A detailed re-analysis of salivary bacterial profiles from adult male volunteers has revealed that habitual areca-nut chewers display a significantly altered oral microbial community composition compared to non-chewing individuals, even though overall bacterial diversity remains unchanged between the two cohorts. The investigation, based on high-throughput sequencing of the bacterial 16S ribosomal RNA gene, identified twenty-one bacterial genera that exhibited differential abundance across groups under strict false discovery rate control. However, the study simultaneously documented a near-complete overlap between areca-nut chewing and cigarette smoking, alongside distinct clinical recruitment pathways between the cohorts. These pervasive confounding factors mean that the observed biological distinctions reflect broad associations with areca-nut user status rather than isolated causal effects of the chewed nut itself, providing a methodologically rigorous framework for interpreting oral microbiome shifts within addiction-exposed populations.
Areca nut represents the harvested seed of the areca palm tree, Areca catechu, and is colloquially termed betel nut due to its historical custom of being wrapped and consumed within leaves of the betel vine, Piper betle. Globally, several hundred million individuals consume areca nut on a daily basis, establishing it as one of the most widely utilized psychoactive substances worldwide after caffeine, nicotine, and alcohol. Prolonged and chronic mastication of areca nut is causally linked to oral submucous fibrosis, diverse oral potentially malignant disorders, and oral squamous cell carcinoma, leading international health agencies to classify areca nut as a Group 1 human carcinogen. While traditional consumption across South Asia, Southeast Asia, and the Western Pacific often incorporates slaked lime, tobacco, and betel leaves, the commercial market across mainland China has popularized packaged, tobacco-free processed areca-nut products containing flavorings and slaked lime, creating a distinctive chemical exposure profile.
Prior scientific investigations examining how areca-nut consumption alters the oral bacterial community have produced contradictory and heterogeneous findings across different geographic regions and populations. Earlier studies conducted in diverse international settings, such as multi-site evaluations in India and salivary metagenomic sequencing in Papua New Guinea, reported higher microbial alpha diversity among chewers, whereas clinical investigations in Guam observed lower diversity in long-term chewers after statistical adjustment for smoking. Many published studies in this field have been constrained by modest sample sizes, an absence of rigorous multiple-testing corrections, and an inability to untangle the microbial effects of areca nut from the potent ecological influence of concurrent tobacco use. To resolve these discrepancies, researchers re-analyzed salivary amplicon sequencing data from forty male volunteers using a fully scripted, compositionally aware analytical pipeline implemented directly in the R statistical environment, relying on DADA2 denoising, ALDEx2 differential abundance testing, and comprehensive permutation checks rather than standard packaging workflows.
The study cohort comprised forty male volunteers enrolled between December 2024 and April 2025, specifically balanced into twenty habitual areca-nut chewers and twenty healthy, non-chewing control participants. Habitual chewers reported consuming commercial processed Chinese areca-nut brands, including Kouweiwang and Hesetianxia, for a minimum duration of six months. Enrollment was intentionally restricted to male individuals because habitual areca-nut use in the source population represents a predominantly male behavioral pattern, thereby eliminating biological sex-related variations in salivary microbial ecology within a modest sample size. Participants in the areca-nut group were recruited through convenience sampling from patients attending the Department of Oral and Maxillofacial Surgery at The First Affiliated Hospital of Anhui Medical University during outpatient consultations unrelated to surgical operations, whereas control participants were recruited directly from the surrounding community. Inclusion criteria mandated that all participants be at least eighteen years of age, while prospective subjects were excluded if they had received systemic antibiotic therapy within three months of sampling, had a history of diabetes, immunodeficiency, or major systemic illness, or had undergone prior head and neck radiotherapy or chemotherapy.
Salivary collection protocols required all forty participants to provide one milliliter of unstimulated saliva between the hours of 09:00 and 11:30 in the morning. Each subject was instructed to strictly avoid eating food, drinking liquids, chewing areca nut, and using oral mouth rinses for at least two hours prior to donation, followed by ten minutes of quiet rest immediately before sample collection. Total bacterial genomic DNA was extracted from the recovered saliva samples utilizing the commercial FastPure Saliva DNA Isolation Kit, after which structural integrity was confirmed through agarose gel electrophoresis and optical purity was quantified on a spectrophotometer. Following gel purification and microplate fluorometric quantification, paired-end libraries were constructed and sequenced on an Illumina NextSeq 2000 platform, generating high-coverage paired-end reads measuring three hundred base pairs in length.
