key: cord-0307418-b1vayzwl authors: Prasad, P.; Parida, A.; Mahapatra, S.; Mishra, R.; Murmu, K. C.; Aggarwal, S.; Sethi, M.; Mohapatra, P.; Ghosh, A.; Yadav, R.; Dodia, H.; Ansari, S. A.; De, S.; Singh, D.; Suryawanshi, A.; Dash, R.; Senapati, S.; Beuria, T. K.; Chattopadhyay, S.; Syed, G. H.; Swain, R.; Raghav, S. K. title: Nanopore 16S rRNA sequencing reveals alterations in nasopharyngeal microbiome and enrichment of Mycobacterium and Mycoplasma in patients with COVID 19 date: 2021-11-10 journal: nan DOI: 10.1101/2021.11.10.21266147 sha: 688f579d859c38f0c68d183ba1748602016c5ad8 doc_id: 307418 cord_uid: b1vayzwl The coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) is a major global health concern. This virus infects the upper respiratory tract and causes pneumonia-like symptoms. So far, few studies have shown that respiratory infections alter nasopharyngeal (NP) microbiome diversity and enrich opportunistic pathogens. In this study, we have sequenced the 16S rRNA variable regions, V1 through V9, extracted from NP samples of control and COVID-19 (symptomatic and asymptomatic) participants using the Oxford Nanopore technology. Comprehensive bioinformatics analysis investigating the alpha/beta diversities, non-metric multidimensional scaling, correlation studies, canonical correspondence analysis, linear discriminate analysis, and dysbiosis index analysis revealed control and COVID-19-specific NP microbiomes. We observed significant dysbiosis in COVID-19 NP microbiome with abundance of opportunistic pathogens such as Cutibacterium, Corynebacterium, Oerskovia, and Cellulomonas in asymptomatic patients, and of Streptomyces and Mycobacteriaceae family in symptomatic patients. Furthermore, we observed sharp rise in enrichment of opportunistic pathogens in symptomatic patients, with abundance of Mycobacteria and Mycoplasma, which strongly correlated with the occurrences of chest pain and fever. Our findings contribute novel insights regarding emergence of opportunistic pathogens in COVID-19 patients and their relationship with symptoms, suggesting their potential role in coinfections. The coronavirus disease 2019 (COVID-19) pandemic, a global health threat, is caused by 25 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The symptoms range from 26 fever, throat pain, loss of taste and smell to severe congestion in the chest, drop in oxygen 27 levels, pneumonia, and acute respiratory distress syndrome (1) . Furthermore, a significant 28 population worldwide remains asymptomatic, which is considered spreaders of the infection The nasopharyngeal tract is inhabited by a large number of microbial communities which 35 maintain normal homeostasis (6). Studies have revealed association between microbial 36 communities that influence viral infections of the lung, such as chronic rhinosinusitis, asthma, 37 pneumonia, and cystic fibrosis in the URT (7, 8). URT microbiome dysbiosis may also 38 enhance the opportunistic pathogen population and promote coinfection in the host (9, 10). 39 Reports have shown that nasopharyngeal (NP) swabs in viral transport media can be used to 40 investigate the NP microbial composition in patients with 12) . Recent studies then tested using the pairwise Wilcoxon rank-sum tests (alpha = 0.05). Finally, the effect size 139 of each differentially abundant feature was estimated using LDA. One-against-all sample 140 groups were compared and a linear discriminant analysis score greater than 3.6 was set as the 141 threshold; all-against-all sample groups were compared and a linear discriminant analysis 142 score greater than 2.0 was set as the threshold. Cladogram was used for identification of taxa 143 at different levels of the taxonomic hierarchy between sample groups (LDA score > 2). Study design and subject attributes 153 The role of the microbiome in viral infections is an emerging field. We collected NP samples 154 from COVID-19 patients between 11th May 2020 and 10th October 2020 to study alterations 155 in the NP microbiome. The schematic representation of the nasal microbiome study with 16S 156 rDNA amplicon sequencing is shown in Figure 1A . In total, 60 NP samples subjects (infected, 8 counts (1 from control and other from symptomatic category) were excluded and the final 162 study was performed with 58 subjects, including the control (C) [n = 12 (21%)], asymptomatic 163 [IA, infected asymptomatic; n = 25 (43%)], and symptomatic [IS, infected symptomatic; n = 164 21 (36%)]. The details of the participants considered for this study are described in Table 1 . 