The immunological profile of SARS-CoV-2 infection in children is linked to clinical severity and age

doi:10.21203/rs.3.rs-2022874/v1

Download PDF

Article

The immunological profile of SARS-CoV-2 infection in children is linked to clinical severity and age

https://doi.org/10.21203/rs.3.rs-2022874/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Coronavirus disease 19 (COVID-19) is clinically less severe in children, even if the large variety and degree of severity of symptoms reported in children pose a still unresolved challenge to clinicians. We performed an in-depth analysis of immunological profiles in 18 hospitalized SARS-CoV-2-infected children; results were compared to those obtained in 13 age- and sex-matched healthy controls (HC). Patients were categorized as paucisymptomatic/moderate (55.6%) or severe/critical (44.5%), according to established diagnostic criteria, and further stratified into infants (1–12 months, 39%), children (1–12 years, 44%), and adolescents (> 12 years, 17%). We assessed SARS-CoV-2-specific RBD antibodies (Ab), neutralizing antibodies (nAb) and circulating cytokines/chemokines in plasma; SARS-CoV-2-specific immune response was measured in peripheral blood mononuclear cells by gene expression and secretome analyses. Our results disclose peculiar circulating cytokine/chemokine profiles in patients sharing a similar clinical phenotype. A cluster of patients consisted of infants with severe symptoms who presented a hyperinflammatory profile, and extremely polarized antibody profiles, ranging from patients who were negative for Abs and nAbs to those who displayed very high levels of both. In a second cluster consisting of paucisymptomatic patients, a less pronounced increase of inflammatory cytokines along with an association between selected cytokines and humoral responses emerged. A third cluster still consisting of paucisymptomatic patients showed a circulating cytokine/chemokine profile which substantially overlapped with that of HC. SARS-CoV-2-stimulated production of pro-inflammatory proteins (IL-1β, IL-2, IL-4, IL-6, IL-7, IL-8, IL-17, MIP-1β, and TNF-α), as well as T lymphocytes activation and migration-specific proteins were significantly increased in SARS-CoV-2 infected children compared to HC. Our findings suggest that immune response activation is directly correlated to clinical severity and to a lesser extent to age.

Children and adolescents account for 1–3% of reported coronavirus disease 2019 (COVID-19) cases globally ¹. It is increasingly accepted that COVID-19 shows a different pattern of clinical presentation in the paediatric population, with a milder disease course and overall better outcomes in comparison with adults. Nevertheless, a rare but severe evolution of SARS-CoV-2 infection in children is a systemic multi-inflammatory syndrome (MIS-C), featuring an intense and dysregulated inflammatory response ². MIS-C remains to be fully characterized from an immune-pathogenetic point of view ³ and our understanding of the correlation between age and the immunological pathogenesis of COVID-19 is hindered by the paucity of studies comparing the immune response between adults and children. Several alternative or complementary hypotheses can be advanced to explain the differences in susceptibility and clinical spectrum severity between adults and children as well as among different paediatric groups. These include social contact patterns, behavior, viral load, and age-related changes in immune and endothelial/clotting function ⁴. Moreover, the high prevalence of frequent infections in the first years of life might have a protective role by enhancing the activation state of the innate immune system while interfering with the replication of the SARS-CoV-2 ⁵.

Another pivotal issue concerns the role of humoral response in bounding the disease clinical course. In adults, it has been shown that patients with neutralizing antibodies (nAbs) to SARS-CoV-2 early after infection had a faster control of the virus ⁶ while the lack of neutralizing capacity correlates with an increased risk of a fatal outcome.

Conversely, the role of nAbs on the clearance of established SARS-CoV-2 infection and their persistence in paediatric patients remains to be elucidated. Few findings reported that children with mild symptoms displayed higher production of anti-SARS-CoV-2 nAbs compared to adults ^7,8.

Dissecting the immunological profile of SARS-CoV-2-infected children with varying degree of clinical manifestations appears key to understanding more in depth the age-related mechanisms of susceptibility to SARS-CoV-2 and might lead to the identification of immunological biomarkers and potential molecular targets. In this study, we analyzed clinical aspects and SARS-CoV-2-specific immunological profiles in a cohort of paediatric patients based on clinical severity (mild/moderate and severe/critical) and age (infants, children, and adolescents). Within severe cases, we also analyzed the immunological state of patients with cardiovascular involvement compared to those with respiratory manifestations only.

Clinical characteristics of patients and controls

Patients were hospitalized for a mean of 11.6 days (range: 4–19 days) (4–19 days) and only one patient was admitted to the Intensive Care Unit (ICU). Respiratory symptoms (RS; cough, rhinorrhea, wheezing, dyspnea) were present in 12 (66%) patients, gastrointestinal symptoms (GS; vomit, diarrhea, abdominal pain, hypoalimentation) in 8 (44%), while 4 (22%) patients had both RS and GS.

