A cross-sectional study on the nasopharyngeal microbiota of individuals with SARS-CoV-2 infection across three COVID-19 waves in India

Abstract

Background: Multiple variants of the SARS-CoV-2 virus have plagued the world through successive waves of infection over the past three years. Independent research groups across geographies have shown that the microbiome composition in COVID-19 positive patients (CP) differs from that of COVID-19 negative individuals (CN). However, these observations were based on limited-sized sample-sets collected primarily from the early days of the pandemic. Here, we study the nasopharyngeal microbiota in COVID-19 patients, wherein the samples have been collected across the three COVID-19 waves witnessed in India, which were driven by different variants of concern.

Methods: The nasopharyngeal swabs were collected from 589 subjects providing samples for diagnostics purposes at the Centre for Cellular and Molecular Biology (CSIR-CCMB), Hyderabad, India and subjected to 16s rRNA gene amplicon - based sequencing.

Findings: We found variations in the microbiota of symptomatic vs. asymptomatic COVID-19 patients. CP showed a marked shift in the microbial diversity and composition compared to CN, in a wave-dependent manner. Rickettsiaceae was the only family that was noted to be consistently depleted in CP samples across the waves. The genera Staphylococcus, Anhydrobacter, Thermus, and Aerococcus were observed to be highly abundant in the symptomatic CP patients when compared to the asymptomatic group. In general, we observed a decrease in the burden of opportunistic pathogens in the host microbiota during the later waves of infection.

Interpretation: To our knowledge, this is the first analytical cross-sectional study of this scale, which was designed to understand the relation between the evolving nature of the virus and the changes in the human nasopharyngeal microbiota. Although no clear signatures were observed, this study shall pave the way for a better understanding of the disease pathophysiology and help gather preliminary evidence on whether interventions to the host microbiota can help in better protection or faster recovery.

Keywords: 16S rRNA gene amplicon-based sequencing; COVID-19 disease; Indian cohort; SARS-CoV-2; nasopharyngeal microbiome; variants of concern.

Figure 1

Distribution of the SARS-CoV-2 lineage among the samples collected from COVID-19 positive individuals. The nasopharyngeal samples were collected in the state of Telangana (India) between March 2020 and June 2022.

Figure 2

Alpha and beta diversity of the nasopharyngeal microbiota in CP and CN samples. Number of observed amplicon sequence variants (ASVs) (A) in all the analyzed samples (overall), and (B) in the three COVID-19 waves are shown. Shannon diversity in term of ASVs (C) in all the analyzed samples, and (D) in the three COVID-19 waves are depicted. (E) Beta diversity of microbiota assessed using principal component analysis (Bray-Curtis distance) represented along first two principal components for the three COVID-19 waves are presented.

Figure 3

Differential abundance of bacterial families between the CP and CN samples. The log2fold change in the mean abundance (along with whiskers representing standard errors) of a bacterial family in CP with respect to CN is depicted in (A) all the analyzed samples (overall) as well as in (B–D) each of the three COVID-19 waves. Significantly different abundance (q-value <0.05) is indicated with blue color.

Figure 4

Differential abundance of bacterial families between the four sub-group of samples – aCP, sCP, aCN, and sCN. The log2fold change in the mean abundance (along with whiskers representing standard errors) of a bacterial family in (A) aCP with respect to aCN, (B) sCN with respect to aCN, (C) sCP with respect to sCN, and (D) sCP with respect to aCP is depicted. Significantly different abundance (q-value <0.05) is indicated with blue color.

Figure 5

The changes in community structure (community shuffling) between the microbial association networks corresponding to the CN – ‘control’ and CP – ‘case’ networks. (A) Nodes belonging to the CN – ‘control’ and CP – ‘case’ networks are plotted along the left half and right half of the circular frame. Same node (microbe) in the two network is connected by an edge for easy viewing of the community shuffling. Node labels are colored (at random) based on sub-network/community affiliations. Grayed out node labels indicate that the node does not interact directly with the common sub-network. The node sizes are proportional to the betweenness centrality measure of the node in the corresponding network. (B) Heatmap representing the communities in the CN – ‘control’ network (along the vertical axis) and the CP – ‘case’ network (along the horizontal axis) with counts of the nodes (microbes) in each community in brackets. Numbers (and color gradient) in the heatmap indicates how the nodes constituting any sub-network/community of the CN – ‘control’ network are shared with the sub-networks/communities of the CP – ‘case’ network, and vice versa.