Sequence processing was executed through a reproducible bioinformatics pipeline in R utilizing cutadapt for automated primer removal, which successfully retained an average of 99.8 percent of read pairs. DADA2 denoised the quality-filtered reads into exact amplicon sequence variants, merged overlapping read pairs, and eliminated chimeric sequences, ultimately yielding a mean of 28,956 non-chimeric reads per sample. After screening and discarding non-bacterial, mitochondrial, and chloroplast lineages, 5,249 distinct amplicon sequence variants remained across the entire dataset. To correct for uneven sequencing depth while ensuring valid ecological comparisons, all forty samples were rarefied to the minimum library depth of 23,782 reads, which preserved a Good’s coverage metric exceeding 99.8 percent for every individual library. Taxonomic assignments were systematically resolved against the SILVA v138.2 reference database, which incorporates updated taxonomic nomenclature that delineates historical microbial classifications into modern, refined generic boundaries.
Evaluation of alpha diversity revealed that overall microbial richness and evenness did not differ significantly between areca-nut chewers and non-chewing controls across any evaluated mathematical index. Non-parametric Wilcoxon rank-sum tests performed on Observed amplicon sequence variants, Chao1 richness, Shannon diversity, Simpson index, and Inverse Simpson values all yielded uniform Benjamini–Hochberg adjusted p-values of 0.49, accompanied by negligible Cliff’s delta effect sizes below 0.18. Furthermore, two one-sided equivalence tests conducted across a pre-specified medium-effect margin failed to establish statistical equivalence, indicating that while no overt diversity divergence existed, the cohort size of forty men possessed eighty percent statistical power only to detect relatively large standardized shifts corresponding to a Cohen’s d of 0.9. Consequently, the authors emphasized that the observed absence of alpha-diversity differences demonstrates a lack of detectable divergence at this specific sample size rather than definitive evidence of biological equivalence across populations.
In marked contrast to overall diversity metrics, community composition and beta diversity separated the two cohorts with robust statistical significance. Permutational multivariate analysis of variance based on Bray–Curtis dissimilarity matrices revealed a significant grouping effect that accounted for 17.8 percent of the total variance in community structure, characterized by an R-squared value of 0.178 and a permutation p-value of 0.001. Tests for the homogeneity of multivariate dispersions showed only a marginal, non-significant difference between groups, confirming that the separation was primarily driven by a true directional shift in the community centroid rather than unequal within-group scatter. Jaccard distance metrics similarly confirmed group separation, though with a smaller variance explained of 4.8 percent. Notably, a technical sensitivity analysis established that read retention during sequence denoising was systematically lower in chewers than in controls, averaging 52.8 percent compared to 57.5 percent; adjusting for this read retention halved the Bray–Curtis grouping effect size to an R-squared of 0.081, while the association remained statistically significant at p = 0.001.
Taxonomic profiling at the broad phylum level revealed descriptive shifts in the salivary landscape, characterized by higher mean relative abundances of Bacillota at 41.6 percent versus 29.9 percent and Actinomycetota at 11.4 percent versus 5.8 percent among areca-nut chewers, alongside a marked reduction in Pseudomonadota from 19.3 percent in controls down to 7.0 percent in chewers. When differential abundance was evaluated at the genus level using ALDEx2 centered log-ratio transformed counts with Benjamini–Hochberg false discovery rate control, twenty-one genera demonstrated significant alterations. Two primary large-effect organisms positioned well above the baseline limit of detection dominated this signature: the genus Actinomyces was significantly increased in the areca-nut cohort, showing a mean abundance shift from 0.48 percent in controls up to 2.31 percent in chewers, whereas Porphyromonas was significantly decreased, dropping from 9.63 percent in controls down to 2.85 percent in chewers. Both primary shifts were verified across leave-one-out cross-validations and CLR linear modeling, retaining statistical significance even after adjusting for differences in sequence read retention.