165 Differential OTUs (n = 795, p ≤ 0.05) were obtained from a total of 3482 OTUs using the 166 deseq2 function by comparing with control NP subjects. For downstream analysis differential, 167 795 OTUs were considered. We used the t-distributed stochastic neighbor embedding (t-SNE) 168 dimension reduction method to obtain the overall distribution of NP samples with 795 OTUs 169 ( Figure 1C ). We found that the control and SARS-CoV-2-infected subjects showed distinct 170 segregation of OTUs in the NP microbiome, while asymptomatic and symptomatic subjects 171 showed modest separation. This indicated that the abundance of 795 differential OTUs 172 potentially determines the compositional distribution patterns. patients compared to control subjects were found to be significantly reduced, no difference 181 was observed between symptomatic and asymptomatic samples ( Figure 2C, D) . Furthermore, 182 we used a linear regression model to establish the association between total OTU read counts 183 for each sample and Shannon/Simpson alpha diversity indices. We found negative correlation 184 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. 9 for both Shannon (IA -R = -0.35, R 2 = 0.44, p = 0.083; IS -R = -0.54, R 2 = 0.48, p = 0.012) 185 and Simpson (IA -R = -0.58, R 2 = 0.68, p = 0.0028; IS -R = -0.77, R 2 = 0.63, p = 7.7  10 -5 ) 186 alpha diversity indices with 95% confidence intervals with total OTU counts ( Figure 2E , F). To further understand the microbial composition dissimilarity within the samples, we analyzed 188 beta diversity using principal coordinate analysis (PCoA) and applied both unweighted 189 (microbial richness) and weighted (microbial richness and abundance) unifrac distance 200 Alterations in the microbial diversity prompted us to determine microbial dysbiosis index (DI) 201 (alterations in the microbial community) across the three groups (C, IA, and IS). We 202 performed PCoA using the Bray Curtis distance matrix and found that NP microbiota was 203 significantly altered (p = 0.001) with 61% variation in distances explained (R 2 = 0.6136) 204 assessed by ADONIS test. Next, we calculated the Euclidean distance from the centroid for preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; 10 observed DI was significant (Kruskal-Wallis test, p = 1.317E-07) across all the groups. Pairwise comparison showed significant dysbiosis between control vs symptomatic (p = 209 5.6E−09) and control vs. asymptomatic (p = 1.1E−09) groups; however, dysbiosis between 210 asymptomatic and symptomatic (p = 0.016) pair was not highly significant ( Figure 2H ). We 211 also observed highly significant dysbiosis (p = 2.2E-12) between the control and infected 212 group (Supplemental Figure 1A, 1C) . This showed that compared to that in the control 213 subjects, the NP microbial community is severely altered in both symptomatic and 214 asymptomatic COVID-19 patients. Distinct microbial composition and abundance at phylum and family levels in patients 216 suffering from SARS-CoV2 infection 217 The alpha and beta diversities, and DI showed that the NP microbiome was significantly 218 altered in COVID-19 patients. Next, we aimed to identify the microbial communities that were 219 altered at the phylum and family levels in three sample groups. We found 795 differential 220 OTUs, out of which, 12 phyla, 65 orders, 126 families, and 240 genera were present in all 221 three groups (C, IA, and IS) (Supplemental Table 1 ). The 12 phyla and their significance is 222 shown in Table 2 . The most significant bacteria in phylum level were Actinobacteria (p = 223 9.96E-07) and Proteobacteria (p = 9.61E-07), including 9 other phyla assessed using the 224 Kruskal-Wallis test. The abundance of phyla Firmicutes (p = 4.65E-02) and Actinobacteria (p 225 = 9.96E-07) were significantly higher in the SARS-CoV-2-infected groups (symptomatic and 226 asymptomatic). In contrast, Bacteroidetes (p = 1.48E-06) and Proteobacteria (p = 6.56E-07) 227 were highly abundant in the control group (non-infected) (Supplemental Figure 2A) . preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. These families contain opportunistic pathogens in both symptomatic and asymptomatic 231 COVID-19 patients, while these families are absent in control subjects. Top families and their 232 significance is shown in Table 3 . The mean and median value of each density plot revealed lack of difference between the C, 250 IA and IS groups at the phylum level. Furthermore, subtle differences were observed at the 251 order and family level. However, at the genus level, we found comprehensible differences 252 between C (mean = 7.95E-01; median = 6.39E-01), IA (mean = 5.65E-01; median = 8.33E-253 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; 12 01) and IS (mean = 6.51E-01; median = 7.01E-01) (Table 4) ( Figure 3B ). To evaluate the 254 statistical significance of densities based on sample segregation, we calculated cumulative 255 distribution distance (D) and significance between C, IA, and IS groups using the 256 Kolmogorov-Smirnov (KS) test for each taxonomic rank (Table 5) . We observed that 257 compared to that at other taxonomic levels, all the comparisons were highly significant at the Table 2 ). The heat map preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 287 The CCA analysis prompted us to select clusters with maximum variance explained. Therefore, we considered all the clusters with ≥ 30% variance, which includes all the clusters preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. Table 3 and Table 6 . We obtained 12 significantly enriched genera of Cellulosimicrobium in asymptomatic samples ( Figure 4C ). The histogram showing the relative 308 abundance of the 12 genera for each C, IA, and IS sample group clearly distinguishes each 309 sample type ( Figure 4D ). Finally, we used weighted correlation network analysis to construct 310 a network (Spearman correlation) with 12 genera identified using the LDA analysis. The 311 network creates two distinct modules, one for control groups and another for both symptomatic 312 and asymptomatic groups. We obtained strong correlation within the genera of C, IA, and IS 313 sample groups (Table 7) . However, the correlation between C vs. IA was extremely weak and 314 correlation was not obtained for C vs. IS groups. The network analysis suggested that the NP 315 microbiota of the control group was clearly distinct from that of the asymptomatic and 316 symptomatic groups. The DI of the 12 genera showed the highest significance between C vs. 317 IS (p = 4.7E-05), while significant dysbiosis was not observed between IA and IS groups 318 ( Figure 4F ). Overall, our analysis confirms the significance of the genera identified and their 319 associations with symptomatic and asymptomatic COVID-19 patients. Distinct correlation of OTUs with clinical symptoms in COVID-19 patients 321 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint To evaluate the accuracy of LDA classification that identified eight bacterial genera in the IA 322 and IS sample group, we tested the ROC (receiver operating characteristics) -AUC (area 323 under the curve) score. We obtained a value of 0.8 with 95% confidence interval for true 324 positive classification, showing 80% sensitivity and specificity of data obtained from LDA 325 analysis ( Figure 5A ). Next, we used the Spearman correlation matrix to identify the preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint 18 may lead to co-infections (18). Our study is the first to use 16S amplification of ~1.6 Kb 390 variable regions to identify the bacterial community associated with infected and control NP. However, 16S rRNA gene amplification may introduce PCR biases, however, more subjects 392 and a robust analysis pipeline may dilute these biases. This study is the first comprehensive have been associated with pathogenicity in humans and were isolated from hospitalized 412 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. 444 The authors declare no competing commercial or financial interests in relation to this work. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Human Respiratory System and its Microbiome at a Glimpse. Biology (Basel). 2020;9(10). 523 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint 29 preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 10, 2021. ; https://doi.org/10.1101/2021.11.10.21266147 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 IA1 IA2 IA3 IA4 IA5 IA18 IA19 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA30 IA35 IA37 IA39 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IS6 IS7 IS9 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS50 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 Relative abundance All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 C1 C10 C11 C12 C2 C3 C4 C5 C6 C7 C8 C9 IA1 IA18 IA19 IA2 IA20 IA21 IA22 IA24 IA25 IA27 IA29 IA3 IA30 IA35 IA37 IA39 IA4 IA40 IA41 IA42 IA43 IA44 IA45 IA46 IA5 IS10 IS11 IS12 IS13 IS14 IS15 IS17 IS23 IS28 IS31 IS32 IS33 IS34 IS36 IS38 IS47 IS48 IS50 IS6 IS7 Main Clinical Features of COVID-19 and Potential 459 Prognostic and Therapeutic Value of the Microbiota in SARS-CoV-2 Infections Lung microbiome and coronavirus disease 2019 (COVID-19): 462 Possible link and implications Single-cell RNA-seq data analysis on the 464 receptor ACE2 expression reveals the potential risk of different human organs vulnerable to 465 2019-nCoV infection CoV-2 Reverse Genetics Reveals a Variable Infection Gradient in the Respiratory Tract 1.617.2 Delta variant replication and immune evasion Homeostatic Immunity and the Microbiota The nasal 474 microbiome in asthma The role of the local microbial 476 ecosystem in respiratory health and disease The microbiome of the upper