key: cord-1041002-flzykj89
authors: Xu, Rong; Liu, Pengcheng; Zhang, Tao; Wu, Qunfu; Zeng, Mei; Ma, Yingying; Jin, Xia; Xu, Jin; Zhang, Zhigang; Zhang, Chiyu
title: Progressive deterioration of the upper respiratory tract and the gut microbiomes in children during the early infection stages of COVID-19
date: 2021-05-29
journal: J Genet Genomics
DOI: 10.1016/j.jgg.2021.05.004
sha: 1afe14842f6b789c0968f76cb62caf032948347c
doc_id: 1041002
cord_uid: flzykj89

Children are less susceptible to COVID-19 and they have manifested lower morbidity and mortality after infection, for which a multitude of mechanisms may be considered. Whether the normal development of the gut-airway microbiome in children is affected by COVID-19 has not been evaluated. Here, we demonstrate that SARS-CoV-2 infection alters the upper respiratory tract and the gut microbiomes in nine children. The alteration of the microbiome is dominated by the genus Pseudomonas, and it sustains for up to 25–58 days in different individuals. Moreover, the patterns of alternation are different between the upper respiratory tract and the gut. Longitudinal investigation shows that the upper respiratory tract and the gut microbiomes are extremely variable among children during the course of COVID-19. The dysbiosis of microbiome persists in 7 of 8 children for at least 19–24 days after discharge from the hospital. Disturbed development of both the gut and the upper respiratory microbiomes and prolonged dysbiosis in these nine children imply possible long-term complications after clinical recovery from COVID-19, such as predisposition to the increased health risk in the post-COVID-19 era.

COVID-19 caused by SARS-CoV-2 has been diagnosed in more than 140 million people Because of the high variability in bacterial composition among individuals within each group, 94 we further characterized the patterns of bacterial community composition using the Dirichlet 95 multinomial mixtures (DMM) method, and identified 8 community types (Fig. 1A) . The specimens

The vast majority (97.8%: 44/45) of stool specimens of COVID-19 children fell into three 121 distinct community types, which were designated as COVID-GUT I-III (Figs. 1A and S4). 122 Similarly, the vast majority (91.5%: 54/59) of nasal and throat swabs formed another three distinct 123 types, which were designated as COVID-TN I-III (Figs. 1A and S4). Three upper respiratory 124 tract-related types and three GUT-related types of COVID-19 children were significantly separated 125 from each other (Fig. 1B) . The community types from the same organ, regardless of the upper 126 respiratory tract and the gut, were also separated from each other and their alpha-diversity orderly 127 decreased from type I to III (Fig. 1C) . The decreased alpha-diversity from types I to III may To characterize eight microbial community types, we identified 35 indicator genera ( Fig. 2A) .

The H-MIX type was characterized by 11 genera with predominant commensal bacteria of 134 Prevotella, Veillonella, Streptococcus, unclassified Pasteurellaceae, Actinomyces, Porphyromonas, 135 Finegoldia and Fusobacterium ( Fig. 2A) Moraxella is a common human respiratory tract pathogen, and it is often present in the individuals 2.5 The dynamic change of the microbiomes in three body sites of children during COVID-19 165 Recently, we observed synchronous restoration of the microbiomes of both the upper 166 respiratory tract and the gut towards more diverse community structure in COVID-19 adults within 167 a short time (6-17 days) after symptom onset (Xu et al., 2021) . Distinct from adults, the 168 microbiome community compositions were extremely variable in children during the course 169 COVID-19 disease, and the changes of the community types were not the same between the upper 170 respiratory tract and the gut (Fig. 3) . The upper respiratory (especially nasopharyngeal) microbiome 171 of 7/8 children (except CV05) appeared to evolve from early healthy (H-MIX) or high-diversity 172 community structure (COVID-TN-I) to late low-diversity dysbiosis structure (COVID-TN-III), 173 indicating a steady deterioration in composition and function of the upper respiratory microbiome 174 despite mild symptoms and clinical recovery (Fig. 3) . In particular, the dysbiosis in the upper 175 respiratory tract was observed to last at least 19-24 days after discharge (i.e., 42-58 days after 176 symptom onset) in three children (CV01, CV02 and CV09).