The pronounced reduction of Porphyromonas, alongside minor downward trends in Fusobacterium, directly challenges the intuitive clinical assumption that habitual areca-nut chewing or deteriorating oral health simply fosters an overgrowth of classical periodontal pathogens. Instead, the taxonomic profile reflects a broad structural rearrangement of the oral habitat toward an aerobe-depleted, Bacillota-weighted community. Several nitrate-reducing aerobic commensal bacteria—specifically the genera Neisseria, Haemophilus, and Capnocytophaga—exhibited substantial depletion among areca-nut users in the baseline differential abundance models. Neisseria dropped from an average relative abundance of 10.6 percent in controls to 4.5 percent in chewers, Haemophilus declined from 8.8 percent to 1.8 percent, and Capnocytophaga decreased from 1.5 percent to 0.6 percent. However, these specific aerobic commensal signals did not maintain statistical significance after analytical adjustment for read retention, underscoring that smaller-effect genera remain exploratory and sensitive to technical modeling parameters.
Crucially, the observed depletion of Neisseria, Haemophilus, and Capnocytophaga exactly recapitulates the widely documented oral microbiome signature of tobacco smoking. Epidemiological evaluation of the study participants revealed near-perfect confounding between the two behavioral exposures: nineteen of the twenty areca-nut chewers were active cigarette smokers, whereas zero of the twenty control subjects smoked, representing a statistically profound disparity confirmed by Fisher’s exact test. Because areca-nut consumption and tobacco inhalation co-occurred almost universally within the sampled chewing group, their independent microbial consequences cannot be disentangled or mathematically adjusted in this dataset. The research team noted that permutation tests and leave-one-out analyses successfully confirm that the taxonomic differences are not statistical artifacts of sequencing noise, but these algorithms remain entirely blind to underlying epidemiological confounding, preventing any direct attribution of these shifts solely to areca nut.
Substantial clinical and recruitment disparities further define the operational boundaries of the investigation. While healthy controls were recruited from the general community, areca-nut chewers were convenience-sampled from patients visiting a hospital oral and maxillofacial surgery clinic. Although saliva donations occurred during outpatient visits unrelated to surgical operations, the clinical dataset did not record oral mucosal lesions, oral submucous fibrosis status, dental caries, or periodontal disease indices, leaving the baseline mucosal burden of the chewers unquantified. Furthermore, chewers exhibited significantly higher body weight and body mass index, along with greater self-reported alcohol intake, which closely aligned with the grouping axis. The authors emphasized that because the study utilized genus-level 16S sequencing without functional metagenomic or metatranscriptomic profiling, and lacked longitudinal sampling of former chewers, all reported alterations must be interpreted strictly as exploratory associations with areca-nut user status that require future validation in larger, lesion-matched, and smoking-stratified prospective cohorts.
Subject of Research: Salivary bacterial community composition and tobacco confounding in male areca nut users
Article Title: Salivary bacterial community composition differs between areca-nut users and non-users: an exploratory 16S rRNA gene sequencing study
Article References: Feng, S., Ren, M., Wang, F., Lu, X., Zhang, H., & Feng, L. (2026). Salivary bacterial community composition differs between areca-nut users and non-users: an exploratory 16S rRNA gene sequencing study. Frontiers in Cellular and Infection Microbiology, 16, Article 1945798. https://doi.org/10.3389/fcimb.2026.1945798
Image Credits: AI Generated
DOI: 10.3389/fcimb.2026.1945798
Keywords: areca nut, salivary microbiota, 16S rRNA sequencing, oral microbiome, tobacco confounding, Actinomyces, Porphyromonas, alpha diversity, beta diversity, PERMANOVA, ALDEx2, DADA2, bacterial dysbiosis, oral health, microbial ecology, Anhui Medical University, compositional data analysis
News Source: Morgan Morrow. (November 20, 2026). Salivary Bacterial Composition Differs in Areca Nut Users Alongside Tobacco Confounding. Scienmag.



