key: cord-0895455-xq1cscei authors: Carl A. Pierce, Sharlene Sy; Benjamin Galen, Doctor Y Goldstein; Erika Orner, Marla J. Keller; Kevan C. Herold, Betsy C. Herold title: Natural Mucosal Barriers and COVID-19 in Children date: 2021-02-13 journal: medRxiv : the preprint server for health sciences DOI: 10.1101/2021.02.12.21251310 sha: 5c16881e860d9c2bb8df0bb352073ff7b162404e doc_id: 895455 cord_uid: xq1cscei COVID-19 is more benign in children compared to adults for unknown reasons. This contrasts with viruses such as influenza where disease manifestations are often more severe in children1. We hypothesized that a more robust early innate immune response to SARS-CoV-2 may protect against severe disease and compared clinical outcomes, viral copies and cellular gene and protein expression in nasopharyngeal swabs from 12 children and 27 adults upon presentation to the Emergency Department. SARS-CoV-2 copies were similar, but compared to adults, children displayed higher expression of genes associated with interferon signaling, NLRP3 inflammasome, and other innate pathways. Higher levels of IFN-alpha2, IFN-gamma, IP-10, IL-8, and IL-1beta were detected in nasal fluid in children versus adults. Anti-SARS-CoV-2 IgA and IgG were detected in nasal fluid from both groups and correlated negatively with mucosal IL-18. These findings suggest that a more robust innate immune response in children compared to adults contributes to favorable clinical outcomes. Epidemiological studies have consistently shown that children infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have a milder clinical course with significantly less morbidity and mortality than adults. The Centers for Disease Control (CDC) estimates that approximately 1.2-3.3% of total hospitalizations and less than 0.21% of deaths from coronavirus disease 2019 (COVID-19) are in children (1) . This experience is in contrast to other respiratory viruses, such as influenza or respiratory syncytial virus, where disease manifestations in children are often more severe than adults (2) . Several hypotheses have been proposed to explain why children are protected from more severe outcomes with COVID-19 including differences in expression of angiotensin-converting enzyme 2 (ACE2), the receptor for viral entry, resulting in lower viral loads, presence of antibodies to common cold coronaviruses that might provide partial protection, and a more robust innate response early in the course of infection that mitigates against a vigorous adaptive response (3, 4) . However, recent studies have shown that ACE2 expression is not reduced in children and may actually be lower in adults (5) . Surveys of children infected with COVID-19 have found that the amount of SARS-CoV-2 RNA detected in nasopharyngeal (NP) swabs is at least as high in children compared to adults (6) . It is also unlikely that antibodies that are cross reactive to other viruses explain the clinical differences as we previously found that antibody levels to other common cold human coronaviruses (229E, NL63, HKU1) were similar in adults and children. In addition, although common cold coronavirus antibody levels may be boosted in response to SARS-CoV-2 infection, they do not provide protection (7, 8) . In a previous analysis of patients hospitalized with COVID-19, we found that serum levels of IL-17A and IFNg collected early in the course of disease were higher in children versus adults and correlated significantly and inversely with age (7) . The source of these cytokines did not appear to be peripheral blood mononuclear cells as adults had more robust T cell responses. These findings suggested that other cells, such as innate immune cells within the respiratory tract, might be the source of the IL-17A and IFN-g and that a more vigorous mucosal innate response could account for the milder clinical course in children. To test the hypothesis that age-related outcomes reflect differences in host responses at the site of viral exposure, we collected nasopharyngeal (NP) swabs from pediatric and adult patients who presented to the Emergency Department at Montefiore Medical Center in The Bronx, NY, analyzed the harvested cells for gene expression, and measured the levels of cytokines and antibodies in the mucosa. The demographics and clinical characteristics of the study population are described in Table S1 . There were no significant differences in number of days of symptoms prior to presentation or the levels of SARS-CoV-2 viral RNA (quantified by cycle threshold [Ct] values) in NP swabs comparing the pediatric (n=12) and adult (n=27) patients (Fig. 1A) . However, the outcomes differed. Adults were more likely to be admitted to the hospital (81% vs 42%, p=0.02), had higher C-reactive protein (10.48 ± 7.19 vs 6.03 ±10.48 mg/dl, p<0.0001) and D-dimer (2.38 ± 3.42 vs 0.81 ± 