per sample) (S1 Fig). Noted that, due to limitations in the sequence length of the 16S rRNA gene V3–V4 region (~465 bp) and database resolution, some OTUs could not be confidently assigned a species (or genus) name; meanwhile these were still distinct OTUs classified at the species-level rank, although unnamed. The Good’s coverage index (estimated percent sequencing coverage to true biodiversity), alpha diversity (numbers of OTUs, Chao richness and Shannon diversity), beta diversity non-metric multidimensional scaling (NMDS) based on the Jclass dissimilarity coefficients, and linear discriminant analysis effect size (LEfSe) for species biomarkers, were computed using Mothur 1.39.5 [20,21]. Microbial metabolic functions were predicted from the microbiota profiles using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) (p < 0.05) and categorized on the basis of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways [22,23].
Clinical data statistics were performed using Stata/MP 14.0 software. Continuous variables with normal distributions are reported as mean ± standard deviation and were compared using an independent samples t-test. Categorical variables are presented as frequencies (percentages) and were analyzed with the χ2 test or Fisher’s exact test, as appropriate. Non-normally distributed continuous variables are reported as median (interquartile range). For microbiota statistics, t-test with centre-log-ratio transformation, permutational multivariate analysis of variance (PERMANOVA), or Tukey’s multiple comparisons test were used, and a p-value < 0.05 was considered significant [16,21]. For microbial metabolic potentials, the Statistical Analysis of Metagenomic Profiles (STAMP) was used [23]. Graph pictures were prepared by GraphPad Prism 9.0 (San Diego, California, USA).
Results
Sixty-one total participants were enrolled, including 41 patients (categorized into 31 stroke and 10 TIA), and 20 matched healthy control participants. Baseline characteristics in age, gender, or BMI were comparable between the patient and control groups, with no statistically significant difference (Table 1, p > 0.05). However, the patients exhibited significantly higher systolic and diastolic blood pressure, and the higher prevalence of hypertension, diabetes, smoking, and lower LDL. Small vessel disease was the most common stroke subtype in the patient group. The median stroke severity in the stroke group, measured by NIHSS was 4 (Table 1).
Microbiota sequencing result and the compositions of oral microbiota between patients and control groups
The number of quality reads were high (average 98,238 reads in patient group, and 70,849 reads in control group), and hence all sequencing samples showed Good’s coverage indices at genus OTUs above 99% and the relatively plateau rarefaction curves at phylum and genus levels (S1 Table and S1 Fig), indicating sufficient sequencing depth to true biodiversity for microbiome data analysis.
As for the compositions, phylum Bacillota predominated in both groups with the statistically higher in stroke individuals (Fig 1A: average 50.31 ± 7.96% in control group, 53.79 ± 8.49% in stroke group) (t-test, p = 0.0285), followed by Bacteroidota(19.36 ± 6.23% control, 16.57% ± 7.44% stroke; p = 0.0006), Actinomycetota (13.04 ± 2.81% control, 12.62 ± 5.48% stroke; p = 0.0058), Pseudomonadota (9.98 ± 6.96% control, 7.83 ± 8.51% stroke; p = 0.0013), and Fusobacteria (4.13 ± 3.17% control, 5.73 ± 4.11% stroke), in orderly. While phylum Bacillota was significantly more abundant in stroke patients, Bacteroidota, Actinomycetota and Pseudomonadota showed a statistically decreasing trend; and Fusobacteria showed slightly more prevalent in stroke group (Fig 1A).
OTUs of less than 0.01% in phylum (or genus) were in Others. Symbol “*” after OTU name represented phylum (or genus) with the statistically different relative abundances between control and stroke groups (p < 0.05). In (B), the OTU where Mothur could not identify the genus name was denoted by small letter to the deepest taxonomic names that could be identified (k_ abbreviates kingdom; and g_, genus).
The genus level OTUs demonstrated the statistically significant difference in the overall microbiota compositions between the control and stroke groups (PERMANOVA, p < 0.001) (Fig 1B). Prevotella was lower (average 14.87 ± 6.58% in control group, 11.30 ± 6.63% in stroke group) and Streptococcus was statistically higher (27.46 ± 7.43% control, 29.89 ± 10.57% stroke) (t-test, p = 0.0171) in stroke group. Overall, genera Streptococcus, Prevotella and Veillonella (10.81 ± 5.88% control, 9.16 ± 6.07% stroke) predominated, followed by Rothia, Neisseria, Actinomyces, Porphyromonas and Fusobacterium, respectively (Fig 1B).
