Dual-transcriptomic analysis of human nasal transcriptome and microbiome reveals host-bacteria interactions associated with symptomatic respiratory infection [RNA-Seq]

The human nasopharynx is colonized by a diverse community of commensal microbiota linked to many respiratory diseases, yet their interactions with the host remain unclear. In this study, we introduced a dual-transcriptomics analysis strategy, which can characterize the host transcriptome and microbiome from nasal samples simultaneously. RNA sequencing reads from human nasal swab samples were pre-processed and aligned to the human genome for host gene expression counting, while unmapped reads were further aligned to microbiota genome. After taxonomic classification, microbial abundance matrix was derived at each taxonomic level for differential and host-microbiota interaction analysis. We applied this workflow to a local SARS-CoV-2 cohort with 76 infected patients, among whom 55 (72.37%) were symptomatic at enrollment. Nasal swabs were collected from all 76 patients at enrollment and from 73 patients at one-week later follow-up. We detected a median of 4.81% reads unmapped from the human genome across all 149 samples, among which around half (median 48.63%) were successfully mapped to microbiome genome. Meta-transcriptomic analysis detected significantly higher SARS-related coronavirus loads in samples from the symptomatic group at enrollment, and both groups showed decreased loads one week later. Compared with benchmarking 16S rRNA sequencing on 53 samples, our computational strategy showed high correlation of relative abundance in all top 20 genus. A total of 685 bacteria species were identified to show a relative abundance >= 0.01% in at least 10% samples. Differential abundance analysis identified 66 species (DASs) from 6 phyla with significantly decreased abundance in samples from the symptomatic group compared to the asymptomatic group at enrollment. Integrating these symptom-associated DASs with host’s gene expression using an expression quantitative trait bacteria (eQTB) model, we found 58 symptom-associated DASs identified at enrollment were significantly associated with one to 16 genes. GSEA showed a series of symptom-associated DASs were significantly correlated with pathways related to the activation of olfactory, keratinocyte differentiation, and DNA methylation. In summary, our dual-transcriptomic analysis strategy effectively characterized host-microbiome interactions, offering insights into microbial contributions to respiratory diseases.