0.59 µg/ml, p<0.0001) at admission, and if hospitalized, had a longer length of stay (10.36 ± 10.85 vs 3.0 ± 1.73 days, p<0.0001) ( Table S1 ). None of the pediatric patients required oxygen whereas 7 adults did and 4 required mechanical ventilation (p=0.03). Four adults but none of the pediatric patients died. To evaluate whether the age-related differences in outcomes reflect differences in host responses at the site of viral exposure, the nasal mucosa, we performed RNA-Seq on cells isolated from NP swabs obtained at the time of presentation to the Emergency Department (n=6 pediatric and 15 adult patients). We identified 538 differentially expressed genes (p<0.05 after FDR correction) including 267 that were upregulated and 271 that were downregulated comparing pediatric and adult patients (Fig. 1B) . Notably, expression of ACE2 trended towards being higher in the pediatric samples (p=0.053, Fig 1A) . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (Fig. 1C) . To understand the gene expression patterns contributing to the segregation of the samples, we interrogated the 50 genes contributing most strongly to the principal components (Table S2 ). This gene set was enriched for interferon-stimulated and other innate immune response genes such as IL-1β, CCL3, CXCL10, and NLRP3 (Fig. 1D ). After dichotomization of adults by clinical outcome, PCA analysis showed that adults who never required supplemental oxygen were intermediate between adults who required respiratory support and children ( Figure S1 ). Conversely, fatty acid metabolism (hallmark:M5935, NES=-1.86, p=0.006) was enriched in adults compared to pediatric patients (Fig. 2B ). Oxidative phosphorylation and glycolysis pathways were also increased in adults versus children, but these were not statistically different after FDR correction (NES=-1.18, p=0.14; NES=-1.09, p=0.3, respectively). These findings further support an enhanced innate response to SARS-CoV-2 infection in children but enhanced metabolic pathways among cells from adults. We then performed RT-qPCR using RNA isolated from NP swabs from an additional 4 pediatric and 5 adult patients who were not included in the RNA-Seq analysis to confirm these findings. IL-17A gene expression was significantly increased in children (Fig. 2C ) and there were similar trends for higher levels of IFN response genes ( Figure S2 ). We also measured cytokine All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint levels in NP fluid and found significantly higher levels of IFN-g, IFN-a2, and IL-1b, IL-8 and IP10 in pediatric patients ( Fig. 2D-H) , consistent with the RNAseq data. SARS-CoV-2-specific IgA in nasal mucosal secretions may contribute to protection (9) and our RNA-Seq data revealed a cluster of B cell-associated genes in the PCA gene set (Dim 2; IGHA1, IGKC, IGHM, IGHG1, FAM30A and PAX5) with higher expression in a subset of both adult and pediatric samples ( Fig 1D) . Therefore, we measured total IgA and IgG by ELISA and SARS-CoV-2 specific IgA and IgG targeting S1, S2, and the receptor binding domain (RBD) of the spike and nucleocapsid (NC) protein by multiplex assay in the NP fluid. We did not identify differences in total IgG or IgA comparing pediatric and adult patients ( Agglomerative hierarchical clustering using only the B cell-associated genes dichotomized the cohort into a "low" or a "high" expression group (Fig. 3E ). All of the adult patients with more severe clinical disease who required supplemental oxygen (SO) were in the low transcript group (Fig. 3E ). Correlation matrices were generated to explore possible associations between NP antibody levels, cytokines, and SARS-CoV-2 Ct values (Fig. 4A ). In general, antibodies and cytokines did not strongly correlate with one another. However, we observed a significant and inverse relationship between SARS-CoV-2-specific antibodies (IgG and IgA) and IL-18 (Fig. 4C , D). There were no differences in IL-18 levels comparing children and adults, but there was a trend towards lower IL-18 in adults who required supplemental oxygen (Fig. 4B , p=0.12). Our studies identify age-related differences in primary immune responses to SARS-CoV-2 at the nasal mucosa, the presumptive site of first viral encounter, which may contribute to the clinical outcomes. Most studies focus on systemic immune responses and have measured cytokines and antibodies in the blood, but our prior and other studies led us to speculate that the innate immune response was more vigorous in children (7) . Our new findings using NP swabs All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint obtained at the earliest time point of presentation establish this and suggest that a more robust mucosal innate response in children overcomes viral evasion strategies and generates an immediate barrier to viral infection. This may dampen