We stratified patients according to disease severity in pauci-symptomatic or with moderate disease (pauci, n = 10) and severe/critical (severe, n = 8). Pauci patients had upper or lower respiratory tract infection without organ disfunction; severe patients required oxygen supplementation or underwent multiorgan failure. In seven children with severe disease a cardiovascular involvement, defined as depressed left ventricular ejection fraction (EF < 55%) on echocardiography, was present.

Antibody responses in patients and controls

Fourteen patients (78%) had SARS-CoV-2 binding Abs and nAbs (Fig. 1 and Supplementary Figure S1). Binding Abs measured IgG in 14 (78%), IgA in 13 (72%) and IgM in 10 (56%) cases. The 4 children without detectable SARS-CoV-2 spike antibodies included both pauci and severe patients and were tested within 8 days from hospitalization. None of the controls had detectable antibodies to SARS-CoV-2 spike and NP. As expected, mean levels of binding and neutralizing Abs to SARS-CoV-2 were significantly increased compared to controls (p adjusted range < 0.05 - <0.001). Few differences in Ab levels across patients stratified by disease severity and age reached statistical significance: RBD IgG were higher in severe vs pauci children (p adjusted < 0.05), RBD IgM were increased in pauci adolescent vs severe infant (p adjusted < 0.05) and nAbs were increased in severe children vs infants (p adjusted < 0.05) (Fig. 1A-B and Supplementary Figure S1). No difference in antibody responses to SARS-CoV-2 were detected between patients with or without a cardiovascular involvement (Fig. 1C and Supplementary Figure S1).

IgG to the seasonal betacoronaviruses OC43 and HKU1 spike S2 protein were present in all but 3 children (1 control and 2 pauci infant patients), and IgG to the HA Flu H1N1 were present in all study subjects. Among patients, seasonal betacoronaviruses IgGs were higher in children vs infants with either severe or pauci disease. IgG to the flu H1N1 HA were lower in infants compared to child (1–12 years) or adolescent patients.

Clustering of antibody responses showed the presence of two major patient subgroups: the first broad and heterogenous in terms of age and disease severity that showed strong responses against multiple SARS-CoV-2 and seasonal betacoronaviruses antigens; the second group included controls and 4 patients without SARS-CoV-2 antibody responses (Fig. 1D).

Cytokine and chemokine plasma levels in paediatric patients

Plasma concentration of cytokines and chemokines was measured in 14 patients and 11 controls. Severe patients (n = 6) showed a significant increase in a subset of plasma cytokine/chemokine levels (IL-2, IL-4, IL-6, IL-8, MIP-1β, and TNF-α; ANOVA p adjusted < 0.05 to < 0.001) in comparison with pauci COVID-19 (n = 8) and/or controls (Supplementary Figure S2). Several plasma cytokines/chemokines showed a trend towards increased plasma levels in infant patients, which reached statistical significance for MIP-1β, TNF-α, and IL-2 (ANOVA p adjusted < 0.05) (Supplementary Figure S2). The stratification of COVID-19 patients according to presence or absence of cardiovascular involvement (CI) showed a trend towards increased plasma levels of several plasmatic cytokines/chemokines which reached statistical significance for IL-1β, IL-2, IL-4, IL-6, IL-7, IL-13, and IL-17 (ANOVA p adjusted < 0.01 to < 0.001) (Supplementary Figure S2). The stratification of patients based on both age and disease severity showed that the highest plasma levels of cytokines and chemokines were almost invariably found in infant patients with severe disease (Supplementary Figure S3).

Clustering of patients according to cytokine/chemokine plasma levels showed the presence of three major subgroups (Fig. 2A). Group 1 included 3 infants and 1 child, all with severe disease and CI and showing strong SARS-CoV-2 binding and neutralizing Abs in 2 patients (2/4, 50%) (Fig. 2B), both sampled in the early acute phase of the disease. Group 1 was characterized by a hyper-inflammatory profile with the elevation of several cytokines/chemokines like IL-1β, IL-2, IL-4, IL-6, IL-7, IL-13, IL-17, IFN-γ, TNF-α (Fig. 2C). Group 2 included patients of heterogeneous age, mostly paucisymptomatic and, save for a single exception, without CI that showed a higher prevalence of nAbs (5/7, 71%) and a less generalized increase of plasmatic cytokines/chemokines (Fig. 2A-C). Group 3 included paucisymptomatic 1 child and 2 adolescent patients positive for nAbs but with a plasma cytokine/chemokine profile essentially overlapping with that of control samples (Fig. 2A-C).

