Alterations in Gut Microbiota Composition in Patients with COVID-19: A Pilot Study of Whole Hypervariable 16S rRNA Gene Sequencing

Abstract

It is crucial to consider the importance of the microbiome and the gut-lung axis in the context of SARS-CoV-2 infection. This pilot study examined the fecal microbial composition of patients with COVID-19 following a 3-month recovery. Using for the first time metagenomic analysis based on all hypervariable regions (V1-V9) of the 16S rRNA gene, we have identified 561 microbial species; however, 17 were specific only for the COVID-19 group (n = 8). The patients' cohorts revealed significantly greater alpha diversity of the gut microbiota compared to healthy controls (n = 14). This finding has been demonstrated by operational taxonomic units (OTUs) richness (p < 0.001) and Chao1 index (p < 0.01). The abundance of the phylum Verrucomicrobia was 30 times higher in COVID-19 patients compared to healthy subjects. Accordingly, this disproportion was also noted at other taxonomic levels: in the class Verrucomicrobiae, the family Verrucomicrobiaceae, and the genus Akkermansia. Elevated pathobionts such as Escherichia coli, Bilophila wadsworthia, and Parabacteroides distasonis were found in COVID-19 patients. Considering the gut microbiota's ability to disturb the immune response, our findings suggest the importance of the enteric microbiota in the course of SARS-CoV-2 infection. This pilot study shows that the composition of the microbial community may not be fully restored in individuals with SARS-CoV-2 following a 3-month recovery.

Keywords: COVID-19; dysbiosis; gastrointestinal tract; gut microbiome; gut microbiota; gut–lung axis; microbiota; recovered COVID-19 patient.

Figure 1

Alpha and beta diversity of the gut microbiota between COVID-19 patients and healthy controls. (A) Richness with operational taxonomic units (OTUs) number and within-sample diversity calculated with the Chao1, Shannon entropy, and Simpson index (median, quartile Q1–Q3, min–max). Asterisks denote statistically significant differences given by the Mann–Whitney test (** p < 0.01, *** p < 0.001) (B). Comparison between sample diversities by principal coordinates analysis (PCoA) with a Bray–Curtis dissimilarity matrix and with unweighted and weighted Unifrac distance. Red and blue represent COVID-19 patients and healthy control data, respectively.

Figure 2

Individual (A,B) and group (C) profiles of gut microbial composition at the phylum and genus level in the COVID-19 patient and healthy control groups. Charts for group profiles show mean relative abundance.

Figure 3

Bacteroidetes/Firmicutes ratio (A) and the main phylum-level abundance (B) in the gut microbiota among COVID-19 patients and healthy controls (median, quartile Q1–Q3, min–max, and mean values with SD, respectively). ** p < 0.01 calculated using the non-parametric Mann–Whitney test based on medians.

Figure 4

Projection of the microbial composition at phylum level on the two-dimensions biplot for COVID-19 patients (A) and healthy controls (B). PC1—principal component 1, PC2—principal component 2.

Figure 5

Differences in microbial composition between COVID-19 patients and healthy controls at class (A), order (B), family (C), and genus (D,E) level (mean values with SD). These data only concern initial statistically significant results calculated using the non-parametric Mann–Whitney test based on medians, and asterisks mark statistical significance obtained after multiple hypothesis testing corrections according to the Benjamini–Hochberg procedure * p < 0.05.

Figure 6

Species abundance distribution and difference in microbial composition between COVID-19 patients and healthy controls. (A) Venn diagram for identified in the study gut microbiota species. Red and blue represent COVID-19 patients and healthy controls data, respectively. The intersection shows the species (543) shared with both groups. (B) Heatmap of abundance differences in studied group at species level (mean values). These data only concern initial statistically significant results calculated using the non-parametric Mann–Whitney. (C) Observed multiple differences in the amount of bacteria abundance (↓—reduction, ↑—increase) between COVID-19 patients’ microbiota compared to healthy controls.

Figure 7

Flowchart of literature search. N—number of entries (research manuscripts). V1–V9—hypervariable regions of the 16S rRNA gene.