Metatranscriptome of human lung microbial communities in a cohort of mechanically ventilated COVID-19 Omicron patients

Abstract

The Omicron variant of the severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) infected a substantial proportion of Chinese population, and understanding the factors underlying the severity of the disease and fatality is valuable for future prevention and clinical treatment. We recruited 64 patients with invasive ventilation for COVID-19 and performed metatranscriptomic sequencing to profile host transcriptomic profiles, plus viral, bacterial, and fungal content, as well as virulence factors and examined their relationships to 28-day mortality were examined. In addition, the bronchoalveolar lavage fluid (BALF) samples from invasive ventilated hospital/community-acquired pneumonia patients (HAP/CAP) sampled in 2019 were included for comparison. Genomic analysis revealed that all Omicron strains belong to BA.5 and BF.7 sub-lineages, with no difference in 28-day mortality between them. Compared to HAP/CAP cohort, invasive ventilated COVID-19 patients have distinct host transcriptomic and microbial signatures in the lower respiratory tract; and in the COVID-19 non-survivors, we found significantly lower gene expressions in pathways related viral processes and positive regulation of protein localization to plasma membrane, higher abundance of opportunistic pathogens including bacterial Alloprevotella, Caulobacter, Escherichia-Shigella, Ralstonia and fungal Aspergillus sydowii and Penicillium rubens. Correlational analysis further revealed significant associations between host immune responses and microbial compositions, besides synergy within viral, bacterial, and fungal pathogens. Our study presents the relationships of lower respiratory tract microbiome and transcriptome in invasive ventilated COVID-19 patients, providing the basis for future clinical treatment and reduction of fatality.

Fig. 1

Analysis of SARS-CoV-2 genomes in the BALF samples. a Coverage distribution of reads mapped to reference SARS-CoV-2 genome, in which we found 8 complete genomes and 43 > 50% completeness. COVID-19 patients (n = 63). b Overview of SNP variant types in the assembled genome, in total we found 34 insertions, 18 deletions, and 583 substitutions. COVID-19 patients (n = 51). c Phylogenetic relationships of the 51 genomes assembled in this study vs. 682 SARS-CoV-2 genomes sampled in Beijing general population from the same period and deposited in GISAID. 20 genomes belong to BA.5 and 24 to BF.7 lineages. COVID-19 patients (n = 44). d The correlation between SARS-CoV-2 genomes of BF.7 and BA.5 lineages and 28-day mortality (survivors or non-survivors) based on Spearman’s rank test, no significant differences were found between the 28-day mortality rate in the two strains. COVID-19 patients (n = 32)

Fig. 2

Gene Differential Expression Analysis in BALF of COVID-19 Patients. a Volcano plot comparing gene expression between the survival and non-survival groups of COVID-19 patients. 489 genes were up-regulated and 187 genes were down-regulated in the survival group (abs(FC) > 1.5, p-value < 0.05), including red-highlighted genes that were functionally characterized or implicated in subsequent analyses. b Gene Ontology (GO) enrichment analysis (Biological Process) of the differentially expressed genes identified in (a), where the representative genes involved in each GO term are indicated in parentheses. c Volcano plot displaying genes with differential expression between COVID-19 patients and HAP/CAP patients, which are also differentially expressed in the survival group of COVID-19 patients. These genes are specific to COVID-19 survival patients and include 268 up-regulated and 97 down-regulated genes. d KEGG enrichment analysis of the differentially expressed genes identified in (c). HAP/CAP patients (n = 27), COVID-19 patients (n = 63), non-survivors of the invasive ventilated COVID-19 patients (n = 45), survivors of the invasive ventilated COVID-19 patients (n = 18)

Fig. 3

Alternation of microbiome and virulence factors in COVID-19 and HAP/CAP patients. a Composition and relative abundance of the top 10 viruses in invasive ventilated COVID-19 patients. b Relative abundance of the top ten bacterial phylum levels in invasive ventilated COVID-19 patients. c Relative abundance of fungi belonging to the genus Candida in COVID-19 patients. d Box plot of differential virus between invasive ventilated COVID-19 and HAP/CAP patients, significance was derived from Wilcoxon tests. e Box plot of differential bacteria between invasive ventilated COVID-19 and HAP/CAP patients, significance was derived from Wilcoxon tests. f Box plot of significantly different fungi between invasive ventilated COVID-19 and HAP/CAP patients. g Box plot of significantly different bacterium between survivors and non-survivors of the invasive ventilated COVID-19 patients. h Box plot of significantly different fungi between survivors and non-survivors of the invasive ventilated COVID-19 patients. i Relative abundance of the top ten virulence factors in invasive ventilated COVID-19 patients. j Heatmap of significantly different virulence factors between survivors and non-survivors of invasive ventilated COVID-19 patients. k Heatmap of differential virulence factors between invasive ventilated COVID-19 and HAP/CAP patients. HAP/CAP patients (n = 27), COVID-19 patients (n = 63), non-survivors of the invasive ventilated COVID-19 patients (n = 45), survivors of the invasive ventilated COVID-19 patients (n = 18). One-tailed wilcoxon rank-sum test was used for all significance statistics, *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, ****p-value < 0.0001

Fig. 4

Multi-omics correlation analysis of COVID-19 patients. a Heatmaps display the correlation analysis of viral, fungal, bacterial, and virulence factor, with differentially expressed genes in COVID-19 patients. Spearman correlation coefficients with adjusted p-values < 0.05 are marked in colored cells. b Heatmaps depicting correlation analysis between differentially expressed genes of viral and bacterial with cytokine levels in invasive ventilated COVID-19 patients. Positions with adjusted p-values < 0.05 are shown in color code, and the Spearman correlation coefficient numbers are within each grid. Only absolute correlation coefficients greater than 0.5 are displayed. COVID-19 patients (n = 63)