Correlation of circulating antibody and cytokine/chemokine responses in patients

We analyzed the correlation of antibody and cytokine/chemokine responses in patients grouped according to the previous cytokine/chemokine clustering. In patients belonging to group 1, i.e. infant or child patients with severe disease and CI, there was a positive correlation between binding and neutralizing SARS-CoV-2 Abs, as well as between SARS-CoV-2 Abs and seasonal betacoronaviruses Abs (Supplementary Figure S4). In particular, the correlation was high and significant (p not adjusted) between nAbs, SARS-CoV-2 RBD Abs of the IgA class, and IgG against the HCoV-OC43 spike protein. Always in group 1, we observed a strong positive correlation within two subsets of cytokines/chemokines: the first subset included IL1β, IL-2, IL-5, IL-6, IL-8, GM-CSF, IFN-γ, MCP-1, while the second subset included IL-4, IL-17, IL-13. Overall, in group 1 there was a negative correlation of antibody and cytokine/chemokine responses, in particular between IL-2 and SARS-CoV-2 binding Abs, with the caveat that the 2 Ab-negative infants were sampled at the time of acute disease presentation i.e. likely too early for the development of SARS-CoV-2 specific antibodies.

Group 2 included 2 infants, 4 children and 1 adolescent patients, which were mostly pauci-symptomatic. In this group, we observed a strong correlation of SARS-CoV-2 IgA and IgG RBD Abs with nAbs. The correlation of seasonal betacoronaviruses and SARS-CoV-2 Abs was instead weak or negative and did not reach statistical significance (Supplementary Figure S5). Always in group 2, a correlation of several chemokine/cytokine responses was present, but this reached statistical significance only for a few subsets like between IL-1β, IL-8, MIP1-β, TNF-α, between IL-13 and G-CSF, and between IL-5 and IFN-γ. SARS-CoV-2 IgG Abs correlated negatively with IL-12 while seasonal betacoronavirus IgG correlated negatively with IL-8 and TNF-α.

The group 3 included 1 child and 2 adolescent pauci patients. Group 3 was characterized by extremely low plasma levels of most circulating cytokines/chemokines. Correlations between antibodies and cytokines/chemokines were weaker and rarely significant (Supplementary Figure S6). In particular, we observed a positive correlation between IgG and IgA RBD Abs, between seasonal betacoronavirus Abs, and between IL-7 and MIP-1β. Negative correlations were found between SARS-CoV-2 IgG Abs and IL-8, IL-5 and MCP-1.

SARS-CoV-2-specific cytokine/chemokine secretion

We compared patients (n = 18) and controls (n = 13) cytokine/chemokine secretion by PBMCs stimulated in the presence/absence of SARS-CoV-2 antigens. In the absence of virus-specific stimulation, secreted levels of IL-1β, IL-2, IL-4, IL-6, IL-8, IFN-γ, and TNF-α was statistically significant higher in SARS-CoV-2-infected patients compared to HCs (Repeated measures ANOVA p adjusted range < 0.05 to < 0.001) (Fig. 3). Stimulation with SARS-CoV-2 antigens led to a significant higher secretion of IL-1β, IL-2, IL-4, IL-6, IL-8, IL-13, GM-CSF, IFN-γ, TNF-α in patients compared to controls (Fig. 3).

Clustering of unstimulated cytokine/chemokine secretion levels before and after stimulation showed the presence of two major largely overlapping groups of patients, characterized by heterogenous clinical and demographic features and differing mainly in the levels of secreted cytokines/chemokines (Supplementary Figure S7).

SARS-CoV-2-specific immune gene expression

We analyzed the mRNA expression of 70 genes involved in the antiviral immune response in unstimulated and SARS-CoV-2 stimulated PBMCs from all enrolled subjects. Results showed that 38 genes were modulated upon stimulation. Unstimulated gene expression clustered the subjects clearly, discriminating patients from controls (Fig. 4A).

Patients’ stratification according to disease severity showed a trend towards increased mRNA for several genes that reached statistical significance for IL-1A, IL1B, IL1RN, and PTGS2 in severe patients compared with controls (ANOVA Tukey post hoc test p adjusted < 0.05 to < 0.001) (Supplementary Figure S8). ITAG4 was instead the only gene whose mRNA was significantly decreased in patients compared to controls (pauci vs controls, ANOVA Tukey post hoc test p adjusted < 0.01). After stratification into age groups, genes that showed a statistically significant differences in mRNA levels between selected patient age groups and controls were: IL1B, IL6R, IL1A, IFI16, CCL3, IL8 (ANOVA Tukey post hoc test p adjusted < 0.05 to < 0.01) (Supplementary Figure S8). ITAG4 expression was decreased in infant patients (ANOVA Tukey post hoc test p adjusted < 0.05). Patients with or without cardiovascular involvement did not segregate into distinct clusters (Fig. 4A). However, stratification of patients based on cardiovascular involvement showed a few statistically significant differences compared to controls in the mRNA levels of the following genes: IL1B and IL1A (COVID-19 with cardiovascular involvement vs controls, ANOVA Tukey post hoc test p adjusted < 0.05) and IL1B, NR1H3, IL1RN, ERAP1 (COVID-19 without cardiovascular involvement vs controls, ANOVA Tukey post hoc test p adjusted < 0.05) (Supplementary Figure S8). ITAG4 expression was decreased in patients without cardiovascular involvement vs controls (ANOVA Tukey post hoc test p adjusted < 0.01).

SARS-CoV-2 antigens stimulation modulated gene expression in both patients and controls. Clustering of gene expression levels led to three patients’ groups heterogeneous for disease severity, age and presence of cardiovascular involvement (Fig. 4B). Stratification according to disease severity showed few statistically significant differences in SARS-CoV-2-stimulated gene expression in severe COVID-19 (ERAP1 and IFI16, ANOVA Tukey post hoc test p adjusted < 0.05) and pauci COVID-19 (ERAP1, CD44, ANOVA Tukey post hoc test p adjusted < 0.05) compared to controls (Supplementary Figure S9). ITAG4 gene expression was significantly decreased also in stimulated PBMC from severe patients compared to controls (ANOVA Tukey post hoc test p adjusted < 0.05). Stratification according to age showed a statistically significant difference in stimulated mRNA levels only for ERAP1 (adolescent COVID-19 vs controls, ANOVA Tukey post hoc test p adjusted < 0.05) (Supplementary Figure S9). Patients’ stratification according to cardiovascular involvement showed a significant difference in stimulated mRNA levels for ERAP1 (COVID-19 with cardiovascular involvement vs controls, ANOVA Tukey post hoc test p adjusted < 0.05) and ERAP1, CD44 (COVID-19 without cardiovascular involvement vs controls, ANOVA Tukey post hoc test p adjusted < 0.05) (Supplementary Figure S9). ITGA4 gene expression was lower in cases without cardiovascular involvement compared to controls (ANOVA Tukey post hoc test p adjusted < 0.05).

It is now well recognized that, in contrast to adults, children infected by SARS-CoV-2 are usually spared from severe respiratory disease. However, in the fraction of children who develop symptomatic COVID-19 the immunopathogenesis is still only partially understood. Our work provides a deep immunological characterization of antibody responses and chemokine/cytokine profiles in a cohort of hospitalized COVID-19 paediatric patients including infants and adolescents with a spectrum of clinical severity. Of note, in our study we considered severe cases eventually presenting cardiovascular involvement without multiorgan failure. Thus, none of the included patients fitted with WHO or CDC criteria for MIS-C diagnosis^9,10.

In paediatric COVID-19, the impact of patients’ SARS-CoV-2 antibodies in clearing the infection and determining specific clinical outcomes is not fully understood ¹¹.

While in adults there is evidence that the early SARS-CoV-2 nAb response is inversely correlated with COVID-19 disease severity ⁶, in children an efficient viral clearance may not depend on a strong antibody response ^12,13. A limited number of previously published studies reported the presence of both binding and neutralizing antibody responses in children and adolescents with mild or asymptomatic SARS-CoV-2 infection ^7,14,15. Despite the limited size of our paediatric cohort, our results show that even within patients of similar age and disease severity the SARS-CoV-2 antibody responses can exhibit extremely different phenotypes. For instance, among infants with severe disease some cases presented without detectable binding Abs and nAbs while others showed very high levels of both. Intriguingly, in almost all our COVID-19 patients we observed high levels of IgG to S1/S2 proteins of two seasonal hCoVs (HKU1 and OC43). This might be in agreement with either maternal transmission of antibodies, particularly in infants, or prior exposure to seasonal betacoronaviruses. In contrast with Aydillo T et al. ¹⁶ and Shrock E. et al. ¹⁷, and in agreement with Sasson J. M. et al. ¹⁸, we found a positive correlation between SARS-CoV-2 and seasonal human betacoronaviruses antibodies, suggesting the likely cross-activation by SARS-CoV-2 of pre-existing humoral immune response against seasonal hCoVs as previously described ¹⁹.

It has been suggested that paediatric COVID-19 patients might be characterized by a different profile of circulating inflammatory cytokines and chemokines compared to adults. Critical and fatal COVID-19 outcomes in adults were associated with elevated levels of mostly IL-6, IL-1β, and TNFα, along with a reduced and delayed production of type I IFNs ²⁰⁻²³. Upon stratification of our cohort on the basis of disease severity, elevated concentrations of pro-inflammatory cytokines/chemokines in plasma were found in severe cases (i.e. IL-2, IL-4, IL-6, IL-8, MIP1β, and TNFα). This was more evident specifically in infants reporting severe disease, showing a picture suggestive of a hyper-inflammatory profile presumably correlated to a cytokine storm.

We further investigated the correlation of cytokines with COVID-19 phenotype by clustering the subjects in our cohort based on age classes, disease severity and plasma cytokines. Subgroups of patients sharing a similar clinical phenotype showed peculiar circulating cytokine/chemokine profiles. This was particularly evident for a cluster of patients consisting of infants with a cardiovascular involvement, which presented a hyperinflammatory profile with significantly elevated plasma IL-2, IL-6, MIP1β and TNF-α levels. While recent findings reported normal IL-6 concentrations in paediatric COVID-19 with mild clinical presentation ²⁴, our severe infant patients presented with elevated circulating IL-6 levels similar to adult COVID-19, suggesting that elevated IL-6 levels might constitute a signature of severe COVID-19 also in children.

In this cluster of patients, SARS-CoV-2 antibodies showed extremely polarized profiles with some patients being completely negative for binding and neutralizing antibodies while others showed very high levels of both. Technical issues are an improbable cause for these contrasting observations, since binding antibodies were determined using validated assays based on alternative SARS-CoV-2 antigens (whole Spike, spike RBD and nucleocapsid proteins). Although the measurement of different Ig classes did not show in these polarized cases the presence of early IgM vs later IgG responses, the variable timing of sampling of the children during hospitalization, i.e. in an acute vs sub-acute disease phase, might have influenced these results. Of note, all infants in whom Abs were not detected were sampled in the acute phase of the disease within 7 days form hospitalization, i.e. likely too early for the development of SARS-CoV-2-specific antibodies. Interestingly, always in this cluster we observed an inverse correlation between the levels of inflammatory cytokines/chemokines and those of antibodies to multiple SARS-CoV-2 antigens. It might therefore be speculated that either a rapid progression of disease can occur before antibody development or that early high cytokine levels might actually impair the rapid establishment of a strong humoral response, particularly in the context of the immature immune system of infants.

Circulating cytokines/chemokines and antibodies highlighted the presence of another distinctive patient cluster characterized by paucisymptomatic COVID-19 without cardiovascular involvement in which the increase of inflammatory cytokines was less pronounced and more heterogeneous and an association was present between selected cytokine and humoral responses, in particular between SARS-CoV-2 IgG Abs and IL-12 and between seasonal betaconoraviruses IgG and IL-8 and TNFα. Always in this group, we observed a strong correlations of IL-5 with IFN-γ and of IL-1β with TNFα, supporting the hypothesis of synergistic Th1 and Th2 responses and a cooperative activity of both type 1 and type 2 macrophages in controlling COVID-19. In addition, the observed inverse correlation between IL-6 and IL-17 in these patients might be understood as an attempt to dampen and avoid excessive inflammatory responses.

Finally, we detected a third cluster of paucisymptomatic patients with a circulating cytokine/chemokine profiles substantially overlapped with that of healthy controls, a finding in agreement with previous reports in children with mild COVID-19, and likely reflecting a lower level of inflammation ²⁵. The most prominent feature distinguishing the two clusters of mild/paucisymptomatic patients was the presence or absence of a good correlation of cytokine and SARS-CoV-2 Ab levels. It could be speculated that a coordinated raise of both cytokines and Ab responses might mitigate disease severity.

Elevated concentrations of cytokines/chemokines were secreted by SARS-CoV-2 paediatric patient PBMCs vs control cells regardless of disease severity and age and that this was true both without or with SARS-CoV-2-specific stimulation. Similarly, the profiling of immune gene transcripts in PBMC revealed a generalized upregulation of multiple inflammatory factors in patients that was particularly evident in severe cases. Interestingly, we instead detected a decreased level of ITGA4 (CD49d) mRNA in severe patients including those with cardiovascular involvement. This is consistent with the reported decrease of CD49d in adults with severe COVID-19 ²⁶ suggesting that CD49d might be biomarker of worse disease progression also in paediatric patients.

Our study has some limitations. In particular, the limited number of enrolled patients and the heterogeneity of the timing of sampling with respect to the time of symptoms onset. Nevertheless, we believe that it provides for the first time, intriguing results with regards to the correlation of cytokine/chemokine and Ab responses in paediatric patients with different clinical manifestations of COVID-19.

Our data support the existence of specific immunological profiles correlated with disease severity and age in paediatric patients that can be easly analysed on peripheral blood samples. Considering that a non-negligible fraction of the pediatric population remains susceptible to the development of serious health consequences upon SARS-CoV-2-infection, our results deserve replication in larger cohorts and contribute to the understanding the immunological factors associated with severe COVID-19 in children, a goal that remains fundamental in order to inform prevention and mitigation strategies. However, we can speculate that a given immune profile can detect a particular underlying clinical condition and may be useful for better managing the evolution of COVD-19 in pediatric patients.

Patient population

We conducted a multicenter prospective analysis of clinical records and biological samples of SARS-CoV-2 infected children hospitalized from February 21st to May 1st, 2020 at the Luigi Sacco Hospital and Vittore Buzzi Hospital, Milan, Italy. The study was approved by the ethical committee of the coordinating center in Milan (protocol number 2020/ST/061). The humoral immunity study was approved by the IRCCS Ospedale San Raffaele Ethics Review Board (protocol number 34/INT/2020). All research was performed in accordance with relevant regulations, and informed consent was obtained from all participant parents or their legal guardians. The study has been performed in accordance with the Declaration of Helsinki. SARS-CoV-2 infection was based on either a positive result of SARS-CoV-2 real-time reverse-transcriptase–PCR (RT-PCR) assay on nasal/pharyngeal swab specimens or the presence of serum anti-SARS-CoV-2-specific antibody IgG using a semi-quantitative anti-SARS-CoV-2 ELISA (Euroimmun, Lübeck, Germany). Data regarding SARS-CoV-2 exposure history, clinical symptoms or signs, and laboratory findings on admission was collected using a standardized form. We enrolled 18 children (12 females and 6 males) with a mean age of 6.1 years (range 0.1–14.7 years) (Table 1). The patients were stratified according to age into 3 groups: 7 infants (1–12 months), 8 children (1–12 years), and 3 adolescents (> 12 years). As a control group we included 13 sex- and age-matched children (HC). None of the patients or the controls had a preexisting medical condition.

Table 1

Clinical characterization of paediatric patients and controls enrolled in the study.
	Patients		Controls
	#	%	#	%
Median Age (Range) in years	6,1 (0,1–14,7)	-	9,4 (1,0–16,6)	-
Female	12	33	5	38,5
Comorbidities	0	0	0	0
White cells count (average ± SD)	8071 ± 3506		-	-
Neutrophils (average ± SD)	2962 ± 2234	35 ± 15,4	-
Lymphocytes (average ± SD)	4215 ± 1976	53,2 ± 14,8	-	-
Positive RT-PCR on respiratory specimen	13/19	68,4	-	-
IgG Anti-SARS-CoV-2	11/13	84,6	-	-
Paucisymptomatic/moderate (PM)	10	55%	-	-
Severe/critical (SC)	8	45%	-	-

Specimens for immunological studies were stratified in two groups according to the time from the hospitalization (cut-off 7 days).

SARS-CoV-2, seasonal HCoVs, and H1N1 flu virus antibody assays

Serum samples of 18 children were tested for antibody responses to SARS-CoV-2, seasonal betacoronaviruses and 2009 pandemic H1N1 influenza virus. Detection of binding IgM, IgG and IgA antibodies to the spike S1/S2 and spike RBD antigens of SARS-CoV-2, IgG to the spike S2 antigens of HCoV-HKU1 and HCoV-OC43, and IgG to the hemagglutinin antigen (HA) of the 2009 A/H1N1 influenza virus was performed by LIPS, as previously described ²⁷. Briefly, each recombinant nanoluciferase-tagged antigen, expressed in transfected eukaryotic cells, was incubated in liquid phase with test sera for 2 hours, followed by capture of immune-complexes with rProtein A-sepharose, washes to remove unbound antigen, and measurement of recovered luciferase activity in a XS960 Centro luminometer (Berthold). Using Ab positive sera as standards, the raw data was then converted into arbitrary units (AU).

Nabs were measured using a neutralization assay based on a lentiviral vector (LV) expressing the SARS-CoV-2 spike protein and a luciferase reporter, as previously described ⁶. Briefly, serum serial dilutions were incubated with the virus for 30 min at 37°C and thereafter added to VeroE6 cells. After 48 hours, the luciferase expression was determined using a luciferase assay system (Bright-Glo, Promega) and measured in a Mitras luminometer (Berthold). Inhibitory serum dilution (ID) 50 was calculated with a linear interpolation method ²⁸.

Isolation and stimulation of Peripheral Blood Mononuclear Cells (PBMCs) with SARS-CoV-2 specific antigens

PBMCs were isolated from whole blood by density gradient centrifugation on Ficoll (Cedarlane Laboratories Limited, Hornby, Ontario, Canada), as previously described ²⁹ and viable cells were counted with the automated cell counter ADAM-MC (Digital Bio, NanoEnTek Inc., Korea). PBMCs were resuspended at the concentration of 1x10⁶/ml in RPMI 1640 medium (Euroclone, Milan, Italy) supplemented with 10% Fetal Bovine Serum (FBS), 1% of L-glutammin (LG), and 2% of penicillin/streptomycin. Subsequently, PBMCs were stimulated with 500 ng/ml nucleocapsid (N) and spike (S) specific SARS-CoV-2 antigens (Novatein Biosciences, MA USA). For gene expression and cytokine analyses, cells were harvested 10 hours after stimulation and cytokines content was quantified in the cell culture supernatants.

Quantigene Plex Gene expression assay

Gene expression analysis was performed by QuantiGene Plex assay (Thermo Scientific, Waltham, MA, USA), upon 10 hours’ stimulation of 8x10⁵ PBMCs with SARS-CoV-2 antigens. The QuantiGene Plex Assay allows the simultaneous quantification of the expression of 70 selected genes involved in the antiviral immune response and 4 housekeeping genes (Table 2).

Table 2

Genes involved in the antiviral immune response and housekeeping genes quantified by the QuantiGene Plex Assay.
IFNA2	CASP1	CD209	TAP1
IL2	ACTB	IL1B	MPO
IL28A	TLR4	IL10	NLRP3
IL17A	IL18	IL1RN	IL6R
CCL2	IL7	ABCA1	HAVCR2
TNFRSF4	PTGS2	ELOVL6	CXCL10
IFNB1	MX1	CD38	IFNG
PPARG	IFITM3	IL12B	NOD2
CRP	CD44	NOS2	CCL3
PPIB	CLEC4M	CH25H	CD69
COVID19	IFITM1	IFI16	AGTR1
CCL5	CD274	ACAT1	ERAP1
GAPDH	IL1A	HPRT1	TBP
PDCD1	IL12A	NOD1	ERAP2
HMGCS1	ITGA4	CSF3
TLR8	NR1H3	AGTR2
IL22	IL8	TLR3
CSF2	ITGB7	IL6
PYCARD	TLR7	ACE
TNF	ACE2	TMPRSS2

Multiplex cytokine analyses

A 17-cytokine multiplex assay was performed on plasma and cell culture supernatants from patients, 10 hours after PBMC stimulation with SARS-CoV-2 specific antigens, as described above, using a multiplexed Luminex magnetic bead immunoassay (Bio-Rad, CA, USA) according to the manufacturer’s instructions.

Statistical analysis

Median values with inter-quartiles ranges (IQR) were used to describe continuous variables while frequencies in percentages and counts were used for categorical variables. Variables were compared using the Chi-square or Fisher’s exact test for categorical variables, and the Wilcoxon rank sum for continuous variables. Comparisons of antibody, cytokine, and chemokine levels across groups were performed using ANOVA with Tukey’s HSD Post-Hoc Test. The correlation between antibody, cytokine, and chemokine levels was assessed using linear regression upon log transformation of the data. Two-tailed p values adjusted for multiple testing are reported for Tukey’s HSD Post-Hoc test analyses with p value < 0.05 considered statistically significant. Statistical analyses were performed using the R software version 3.4.0 (R Core Team (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/).

Antibody (Ab); Coronavirus disease-2019 (COVID-19); human coronavirus (hCoV); Immunoglobulin (Ig); Multisystem inflammatory syndrome in children (MIS-C); neutralizing antibody (nAb); Reverse transcription polymerase chain reaction (RT-PCR); Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)

AUTHORS CONTRIBUTION

D.T. and V.G. conceived and coordinated this project; C.V., M.S. and V.L. further developed the project with the help of I.M., C.F. (Claudio Fenizia), M.B., I.S. and F.L.; L.P. (Laura Paradiso), E.L., L.B., and L.P. (Lorenzo Piemonti) collected the data; C.V., C.F. (Claudio Fenizia), M.B. A.A. (Antonella Amendola), D.S., A.A. (Ahmed Abdelsalam), C.L. and C.F. (Clara Fappani) performed the experiments; C.V., A.A. (Antonella Amendola), E.T., V.L. and I.M. performed the formal analysis; C.V., M.S., V.L., D.T. G.S. and V.G. discussed the data and wrote the manuscript, with contributions from all authors; M.C. and G.V.Z. critically reviewed the manuscript. C.V., M.S. and V.L. equally contributed to the manuscript. V.G. and D.T. equally contributed to the manuscript.

DATA AVAILABILITY STATEMENT

The data sets generated and analysed during the current study are available from the corresponding author on reasonable request.

COMPETING INTERESTS STATEMENT

The authors declare no competing interests

FUNDINGS

This research and the APC were funded by Falk Renewables, grant number REC18ZUCC n. 27767. The work performed at IRCCS Ospedale San Raffaele (OSR) was funded by Program Project COVID-19 OSR-UniSR and Ministero della Salute (COVID-2020-12371617). We thank Foundation Dormeur, Vaduz for the donation of laboratory instruments relevant to this project to the Viral Evolution and Transmission Unit at OSR. DS salary was paid by EAVI2020, the European Union’s Horizon 2020 research and innovation program under grant agreement No. 681137.

INFORMED CONSENT STATEMENT

Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patients to publish this paper.

CORRISPONDING AUTHORS

Correspondence to Claudia Vanetti ([email protected]) and Marta Stracuzzi ([email protected])

Viner, R. M. et al. Susceptibility to SARS-CoV-2 Infection Among Children and Adolescents Compared With Adults: A Systematic Review and Meta-analysis. JAMA Pediatr. 175, 143 (2021).
Sharma, C. et al. Multisystem inflammatory syndrome in children and Kawasaki disease: a critical comparison. Nat. Rev. Rheumatol. 17, 731–748 (2021).
Patel, J. M. Multisystem Inflammatory Syndrome in Children (MIS-C). Curr. Allergy Asthma Rep. 22, 53–60 (2022).
Zimmermann, P. & Curtis, N. Why is COVID-19 less severe in children? A review of the proposed mechanisms underlying the age-related difference in severity of SARS-CoV-2 infections. Arch. Dis. Child. 106, 429 (2021).
Brodin, P. SARS-CoV-2 infections in children: Understanding diverse outcomes. Immunity 55, 201–209 (2022).
Dispinseri, S. et al. Neutralizing antibody responses to SARS-CoV-2 in symptomatic COVID-19 is persistent and critical for survival. Nat. Commun. 12, 2670 (2021).
Petrara, M. R. et al. Asymptomatic and Mild SARS-CoV-2 Infections Elicit Lower Immune Activation and Higher Specific Neutralizing Antibodies in Children Than in Adults. Front. Immunol. 12, 741796 (2021).
Karron, R. A. et al. Binding and neutralizing antibody responses to SARS-CoV-2 in very young children exceed those in adults. JCI Insight 7, e157963 (2022).
CDC. Multisystem Inflammatory Syndrome (MIS). Centers for Disease Control and Prevention https://www.cdc.gov/mis/index.html (2020).
Multisystem inflammatory syndrome in children and adolescents with COVID-19. https://www.who.int/publications-detail-redirect/multisystem-inflammatory-syndrome-in-children-and-adolescents-with-covid-19.
Garcia-Beltran, W. F. et al. COVID-19-neutralizing antibodies predict disease severity and survival. Cell 184, 476-488.e11 (2021).
Weisberg, S. P. et al. Distinct antibody responses to SARS-CoV-2 in children and adults across the COVID-19 clinical spectrum. Nat. Immunol. 22, 25–31 (2021).
Bahar, B. et al. Kinetics of Viral Clearance and Antibody Production Across Age Groups in Children with Severe Acute Respiratory Syndrome Coronavirus 2 Infection. J. Pediatr. 227, 31-37.e1 (2020).
Cotugno, N. et al. Virological and immunological features of SARS-CoV-2-infected children who develop neutralizing antibodies. Cell Rep. 34, 108852 (2021).
Garrido, C. et al. Asymptomatic or mild symptomatic SARS-CoV-2 infection elicits durable neutralizing antibody responses in children and adolescents. JCI Insight 6, e150909 (2021).
Aydillo, T. et al. Immunological imprinting of the antibody response in COVID-19 patients. Nat. Commun. 12, 3781 (2021).
Shrock, E. et al. Viral epitope profiling of COVID-19 patients reveals cross-reactivity and correlates of severity. Science 370, eabd4250 (2020).
Sasson, J. M. et al. Diverse Humoral Immune Responses in Younger and Older Adult COVID-19 Patients. mBio 12, e01229-21 (2021).
Dowell, A. C. et al. Children develop robust and sustained cross-reactive spike-specific immune responses to SARS-CoV-2 infection. Nat. Immunol. 23, 40–49 (2022).
Giamarellos-Bourboulis, E. J. et al. Complex Immune Dysregulation in COVID-19 Patients with Severe Respiratory Failure. Cell Host Microbe 27, 992-1000.e3 (2020).
Zhou, F. et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet 395, 1054–1062 (2020).
Acharya, D., Liu, G. & Gack, M. U. Dysregulation of type I interferon responses in COVID-19. Nat. Rev. Immunol. 20, 397–398 (2020).
Hadjadj, J. et al. Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients. Science 369, 718–724 (2020).
Soraya, G. V. & Ulhaq, Z. S. Interleukin-6 levels in children developing SARS-CoV-2 infection. Pediatr. Neonatol. 61, 253–254 (2020).
Jia, R. et al. Mild Cytokine Elevation, Moderate CD4+ T Cell Response and Abundant Antibody Production in Children with COVID-19. Virol. Sin. 35, 734–743 (2020).
Müller, T. M. et al. Circulating Adaptive Immune Cells Expressing the Gut Homing Marker α4β7 Integrin Are Decreased in COVID-19. Front. Immunol. 12, 639329 (2021).
Secchi, M. et al. COVID-19 survival associates with the immunoglobulin response to the SARS-CoV-2 spike receptor binding domain. J. Clin. Invest. 130, 6366–6378 (2020).
Fenyö, E. M. et al. International Network for Comparison of HIV Neutralization Assays: The NeutNet Report. PLoS ONE 4, e4505 (2009).
Saulle, I. et al. A New ERAP2/Iso3 Isoform Expression Is Triggered by Different Microbial Stimuli in Human Cells. Could It Play a Role in the Modulation of SARS-CoV-2 Infection? Cells 9, 1951 (2020).

No competing interests reported.

SupplementaryFigures.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

The immunological profile of SARS-CoV-2 infection in children is linked to clinical severity and age

Status:

Version 1

Abstract

Figures

Introduction

Results

Clinical characteristics of patients and controls

Antibody responses in patients and controls

Cytokine and chemokine plasma levels in paediatric patients

Correlation of circulating antibody and cytokine/chemokine responses in patients

SARS-CoV-2-specific cytokine/chemokine secretion

SARS-CoV-2-specific immune gene expression

Discussion

Materials And Methods

Patient population

SARS-CoV-2, seasonal HCoVs, and H1N1 flu virus antibody assays

Isolation and stimulation of Peripheral Blood Mononuclear Cells (PBMCs) with SARS-CoV-2 specific antigens

Quantigene Plex Gene expression assay

Multiplex cytokine analyses

Statistical analysis

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1