the subsequent adaptive immune response as evidenced by our prior findings of lower neutralizing antibodies, decreased antibody-dependent cell mediated phagocytic activity (ADCP), and less robust T cell responses in children versus adults (7) . A reduced adaptive response in children compared to adults has been confirmed by others who documented decreased neutralizing antibody titers (10) and less robust T cell responses (11) . In contrast, a reduced or delayed mucosal response in adults, evidenced by the decreased expression of transcripts associated with innate pathways in their NP swabs may lead to an inability to escape viral immune evasion and a more vigorous adaptive response (12, 13) . The latter contributes to high systemic levels of other inflammatory cytokines (e. g. IL-6 and TNF) and an increased risk of acute respiratory distress syndrome, which has also been observed with SARS-CoV-1 (14-17). Other studies have found an impaired innate response to SARS-CoV-2 in adults compared to other respiratory viral infections (18). Children have more frequent respiratory infections than adults. This, as well as recent childhood immunizations, could contribute to a higher basal level of activation of innate responses as suggested by studies of rhinovirus in which an early interferon response was associated with rapid viral clearance (19). Notably, a recent study found that infection of organoid cultures with rhinovirus protected against subsequent SARS-CoV-2 challenge (20). The prior respiratory infections in children may prime innate immune cells that can rapidly respond to virus. The higher gene expression of metabolic flux pathways in adults may reflect the metabolic demands needed to activate innate pathways (21) and may account for a delayed response to COVID-19 (22). We detected similar levels of SARS-CoV-2 specific IgA and IgG in NP samples obtained from adults and children at this early time point before a fully mature antibody response would be expected. Notably these early antibody responses correlated inversely with mucosal IL-18 levels. All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint IL-18, an IL-1 superfamily cytokine predominantly produced by macrophages, is cleaved to its active form by the NLRP3 inflammasome and ultimately promotes the production of IFN-g. We speculate that the early release of this cytokine may temper the adaptive response which is consistent with the more severe outcomes with lower levels of IL-18. The kinetics of its secretion may be important since at later time points, elevated serum IL-18 levels are associated with increased inflammasome activation and disease severity (23). In summary, we show, for the first time, direct evidence for a more vigorous early innate immune response in the nasopharynx of children compared to adults at clinical presentation with COVID-19. Innate immune responses to other pathogens have also been shown to decrease with age (24, 25) The cellular source of these protective cytokines is not clear but could include mucosal and airway epithelial or invariant natural killer T cells (26, 27). Nonetheless, our findings suggest that airway resident cells establish the response to virus that ultimately determines the clinical outcomes. The reasons for the differences in the early innate responses with age are not clear, but therapies that enhance these pathways may be an effective form of treatment and protect from further damage from viral invasion as suggested by ongoing trials of inhaled interferon beta-1 for early treatment of COVID-19 (28). NP swabs were obtained from the Montefiore Clinical Laboratory from patients with confirmed SARS-CoV-2 infection by PCR assay who presented to the Emergency Department at Montefiore Medical Center between November 2020 and January 2021. Patients were excluded if they had pre-existing medical conditions that might impact immune responses including cancer and HIV, pregnancy, or were receiving chronic immunosuppressive therapy. Demographics, length of stay, peak respiratory support required (1=room air, 2=nasal cannula; 3=CPAP or high flow nasal oxygen, and 4=mechanical ventilation), outcome and the results of clinical laboratory All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint studies including SARS-CoV-2 PCR (Ct values), complete blood counts, D-dimer, and C-reactive protein were obtained by chart review. The NP swabs were collected in viral transport medium. A portion was removed for measurement of SARS-CoV-2 RNA by PCR in the Clinical Microbiology Lab and the remainder of the sample was transported. NP swabs were also obtained from 7 healthy controls. The swab was transferred to a tube containing media supplemented with 13 µM dithiotrietol to inactivate virus and then incubated in a thermomixer for 10 minutes at 37°C and 500 RPM. The