Analyzing the top ten predominant species from the saliva samples belonged in genera Streptococcus, Prevotella, Actinomyces, Fusobacterium, Veillonella, Gemella, Porphyromonas, Neisseria and Rothia (Fig 2). A significant increase in the relative abundance of Streptococcus infantis (t-test, p = 0.0487) and a clear decrease in the relative abundance of Prevotella melaninogenica (p > 0.05), were observed in the stroke group.
The p-value was displayed only for the species with the statistically different relative abundances between groups (p < 0.05), and the OTUs where Mothur could not identify the species were denoted by small letter to the deepest taxonomic names that could be identified (g_ abbreviates genus).
Alpha and beta diversity analyses
In alpha diversity analysis to assess the diversity degree in individual samples, the Shannon diversity showed non-different diversity. Nonetheless, the number of OTUs and Chao1 genus richness were greater in the patients (p < 0.001) (Fig 3), highlighted the greater numbers of OTUs in stroke patients.
Statistic comparisons were performed by t-test between patient and healthy control groups (p < 0.05).
Consistent with the statistic microbiota comparison between groups in Fig 1B, the beta diversity non-metric multidimensional scaling (NMDS) revealed relatively separate clusters for the control vs. the stroke groups (PERMANOVA, p < 0.001) (Fig 4A). The saliva microbiota structures in control individuals exhibited intra-group similarity, while wider distributions in stroke microbiota individuals might reflects their OTUs and Chao1 richness (Fig 4A).
Oral bacterial species biomarkers for the control vs. stroke groups
Oral bacterial species OTU biomarkers determined by the linear discriminant analysis effect size (LEfSe) for the control vs. stroke groups, suggested Prevotella melaninogenica, Neisseria subflava, Veillonella parvula and Haemophilus parainfluenzae were enriched in the control group. Conversely, the stroke group was enriched in unidentified species in genus Granulicatella, Porphyromonas endodontalis, Rothia dentocariosa, Streptococcus anginosus, Prevotella nigrescens and Atopobium rimae, for examples (Fig 4B).
Microbial metabolic potentials for the control vs. stroke groups
Several differences in the microbial related metabolic processes were estimated between the control and stroke individuals, including genetic information processing and environmental information processing (Welch’s t-test, p < 0.05) (Fig 5A). For a subcategory (Fig 5B, KEGG level 2), the healthy control generally showed greater microbial metabolic potentials in most functions, for instances, replication and repair, energy metabolism, and cellular processes and signaling. In contrast, the stroke patients showed the greater in membrane transport, xenobiotics biodegradation and metabolism, cardiovascular diseases, and signal transduction (Fig 5B). Many of the microbial metabolic functions were consistent with the clinical characteristics of stroke (i.e., cardiovascular diseases, xenobiotics biodegradation and metabolism) vs. healthy (i.e., replication and repair).
Only the statistically different metabolic functions between groups were displayed (Welch’s t-test with Benjamini-Hochberg False Discovery Rate correction, p < 0.05).
Discussion
Meanwhile some overlap similarities, this study highlighted differences in the oral (saliva) microbiome in AIS or TIA stroke patients compared with healthy controls, along their differential clinical characteristics. The data help provided a novel finding association between oral microbiome dysbiosis and stroke etiology.
The stroke individuals showed relatively significant greater number of OTUs and Chao1 richness, along the statistical distinct microbial compositions and beta diversity clustering, from the healthy controls, all indicating an oral saliva microbiome dysbiosis in the stroke compared with the healthy. The oral cavity of stroke patients may harbor a more diverse bacterial colonization, including potential pathogens and opportunistic pathogens, which could contribute to the acute post-stroke inflammatory environment and potentially influence systemic effects in AIS and TIA [6]. It is now well recognized that microbiome dysbiosis is associated with both disease and health [24,25]. Saliva microbiome diversity appears to vary across different stages of stroke. Preclinical studies have demonstrated an inverse association between microbial richness and stroke risk [26], whereas elevated diversity has been reported during the hyperacute phase (within 24 hours of stroke onset) [27] and the early subacute phase (within 2 weeks post-stroke) [28]. Our findings of increased diversity within 72 hours post-stroke further support the notion that microbial diversity dynamically shifts with disease progression and inflammatory responses.
At the phylum level, our study demonstrated an increase in Bacillota, particularly Streptococcus, and a decrease in Bacteroidota, Actinomycetota, Pseudomonadota, and Prevotella in stroke patients, consistent with findings from previous studies [27,28]. Elevated levels of Streptococcus have been associated with systemic inflammation and atherosclerosis, with Streptococcus sanguinis linked to aortic inflammation [29] and Streptococcus viridans implicated in the formation of atherosclerotic plaques [30]. Additionally, Streptococcus mutans carrying the cnm gene has been associated with cerebral microbleeds, suggesting a role in localized inflammation [31]. However, our results differ from those reported in cases of cryptogenic stroke, where Prevotella was found to be more abundant [32]. This discrepancy may reflect differences in stroke etiology, as cryptogenic stroke is often associated with non-atherosclerotic mechanisms and lower levels of systemic inflammation; and additionally, the difference in Prevotella is non-statistically significant in this finding.
Microbial-related functional pathway analysis predicted significant enrichment of pathways such as environmental information processing, cardiovascular diseases, signal transduction, membrane transport, and xenobiotics biodegradation and metabolism in the saliva microbiome of stroke patients. These shifts may reflect microbial adaptation to the pro-inflammatory and oxidative stress conditions present in AIS and TIA. In contrast, healthy controls exhibited higher predicted microbial functional potentials related to genetic information processing, energy metabolism, cellular processes and signaling, as well as replication and repair. These functions have been linked to microbial community stability and homeostatic support of host health in non-disease states [9,25]. It is important to note that such microbial functions were inferred using PICRUSt-based predictions from the 16S rRNA gene sequences, which estimated gene content and pathway potentials based on taxonomic profiles and reference genome databases [22,23]. Therefore, while these inferences are widely used, they represent approximations rather than direct functional measurements. The observed enrichment of predicted pathways involving xenobiotics metabolism and cardiovascular diseases in stroke patients also align with clinical features of stroke and systemic inflammation [33–35]. Meanwhile, the reduced potentials for replication and repair and energy metabolism in the stroke-associated microbiome may suggest signs of microbial community stress or dysfunction under pathological conditions [4,36]. Collectively, the results highlight the potential role of microbial dysbiosis contributes to disease progression by promoting inflammation and activating stress-response pathways and/or responses post-strokes.
Furthermore, the distinct microbial profiles in stroke patients suggest that saliva microbiome analysis could serve as a non-invasive screening for stroke risk stratification. Additionally, targeting the oral microbiome through interventions such as oral hygiene practices, dietary modifications, or probiotic therapies might help mitigate systemic inflammation and potentially reduce stroke risks [9–11].
Although direct causal links between the oral microbiome dysbiosis and stroke pathophysiology remain to be fully elucidated, this study findings parallel with patterns observed in other inflammatory-driven diseases. This study still posed limitations, including small sample size, and potential confounding factors such as diet, oral hygiene practices and periodontal status, which could influence microbiota compositions, and 1 sample (S_s15) was later diagnosed confounded by migraine with aura. While a thirteen-year longitudinal study in Thailand did not report a significant association between periodontitis and stroke incidence [37,38], oral health represents one key and may influence systemic inflammation. Future research should prospectively include dental and periodontal assessments to recruit this potential variable factor from the microbiota anaylsis. Additionally, the use of 16S rRNA gene sequencing, while valuable for taxonomic profiling, limits the full resolution of functional characterization and does not capture viral or fungal components of the microbiome.
Periodontitis is an inflammatory condition of the oral cavity that has been consistently associated with cardiovascular risk [6,37,38]. Oral pathogens such as Porphyromonas, Prevotella, Fusobacterium, and Streptococcus species can translocate into the bloodstream, and contributed to endothelial dysfunction, atherogenesis, and systemic immune activation [7,33]. In particular, Porphyromonas endodontalis and Prevotella nigrescens, which were enriched in our stroke cohort, are well-recognized periodontal pathogens that can stimulate pro-inflammatory cytokine production and have been implicated in both periodontal tissue destruction and vascular inflammation. Similarly, Streptococcus anginosus has been linked to oral infections and bacteremia, with potential cardiovascular involvement. Although we did not directly assess periodontal status in our participants, the overlap between stroke-associated oral taxa and known periodontal pathogens suggests a possible mechanistic bridge between periodontitis, oral microbiome dysbiosis, and stroke pathophysiology. Future studies should include alongside the dental and periodontal assessments to clarify this link and to evaluate whether periodontal management could mitigate stroke risk.
Conclusion
This study demonstrated distinct alterations in the saliva microbiome profiles of patients with AIS and TIA compared to healthy controls. Stroke patients exhibited greater microbial richness, significant microbiota shifts, and enrichment of microbial pathways associated with inflammation, cardiovascular diseases, and environmental stress responses. These findings suggest that oral microbiome dysbiosis may contribute to systemic inflammatory environments and potentially influence stroke pathophysiology. Although causal relationships remain to be established, the observed associations highlight the potential of the oral microbiome both as a non-invasive biomarker for stroke risk assessment and as a novel target for supportive intervention and preventive care. Future large-scale, multi-center, and longitudinal studies incorporating multi-omics and functional analyses are warranted to further elucidate the role of the oral microbiome in stroke development and progression.
Supporting information
S1 Fig. Rarefaction curves of quality reads per sample at (A) phylum and (B) genus level operational taxonomic units (OTUs), in patient and control samples.
Noted that since the number of quality reads differed across samples, we normalized all samples to an equal sequencing depth (37,483 quality reads per sample).
https://doi.org/10.1371/journal.pone.0333676.s001
(TIFF)
S1 Table. Saliva sample IDs, group, number of quality reads, Good’s coverage indices at genus OTUs, and subject information.
https://doi.org/10.1371/journal.pone.0333676.s002
(DOCX)
References
- 1. Li X-Y, Kong X-M, Yang C-H, Cheng Z-F, Lv J-J, Guo H, et al. Global, regional, and national burden of ischemic stroke, 1990-2021: an analysis of data from the global burden of disease study 2021. EClinicalMedicine. 2024;75:102758. pmid:39157811
- 2. Feigin VL, Norrving B, Mensah GA. Global Burden of Stroke. Circ Res. 2017;120(3):439–48. pmid:28154096
- 3. 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(2):e0171521. pmid:28166278
- 4. Banerjee A, Mohapatra S, McCullough LD. Stroke and the Microbiome. Stroke. 2024;55(3):762–4. pmid:38258571
- 5. Peh A, O’Donnell JA, Broughton BRS, Marques FZ. Gut Microbiota and Their Metabolites in Stroke: A Double-Edged Sword. Stroke. 2022;53(5):1788–801. pmid:35135325
- 6. Bui FQ, Almeida-da-Silva CLC, Huynh B, Trinh A, Liu J, Woodward J, et al. Association between periodontal pathogens and systemic disease. Biomed J. 2019;42(1):27–35.
- 7. Leonov G, Salikhova D, Starodubova A, Vasilyev A, Makhnach O, Fatkhudinov T, et al. Oral Microbiome Dysbiosis as a Risk Factor for Stroke: A Comprehensive Review. Microorganisms. 2024;12(8):1732. pmid:39203574
- 8. Belstrøm D. The salivary microbiota in health and disease. J Oral Microbiol. 2020;12(1):1723975. pmid:32128039
- 9. Rosier BT, Marsh PD, Mira A. Resilience of the Oral Microbiota in Health: Mechanisms That Prevent Dysbiosis. J Dent Res. 2018;97(4):371–80. pmid:29195050
- 10. Scannapieco FA, Cantos A. Oral inflammation and infection, and chronic medical diseases: implications for the elderly. Periodontol 2000. 2016;72(1):153–75. pmid:27501498
- 11. Devine DA, Marsh PD. Prospects for the development of probiotics and prebiotics for oral applications. J Oral Microbiol. 2009;1:10.3402/jom.v1i0.1949. pmid:21523212
- 12. Rathburn CM, Mun KT, Sharma LK, Saver JL. TOAST stroke subtype classification in clinical practice: implications for the Get With The Guidelines-Stroke nationwide registry. Front Neurol. 2024;15:1375547. pmid:38585349
- 13. Banks JL, Marotta CA. Outcomes validity and reliability of the modified Rankin scale: implications for stroke clinical trials: a literature review and synthesis. Stroke. 2007;38(3):1091–6. pmid:17272767
- 14. Rosenbaum J, Usyk M, Chen Z, Zolnik CP, Jones HE, Waldron L, et al. Evaluation of Oral Cavity DNA Extraction Methods on Bacterial and Fungal Microbiota. Sci Rep. 2019;9(1):1531. pmid:30728424
- 15. Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41(1):e1. pmid:22933715
- 16. Wongsamart R, Somboonna N, Cheibchalard T, Klankeo P, Ruampatana J, Nuntapaitoon M. Probiotic Bacillus licheniformis DSMZ 28710 improves sow milk microbiota and enhances piglet health outcomes. Sci Rep. 2025;15(1):17. pmid:39747535
- 17.
Illumina. 16S metagenomic sequencing library preparation. Preparing 16S ribosomal RNA gene amplicons for the Illumina MiSeq System. 2013. Available from: https://support.illumina.com
- 18. Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol. 2013;79(17):5112–20. pmid:23793624
- 19. Wongsaroj L, Chanabun R, Tunsakul N, Prombutara P, Panha S, Somboonna N. First reported quantitative microbiota in different livestock manures used as organic fertilizers in the Northeast of Thailand. Sci Rep. 2021;11(1):102. pmid:33420281
- 20. Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J, et al. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res. 2007;35(21):7188–96. pmid:17947321
- 21. Cheibchalard T, Leelahavanichkul A, Chatthanathon P, Klankeo P, Hirankarn N, Somboonna N. Fungal microbiome in gut of systemic lupus erythematosus (SLE)-prone mice (pristane and FCGRIIb deficiency), a possible impact of fungi in lupus. PLoS One. 2024;19(12):e0314662. pmid:39637140
- 22. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30. pmid:10592173
- 23. Parks DH, Tyson GW, Hugenholtz P, Beiko RG. STAMP: statistical analysis of taxonomic and functional profiles. Bioinformatics. 2014;30(21):3123–4. pmid:25061070
- 24. Wilkins LJ, Monga M, Miller AW. Defining Dysbiosis for a Cluster of Chronic Diseases. Sci Rep. 2019;9(1):12918. pmid:31501492
- 25. Human Microbiome Project Consortium. Structure, function and diversity of the healthy human microbiome. Nature. 2012;486(7402):207–14. pmid:22699609
- 26. Wang C, Yang Y, Cai Q, Gao Y, Cai H, Wu J, et al. Oral microbiome and ischemic stroke risk among elderly Chinese women. J Oral Microbiol. 2023;15(1):2266655. pmid:37822701
- 27. Sun W, Huang S, Yang X, Luo Y, Liu L, Wu D. The oral microbiome of patients with ischemic stroke predicts their severity and prognosis. Front Immunol. 2023;14:1171898. pmid:37138888
- 28. Boaden E, Lyons M, Singhrao SK, Dickinson H, Leathley M, Lightbody CE, et al. Oral flora in acute stroke patients: A prospective exploratory observational study. Gerodontology. 2017;34(3):343–56. pmid:28543778
- 29. Hashizume-Takizawa T, Yamaguchi Y, Kobayashi R, Shinozaki-Kuwahara N, Saito M, Kurita-Ochiai T. Oral challenge with Streptococcus sanguinis induces aortic inflammation and accelerates atherosclerosis in spontaneously hyperlipidemic mice. Biochem Biophys Res Commun. 2019;520(3):507–13. pmid:31610917
- 30. Patrakka O, Tuomisto S, Pienimäki J-P, Ollikainen J, Oksala N, Lampinen V, et al. Thrombus Aspirates From Patients With Acute Ischemic Stroke Are Infiltrated by Viridans Streptococci. J Am Heart Assoc. 2023;12(22):e030639. pmid:37982253
- 31. Tonomura S, Ihara M, Kawano T, Tanaka T, Okuno Y, Saito S, et al. Intracerebral hemorrhage and deep microbleeds associated with cnm-positive Streptococcus mutans; a hospital cohort study. Sci Rep. 2016;6:20074. pmid:26847666
- 32. Manzoor M, Leskelä J, Pietiäinen M, Martinez-Majander N, Ylikotila P, Könönen E, et al. Multikingdom oral microbiome interactions in early-onset cryptogenic ischemic stroke. ISME Commun. 2024;4(1):ycae088. pmid:38988699
- 33. Xu Q, Wang W, Li Y, Cui J, Zhu M, Liu Y, et al. The oral-gut microbiota axis: a link in cardiometabolic diseases. NPJ Biofilms Microbiomes. 2025;11(1):11. pmid:39794340
- 34. Koh JY, Choi DW. Cultured striatal neurons containing NADPH-diaphorase or acetylcholinesterase are selectively resistant to injury by NMDA receptor agonists. Brain Res. 1988;446(2):374–8. pmid:2836034
- 35. Rizzatti G, Lopetuso LR, Gibiino G, Binda C, Gasbarrini A. Proteobacteria: A Common Factor in Human Diseases. Biomed Res Int. 2017;2017:9351507. pmid:29230419
- 36. Li Y, Zhu M, Liu Y, Luo B, Cui J, Huang L, et al. The oral microbiota and cardiometabolic health: A comprehensive review and emerging insights. Front Immunol. 2022;13:1010368. pmid:36466857
- 37. Tiensripojamarn N, Lertpimonchai A, Tavedhikul K, Udomsak A, Vathesatogkit P, Sritara P, et al. Periodontitis is associated with cardiovascular diseases: A 13-year study. J Clin Periodontol. 2021;48(3):348–56. pmid:33386631
- 38. Holmlund A, Lampa E, Lind L. Oral health and cardiovascular disease risk in a cohort of periodontitis patients. Atherosclerosis. 2017;262:101–6. pmid:28531825