Compared to the upper respiratory tract microbiome, the gut microbiome alteration was more 178 diverse among these COVID-19 children. Improvement or restoration in the gut microbiome was 179 observed in three children (CV01, CV02 and CV05), but a deterioration occurred in another three 180 children (CV03, CV04 and CV09) (Fig. 3) . For example, the bacterial community type of CV09 181 improved from COVID-GUT-II to COVID-GUT-I on day 7 after symptom onset, but returned back 182 to COVID-GUT-III on day 37. For CV03, whose microbiome became worse from a gut community 183 type COVID-GUT-II on day 19 to a respiratory community type COVID-TN-III on day 27, and

The restoration or deterioration of the gut microbiome showed no association with clinical recovery 187 (discharge from the hospital) or the presence or absence of SARS-CoV-2 RNA in the gut (Fig. 3) . 188 The detailed dynamic changes of bacterial compositions are shown in Fig. S5 .

To further assess their dynamic change in relative abundance over time, the top indicator 2.6 Bacteria-bacteria co-occurrence network 205 We identified two main co-occurrence networks (Fig. S6 ). One is predominated by the To reveal the progress of the microbiome in these patients, we further performed the 218 co-occurrence network analysis using data from three disease stages: acute phase (the first three 219 days since symptom onset), middle phase (from the 4th day to the day of discharge) and 220 convalescent (after discharge) phase. We found that the bacteria-bacteria co-occurrence network 221 enlarged and became more complex from early phase to convalescent phase, accompanied by the 222 involvement of more commensals (Fig. 5) . In each phase, two core co-occurrence networks were 223 identified. The bacteria in the same core network were often linked by positive correlation, whereas 224 the bacteria in different networks were linked by negative correlation, especially in the convalescent 225 phase (Fig. 5) . Of particular importance is that a core co-occurrence network was found to be 226 mainly formed by Pseudomonas, Herbaspirillum and Comamonadaceae_U in early phase, and it 227 maintained to late convalescent phase (Fig. 5) . Pseudomonas is a pathogenic bacteria;

Prevotella and unclassified Clostridia, as well as non-hub bacteria Faecalibacterium and 231 Roseburia). The persistent existence of these bacteria implies a slow improvement of the 232 microbiome in these COVID-19 children. One limitation of our study is that only nine COVID-19 children were recruited in this study.

This is due to fact that the epidemic was quickly controlled in Shanghai and surrounding areas, and 320 very few children were infected and available for study. Although the relatively small patient of 103 samples, including 31 nasal swabs, 28 throat swabs and 44 feces, were collected from these 342 patients (Fig. S1 ). Twenty-five samples from 14 age-matched healthy children were used as controls 343 (Table S1 ). The upper respiratory samples were collected using flexible, sterile, dry swabs, which analysis. The Pearson's "r" higher than 0.5 or lower than -0.5 with P value that was below 0.001 443 after the FDR adjustment was considered significant correlation. Co-occurrence network of 672 We selected those microbial genera appearing in at least 30% of samples with at least 0.3% average 673 abundance as core microbiomes to perform Pearson correlation analysis among microbial taxa. The COVID-GUT-I  COVID-GUT-II   COVID-GUT-III  COVID-TN-I  COVID-TN-II  COVID-TN-III   H-GUT  H-MIX   DMM  Cluster   COVID-GUT-I  COVID-GUT-II   COVID-GUT-III  COVID-TN-I  COVID-TN-II  COVID-TN-III   H- 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58 Days after symptom onset 

Deblur rapidly resolves single-nucleotide community 481 sequence patterns

Regulation of bacterial virulence by two-component systems

Gut bacteria that prevent growth impairments 486 transmitted by microbiota from malnourished children

A case 488 series of children with 2019 novel coronavirus infection: Clinical and epidemiological features

Qiime allows analysis of high-throughput community 492 sequencing data

Temporal development of the gut microbiome in early 606 childhood from the teddy study

The infant nasopharyngeal microbiome impacts severity of lower respiratory infection 609 and risk of asthma development

Naive bayesian classifier for rapid assignment of rrna 611 sequences into the new bacterial taxonomy

Characteristics of and important lessons from the coronavirus disease 2019 613 (covid-19) outbreak in china smmary of a report of 72314 cases from the chinese center for disease 614 control and prevention

Temporal association between human upper respiratory and gut bacterial microbiomes during the 617 course of covid-19 in adults

Alterations in gut microbiota of patients with covid-19 during time of hospitalization

Zhigang Zhang: Conceptualization, Project administration, Analyzation, Software, Data curation, Visualization, Writing-original draft, Writing-review & editing, Funding acquisition

Rong Xu: Investigation, Methodology, Data curation, Visualization, Writing-original draft. Pengcheng Liu: Investigation, Collection of clinical samples and data, Analyzation. Tao Zhang: Investigation, Software, Validation, Visualization, Writing-original draft. Qunfu Wu: Investigation, Software. Mei Zeng: Analyzation, Resources. Yingying Ma: Methodology. Xia Jin: Analyzation, Writing-review & editing