cells were washed and frozen in 500µl 90% FBS/10% DMSO. The original transport medium (nasal fluid) was aliquoted and stored at -80°C for measurement of cytokines and antibodies. Samples with sufficient RNA quantity and quality were used for analysis. The methods for library preparation, sequencing, and analysis are discussed in Supplementary Methods. For quantitative RT-PCR, RNA was isolated as for RNAseq and cDNA synthesis was performed with methods described in Supplementary Methods. Media from NP swabs was thawed and treated with UV light to inactivate virus prior to use. SARS-CoV-2-specific IgG and IgA were measured using MILLIPLEX SARS-CoV-2 Antigen Panel 1 IgG and IgA kits (Millipore HC19SERG1-85K and HC19SERA1-85K, respectively). Cytokines were measured using the MILLIPLEX Human Cytokine/Chemokine/Growth Factor Panel A kit (Millipore, HCYTA-60K). Because of sample volume limitations not all assays could be performed with all samples. Missing data was at random and the number of samples used are indicated. Statistical analyses were performed in GraphPad Prism (v9.0.1, GraphPad Software, Inc). Cytokine and antibody data were All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint log transformed prior to analysis. Normality was tested and a parametric or non-parametric test was used for comparison of groups as indicated. College of Medicine (IRB# 2020-11278). Written informed consent was obtained for samples from healthy controls. (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. (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 preprint this version posted February 13, 2021. All rights reserved. No reuse allowed without permission. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Bars show mean ± 95% CI. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (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. (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. (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. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (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 preprint this version posted February 13, 2021. ; https://doi.org/10.1101/2021.02.12.21251310 doi: medRxiv preprint Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019 -COVID-NET, 14 States Pediatric influenza and illness severity: what is known and what questions remain? Inflammatory syndromes associated with SARS-CoV-2 infection: dysregulation of the immune response across the age spectrum Nasal Gene Expression of Angiotensin-Converting Enzyme 2 in Children and Adults Distinct disease severity between children and older adults with COVID-19: Impacts of ACE2 expression, distribution, and lung progenitor cells Age-Related Differences in Nasopharyngeal Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Levels in Patients With Mild to Moderate Coronavirus Disease Immune responses to SARS-CoV-2 infection in hospitalized pediatric and adult patients All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder Enhanced SARS-CoV-2 neutralization by dimeric IgA Distinct antibody responses to SARS-CoV-2 in children and adults across the COVID-19 clinical spectrum Immune response in children with COVID-19 is characterized by lower levels of T-cell activation than infected adults SARS-CoV-2 desensitizes host cells to interferon through inhibition of the JAK-STAT pathway Activation and evasion of type I interferon responses by SARS-CoV-2 Interleukin 6, soluble interleukin 2 receptor alpha (CD25), monocyte colony-stimulating factor, and hepatocyte growth factor linked with systemic hyperinflammation, innate immunity hyperactivation, and organ damage in COVID-19 pneumonia Pathological findings of COVID-19 associated with acute respiratory distress syndrome. The Lancet Respiratory Medicine Heatmaps were generated using the pheatmap package. Principal components analysis was performed with the prcomp functionusing all genes with a nonzero total read count. Prior to PCA, data were transformed with the vst function in DESeq2. PCA results were visualized with the factoextra package. For gene set enrichment analyses, Hallmark (h) and GO (c5.go) datasets were downloaded from MSigDB (Broad Institute) and analysis performed in R with the fgsea package using 1000 permutations Primers/probes were from ThermoFisher: MX1 (Hs00895608_m1), MX2 (Hs01550811_m1), IFNA1 (Hs00256882_s1), IFI44 (Hs00951349_m1), IFIT1 (Hs03027069_s1), IL17A (Hs00174383_m1), IFNG (Hs00989291_m1), RPLPO (Catalog No. 4326314E) Total IgA was measured using the IgA Human ELISA Kit (Invitrogen, BMS2096), and total IgG measured using the IgG Human ELISA All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder ELISA data were acquired on a Spectra Max M5 using SoftMax Pro GxP software (both Molecular Devices) No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity