Immune perturbations induced by SARS-CoV2 in infants vary with disease severity and differ from adults’ responses

doi:10.21203/rs.3.rs-5176621/v1

Download PDF

Article

Immune perturbations induced by SARS-CoV2 in infants vary with disease severity and differ from adults’ responses

https://doi.org/10.21203/rs.3.rs-5176621/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Differences in immune profiles of children and adults with COVID-19 have been previously described. However, no systematic studies have been reported from infants hospitalized with severe disease. We applied a multidimensional approach to decipher the immune responses of SARS-CoV-2 infected infants (n=26; 10 subacute, 11 moderate and 5 severe; median age=~1.6 months) and matched controls (n=14; median age=~2 months). Single cell (scRNA-seq) profiling of PBMCs revealed substantial alterations in cell composition in SARS-CoV-2 infected infants; with most cell-types switching to an interferon-stimulated gene (ISG^hi) state including: (i) CD14⁺ monocytes co-expressing ISGs and inflammasome-related molecules, (ii) ISG^hi naïve CD4⁺ T cells, (iii) ISG^hi proliferating cytotoxic CD8⁺ T cells, and (iv) ISG^hi naïve and transitional B cells. Concurrently, we observed increased serum concentrations of both interferons and inflammatory cytokines in infected infants. Antibody responses to SARS-CoV-2 were also consistently detected in the absence of anti-IFN autoantibodies. Compared with infected adults, infants displayed a similar ISG signature in monocytes but a markedly enhanced ISG signature in T and B cells. These findings provide new insights into the distinct immune responses to SARS-CoV-2 in the first year of life and underscore the importance of further defining the unique features of early life immunity.

Biological sciences/Immunology/Infectious diseases/Viral infection

Health sciences/Diseases/Respiratory tract diseases

During the first year of the COVID-19 pandemic, SARS-CoV-2-related morbidity and mortality disproportionately affected adults. However, as the pandemic evolved and vaccines became available for adults, an increasing number of infected symptomatic infections were observed for children and infants, in part related to the emergence of new variants. This shift challenged the initial belief that children were mostly resistant to SARS-CoV-2 infections. Indeed, recent data from the CDC indicate that infants < 6 months of age with SARS-CoV-2 infection had high rates of hospitalization in this later phase of the pandemic¹.

Studies in adults with severe COVID-19 have shown increased plasma concentrations of pro-inflammatory cytokines, including IL-1β and the interferon (IFN)-induced chemokine IP-10², together with significant alterations in transcriptional profiles and leukocyte composition such as lymphopenia and increased neutrophil counts^3,4. In adult with mild disease, a marked early interferon-stimulated gene (ISG) signature was identified, though this signature was absent in more severe cases⁵. Conversely, other studies reported that interferons (i.e., Type I and Type III) and ISG signatures were induced only at late stages of severe infection, suggesting that IFNs might play a detrimental rather than protective role during COVID-19^6,7.

Comparing systemic immune responses to SARS-CoV-2 infection between children and adults revealed expansions of naive T and B lymphocytes and contraction of NK cells in children, whereas adults exhibited significantly increased cytotoxic and ISG^high T cell subsets⁸. In addition, school-age children with COVID-19 were found to have enhanced mucosal expression of innate immune pathways, including the NLRP3 inflammasome, and increased production of cytokines such as IFN-α2, IFN-γ, IP-10, IL-8, and IL-1β⁹, compared to adults. Another study highlighted enhanced local IFN responses in the respiratory tract of children, in contrast to a more significant systemic IFN response in adults. Higher baseline expression of pattern recognition receptors MDA5 and RIG-I was observed in upper airway epithelial cells, macrophages, and dendritic cells of older children compared with adults, suggesting that children might be primed to produce IFN in response to viral triggers¹⁰. In summary, while most studies suggest that school-age children exhibit a more effective local innate immune response to SARS-CoV-2 than adults, it remains unclear whether they also mount stronger systemic IFN responses.

Despite significant advances in understanding COVID-19's impact on adults and older children, we still know surprisingly little about how SARS-CoV-2 affects the immune system of infants. A recent study comprehensively analyzed the immune responses induced by SARS-CoV-2 infection (including both Omicron and non-Omicron variants) in infants and young children with mild disease (median age 9 months), who were not hospitalized and did not require high level of care. That study showed the emergence of spike antibodies 4–5 days post infection, which persisted for up to 300 days. Infants and young children showed a robust Th17 response in the nasal mucosa, but no significant detection of pro-inflammatory molecules/cytokines in plasma. However, this cohort did not include cases of severe infection, leaving a significant gap in our understanding of how developing immune system handles severe SARS-CoV-2 infection, particularly during the critical first weeks of life. To address this question, we recruited 26 infants hospitalized with SARS-CoV-2 infection during the first weeks of life (median age ~ 1.6 months) and 14 healthy matched controls. We analyzed their blood immune cell transcriptome at the single-cell level and evaluated serum cytokine profiles, viral loads, and antibody responses. We subsequently compared our findings with published data from both infected adults and infants with mild disease not requiring hospitalization. Unique features of the immune responses to SARS-CoV-2 infection through a wide range of clinical severity during the first year of life are reported.

Clinical categorization of infected infants and changes in PBMC compositions

We collected blood samples and processed serum and peripheral blood mononuclear cells (PBMCs) from 26 infants (median, IQR age: 1.63 [0.93–7.59] months) hospitalized with SARS-CoV-2 infection (pediatric COVID; pCoV) and 14 matched pediatric healthy controls (pHC; median IQR age: 2.01 [1.86–4.28] months) (Fig. 1a). Patient demographics, clinical and laboratory data and treatment are summarized in Supplementary Table 1a, b. We categorized pCoV patients into three groups (G1, G2 and G3; Fig. 1b) based on (i) the time (days) since exposure (DSE) to a known COVID-19 case, (ii) disease severity, and (iii) viral loads (VL) measured within a median of 17 [11–36] hours since hospitalization. Infants in G1 (n = 10) were identified as part of the routine diagnostic work-up during hospitalization and had both milder disease and lower SARS-CoV-2 viral loads (subacute group). Infants in G2 (n = 11) had moderate disease and high SARS-CoV-2 loads, but did not require significant medical interventions (i.e., no supplemental oxygen or intravenous fluids). G3 (n = 5) infants also had high SARS-CoV-2 viral loads but required a higher level of care, such as oxygen supplementation or intensive care unit (ICU) admission (Supplementary Table 1a-d).

To assess the immune response to SARS-CoV-2 in these infants and to determine how these responses correlated with their varying clinical phenotypes, we performed: (i) single cell RNA-seq (scRNA-seq) of PBMCs, (ii) serum cytokine concentration assessment using the 92-Olink platform with a focus on inflammatory pathways, (iii) quantification of antibodies against coronaviruses (anti-Spike and anti-RBD) and anti-IFN activity, and (iv) viral load measurement in nasopharyngeal (NP) swabs (Extended Data Fig. 1a).

For scRNA-seq data analysis, raw data from pHC PBMCs (38% of the pool) and pCoV PBMCs (62% of the pool) were integrated (Fig. 1a). Hybrid transcriptomes (multiplets) were identified using scrublet¹¹ and excluded from the rest of the analysis. After filtering steps (see Methods for details), pHC and pCoV samples yielded, respectively, a mean (± S.D.) of 6,559 (± 2,016) cells and 5,981 (± 1,807) cells per individual, and a mean of 1,411 and 1,603 genes per cell (Extended Data Fig. 1b,c). scRNA-seq profiles that passed quality control (n = 203,402 cells) were then corrected for batch effect using BBKNN¹² (Extended Data Fig. 1d,e, see Methods). Unsupervised clustering of the corrected data, followed by a two-dimensional uniform manifold approximation and projection (UMAP), yielded 25 clusters independently of 10x run (Extended Data Fig. 1e) and subject (Extended Data Fig. 1f) batch effects. Using marker genes, clusters were first mapped to major immune cell types, including CD4⁺ and CD8⁺ T cells, B cells, NK cells, CD14⁺ and CD16⁺ monocytes, conventional dendritic cells (cDCs), plasmacytoid dendritic cells (pDCs), plasma cells (PCs), hematopoietic stem cells (HSCs) and Erythroblasts/Erythrocytes (Eryth) (Fig. 1c,d, Extended Data Fig. 1g,h). To analyze the cellular distribution of Interferon Stimulated Genes (ISGs) within different immune subsets, we calculated an ISG score based on expression levels of genes in interferon modules¹³ (Supplementary Table 3). This revealed increased ISG scores in most cell types, including monocytes, T cells and B cells (Fig. 1e). Cell compositional analyses in pHC, G1, G2 and G3 pCoV groups revealed comparable distribution of pHC and G1 groups, contrasting with striking differences observed in G2 and G3 groups (Fig. 1f). Overall, these high-level analyses revealed significant alterations in the PBMC composition of infants upon SARS-CoV-2 infection and a robust ISG signature in most immune cell types in response to the infection. To refine these alterations, we further subclustered and analyzed immune cell subset.

Expansion of ISG IL-1B monocyte In mild to severe disease

Previous studies have reported alterations in the CD14⁺ compartment of adults with severe COVID-19, including decreased HLA-DR surface expression and the accumulation of immature circulating myeloid cells, suggesting dysregulated myelopoiesis/innate immune response^14,15. To investigate this in our infant cohort, we further clustered CD14⁺ monocytes (n = 16,906) into four subclusters (SCs) (Fig. 2a). SC0 (n = 5,495) expressed antigen presentation-related transcripts (e.g., “HLA-DRA”, “CD74”) together with small and large ribosomal subunits (e.g., “RPS3A”, “RPL3”), while SC2 (n = 4,590) expressed high levels of alarmins “S100A8” and “S100A9”. SC1 (n = 5,017) displayed an ISG signature, including “ISG15”, “IFI6” and “MX1”. SC3 (n = 1,804) was characterized by the upregulation of “IL1R2”, “CD163”, “FKBP5” as well as ISGs (Fig. 2b,c). SC1 and SC3 were expanded in G2/G3 and G3 respectively, while SC0 and SC2 were predominantly present in pHC and G1 (Fig. 2d). SC3 was almost entirely contributed by 3/5 severe patients (G3), two of whom had received systemic steroids. In fact, the SC3 transcriptional profile (Fig. 2b,c) was reminiscent of what was previously described in adults with SARS-CoV-2 infection receiving steroid treatment^16,17,18.

The ISG^hi subcluster (SC1) was rare in pHC and significantly expanded in 2/10 of G1, 11/11 of G2, and 2/5 of G3 patients (Fig. 2e). As both inflammation and type I IFN pathways were previously found activated in monocytes from adults with SARS-CoV-2 infection⁵, we analyzed the co-expression of inflammatory transcripts and ISGs at the single cell level in these infants. Approximately 49% of SC1 monocytes co-expressed “IL1B” and “ISG15” (Extended Data Fig. 2a-c).

Thus, infants with moderate and severe COVID-19 were separated in two groups: the first included all G2 and 2/5 G3 infants and was characterized by expansion of monocytes co-expressing inflammatory molecules and ISGs with low HLA gene expression. The second group consisted of 3/5 G3 infants whose monocytes expressed ISGs and IL1R2 along with CD163, likely reflecting the influence of steroid treatment.

SARS-CoV-2 infected infants maintain CD16 ⁺ monocyte percentages with increased interferon expression.

Contraction of CD16⁺ monocytes has been previously described as a feature of severe COVID-19 in adults with COVID-19¹⁵. Subclustering of CD16⁺ monocytes in infants (n = 5,987) generated three SCs (Figure. 2f). SC0 (n = 3,009) expressed an ISG signature (e.g., “ISG15”, “IFI6” and “MX1”). SC1 (n = 2,217) expressed ribosomal-associated genes “RPL5”, “RPL10A”, or “RPL3” and SC2 (n = 761) was characterized by the expression of inflammatory molecules (e.g., “IL1B, “CCL3”) (Figure. 2g). SC0 (n = 3,009) was significantly expanded in G2 and G3, SC1 was enriched in pHC and G1, while SC2 showed a mild elevation in G2/G3 (Figure. 2h). In contrast to the CD14⁺ population, only about 13% of CD16⁺ SC0 monocytes co-expressed “IL1B” and “ISG15” (Extended Data Fig. 2d,e). The ISG^hi subcluster SC0 was rare in pHC and significantly expanded in 9/11 of G2 and 4/5 of G3 infants (Extended Data Fig. 2f). Unlike severe COVID-19 in adults, which was characterized by CD16⁺ monocyte depletion¹⁹, infants with moderate to severe disease (G2/G3) exhibit normal frequencies of monocytes due to the expansion of ISG^hi CD16⁺ monocytes.

To explore how our findings compare with previous studies, we integrated our scRNA-seq data with a recent study by Wimmers et al.²⁰, which analyzed PBMCs from older infants with mild disease (average 9 months of age, not hospitalized and untreated) during both Omicron and non-Omicron variants of SARS-CoV-2 infection. This cohort included: (i) nine samples collected before (pre, n = 9), during (acute; n = 10) and after infection (convalescent or conv; n = 9) with non-Omicron variants, (ii) eight samples during acute Omicron infection (acute omicron) and (iii) seven matched controls. Our integrative analysis of CD14⁺ and CD16⁺ monocytes from 83 infants in total confirmed the presence of an ISG signature in infants with moderate to severe disease (10/10 G2 and 5/5 G3 patients in our cohort), which contrasted with only 5/10 acute non-Omicron and 1/10 acute Omicron patients in the previous cohort (Extended Data Fig. 2g&i). The number of CD14⁺ monocytes co-expressing “ISG15” and “IL1B” in the acute non-Omicron group in the Wimmers cohort was lower than in G2, G3 groups in our cohort (Extended Data Fig. 2h). These differences are likely attributable to variations in the time from exposure to the virus, the age of the infants, and, most importantly, the clinical disease severity between the two cohorts.

Contraction of cDCs in SARS-CoV-2 infected infants

Similar to classical monocytes, DCs from SARS-CoV-2 infected adults exhibit lower levels of HLA-DR compared to healthy controls²¹. Subclustering conventional DCs (cDCs) in infants (n = 440) generated seven SCs (Fig. 2i and Extended Data Fig. 3a), which were further classified into (i) cDC1 (SC5; expressing “CLEC9A” and “XCR1”), (ii) cDC2 (SC0/SC3; expressing “CD1C” and “CLEC10A”), (iii) monocyte-derived cDCs (Mo-DCs; SC1/SC4; expressing “CD14” and S100s) and (iv) AXL⁺ DCs (SC6; expressing “AXL” and ”DAB2”). The AXL + DCs (SC6) expressed the highest levels of ISGs (Fig. 2j,k and Extended Data Fig. 3b-d). G2/G3 groups showed a marked contraction of cDC1, cDC2 and Mo-DCs (Fig. 2l), while two G2 patients showed an expansion of AXL⁺ ISG^hi SC (SC6) (Fig. 2l and Extended Data Fig. 3e). Overall, our data showed that profiles of G2 and G3 infants were characterized by a decrease in cDC1, cDC2 and Mo-DCs.

The frequency of pDCs in peripheral blood has been found to decrease in adults with COVID-19²², possibly due to pDC apoptosis¹⁶. Subclustering of pDCs in infants (n = 386) generated two SCs (Fig. 2m). SC0 (n = 226) was significantly contracted in G2/G3 patients, while SC1 (n = 160) was significantly expanded in these infants and characterized by the upregulation of ISGs (Fig. 2n-o). The depletion of ISG^low pDCs is consistent with reports in adults with COVID-19; however, a subset of ISG^hi circulating pDCs was found to be expanded in G2/G3 infants.

Circulating megakaryocytes (MGKs) have previously been linked to COVID-19 outcomes in adults²³. Two of the infant’s SCs (in C22, Extended Data Fig. 1g,h) were annotated as megakaryocyte (MGK; n = 408; expression of “PPBP”, “CLU” and “PF4”) and hematopoietic stem cells (HSCs; n = 305; expression of “CD34” and “ITM2C”) (Extended Data Fig. 3f-h). MGK were significantly expanded while HSC were significantly contracted in G1 and G3 groups (Extended Data Fig. 3i).

ISG naïve CD4 T cells are expanded in SARS-CoV-2 infected infants

Activation and proliferation of both CD4⁺ and CD8⁺ T cells have been associated with COVID-19 disease severity^16,4, though no specific T cell signatures have been identified as primary correlates in this context. In our infant study, subclustering CD4⁺ T cells (n = 100,482) yielded four SCs (Fig. 3a). SC0 (n = 43,567), expressing ribosomal protein-associated genes such as “RPL5”, “RPL10A” and “RPL3”, was significantly decreased in G2 and G3 infants. SC1 (n = 22,847), which exhibited an ISG signature (e.g., “IFI44L”, “ISG15” and “XAF1”), was expanded in G2 and G3 infants. SC2 (n = 14,383), characterized by expression of “JUNB/JUND”, “CD69” and “CXCR4” did not differ between infants with COVID-19 and pHC. Finally, SC3 (n = 14,383), expressing the memory marker “S100A4”, was expanded in all three pCoV groups (Fig. 3b,c). The ISG^hi SC1 was present in 2/10 G1, 8/11 G2 and 5/5 G3 infants. Notably, more than 90% of CD4⁺ T cells from a fraction of G2 and G3 patients (e.g., pCoV11 and pCoV26) switched to an ISG^hi state (Fig. 3d-f). SC0 and SC1 exhibited naïve (“CCR7”, “LEF1” and “SELL”) markers, while SC3 displayed memory markers (Fig. 3e-g). Further analysis revealed that SC3 contained both cytotoxic-associated transcripts (GZMK and “CCL5”), and Treg-associated transcripts (”FOXP3” and “IL2RA”) (Fig. 3h) indicating the presence of both CD4⁺ populations within the SC3 memory compartment. However, no Tfh (”CXCR5”), Th1 (“TBX21”) or Th2 (“GATA3”)-related transcripts were not detected in any SC. Analysis of granzyme (GZM) family expression revealed a broad upregulation of “GZMM”, contrasting with an absence of “GZMB” and restriction of “GZMK” to SC3. Our analysis also revealed the expression of the transcription factor “SOX4” within the naïve compartment, which is in line with an immature/naive phenotype. The absence of CX3CR1 expression a lack of CD4⁺ TEMRA cells in this age group (Fig. 3e). Interestingly, while the majority of ISG^hi T cells originated from infected patients, we also detected these cells in pHC, indicating constitutive expression of ISGs (Extended Data Fig. 4a,b).

The integration analysis with the Wimmers et al. dataset further confirmed the expansion of an ISG signature in all G2 and G3 individuals in our cohort, contrasting with 4/10 acute non-Omicron and 1/10 acute Omicron individuals in the Wimmers cohort (Extended Data Fig. 5a). Overall, these data indicate that the blood CD4⁺ T cells of COVID-19 infected infants fall into two main compartments: (i) naïve, as expected the predominant population, that had switched to an ISG^hi state, and (ii) memory, which encompassed a mixture of effector memory T cells, CD4⁺ CTLs and Tregs and was expanded in SARS-CoV-2 infected infants.

ISG CD8 T cells are expanded in infants with COVID-19

COVID-19 associated lymphopenia³ is reported to preferentially impact CD8⁺ T cells⁴, possibly due to T cell exhaustion²⁴. However, the overall CD8⁺ T cell compartment (n = 34,366), however, was not decreased in infants with COVID-19 and was distributed into five SCs (Fig. 4a). CD8⁺T-SC0 (naïve, n = 23,768) expressed ribosomal-associated genes (e.g., “RPL34” and “RPL39), and was contracted in G2 and G3 infants; CD8⁺T-SC1 (ISG^hi; n = 3,483) exhibited an ISG signature (e.g., “IFI44L”, “ISG15” and “XAF1”) and was expanded in G2/G3 infants. CD8⁺T-SC2 (n = 2,967) displayed similar frequencies in infected infants and pHC and expressed cytotoxic genes, including “PRF1”, “GZMA” and “GNLY”, inflammatory chemokines (“CCL4” and “CCL5”), “S100A4” and “CX3CR1” consistent with a TEMRA phenotype. CD8⁺T-SC3 (GzK; n = 2,201) expressed “GZMK” and “CCL5” and was mostly present in G1 but contracted in G2/G3 infants. Finally, CD8⁺T-SC4 (Prolif.; n = 1,947), which expressed proliferation markers such as “MKI67” and HMGBs, was expanded in G2 and G3 infants (Fig. 4b,c). ISG^hi SC was rare in pHC but present in 2/10 G1, 8/11 G2 and 3/5 G3 infants. Notably, more than 50% of CD8⁺ T cells in some G2 (pCoV13, pCoV24, pCoV26) and G3 (pCoV2 and pCoV11) infants demonstrated an ISG^hi state (Fig. 4d). The proliferative SC4 included a subset of cells expressing TEMRA markers, such as “CX3CR1”, “FCGR3A” and “GZMB” (Fig. 4e,f). This SC represented 15–20% of CD8⁺ T cells in 6/16 of the G2/G3 infants (Fig. 4d). Comparison with published data²⁰ on infants with mild disease further confirmed the presence of a robust ISG signature in all G2 and G3 patients in our cohort, contrasting with 4/10 acute non-Omicron and 1/10 acute Omicron individuals in the Wimmers cohort (Extended Data Fig. 5b). In summary, infants with more severe disease exhibited an expansion of ISG^hi CD8⁺ T cells and an induction of CX3CR1⁺ proliferating CD8⁺ T cells, a finding that contrasts with previous reports adults²⁵.

ISG^hi NK cells during severe SARS-CoV-2 infection in infants

A recent study reported an increase of “adaptive” NK cells in the blood of adults with severe COVID-19²⁶. However, little is known about the role of these cells in infants. Herein, NK cell (n = 13,403) subclustering yielded four SCs (Fig. 4g). NK-SC0 (n = 5,013), contracted in G2 and G3 infants, expressed ribosomal-associated genes. NK-SC1 (n = 3,393), contracted in G2 infants, expressed “CD160” and “KLRB1”. NK-SC2 (n = 2,721), expanded in G2/G3 infants, exhibited an ISG signature. NK-SC3 (n = 2,276) displayed similar frequencies in pHC and infected infants and upregulated transcripts such as “XCL1”, “XCL2” and “GZMK” (Fig. 4h-k). While very rare in pHC, the ISG^hi SC2 was present in 2/10 G1, 11/11 G2 and 5/5 G3 infants. Notably, more than 70% of NK cells from 9 G2 and all G3 infants switched to an ISG^hi state (Fig. 4l).

The comparison with previously reported cohorts confirmed the presence of a strong ISG signature in all G2 and G3 patients in our cohort, contrasting with 5/10 acute non-Omicron and 1/10 acute Omicron patients in the Wimmers cohort (Extended Data Fig. 5c).

SARS-CoV-2 infection in infants is associated with increased ISG transitional and naïve B cells

In adults with COVID-19, activated B cells and plasmablasts are expanded^27,4, and increased frequency of extrafollicular (Tbet⁺CD11c⁺CXCR5^neg) B cells correlates with disease severity²⁸. In our study, further clustering of infant’s B cells (n = 30,119) generated five SCs (Fig. 5a). In G2/G3 infants, B-SC1 (n = 5,228) and B-SC2 (n = 4,985) were expanded, while B-SC0 (n = 13,820), B-SC3 (n = 4857) and B-SC4 (n = 1,229) were contracted (Fig. 5b). Differential marker expression distinguished three compartments: (i) Naïve B cells (NBC; “IGHD”, “CCR7” and “SELL”), (ii) Transitional B cells (TrBC; “MME (CD10)”, “CD9”, and “CD24”) and (iii) Memory B cells (MBCs; “CD27”, “IGHA1”, “IGHG1” and “S100A4”). An ISG signature (e.g., “IFI44L”, “ISG15” and “IRF7”) was detected in both the NBC and TrBC compartments. Thus, NBCs included B-SC0 (ISG^low; expanded in pHC/G1) and B-SC2 (ISG^hi; expanded in G2/G3); TrBCs encompassed B-SC3 (ISG^low; expanded in pHC/G1) and SC1 (ISG^hi; expanded in G2/G3). SC4 included transcripts characteristic of double negative (DN)2 memory cells (e.g. “ITGAX”, “TFEC” and ”FGR”). Interestingly, the ISG^hi naive B-SC2 expressed “TLR7”, a marker of both extrafollicular activated naïve and DN2 cells²⁹ (Fig. 5c,d). In addition to memory markers, B-SC4 also expressed B cell survival and plasma cell differentiation markers such as “TNFRSF13B” (encoding TACI) and “TNFRSF17” (encoding BCMA), respectively. Although very rare in pHC, the ISG^hi SC was present in 2/10 G1, 11/11 G2 and 5/5 G3 infants. Overall, > 70% of naïve (B-SC2) and transitional (B-SC1) B cell compartments from 13/16 of the G2/G3 infants switched to an ISG^hi state (Fig. 5e). While the majority of the ISG^hi B cells originated from infected patients, we also identified an ISG signature in both naïve and transitional B cells from healthy controls(Extended Data Fig. 4c,d).

The comparison between cohorts demonstrated the presence of a strong ISG signature in all G2 and G3 patients in our cohort, contrasting with 10% and 40% of acute Omicron and non-Omicron individuals, respectively, in the Wimmers cohort (Extended Data Fig. 5d).

Finally, PC (n = 544) subclustering generated three SCs (Fig. 5f), which were distinguishable based on IGH expression (Fig. 5g). Although the low number of cells is a limitation, cell compositional analysis did not reveal significant differences between COVID-19 infant groups and pHC (Fig. 5h), which contrasts with what was reported in adults with COVID-19³⁰.

Patient stratification using cell frequencies.

To integrate subcluster analysis data (Extended Data Fig. 6a-d) with clinical disease severity, we performed an unsupervised clustering based on the abundance of detected subclusters (n = 40) across infants with COVID-19 (n = 26) and pHC (n = 14). The study subjects were accordingly clustered into two main sets (Fig. 6a). The first set, which included 10 infants with COVID-19 (3/5 G3, 6/11 G2 and 1/10 G1 infants), was characterized by the expansion of ISG^hi SCs. The second set, which included the remaining 16 COVID-19 infants together with the 14 pHC, displayed ISG^low SCs including naïve CD4⁺, CD8⁺ T cells and B cells (Fig. 6b). Interestingly, the two G3 patients who clustered with pHC/G1 (pCoV17 and pCoV21) had received systemic steroids, consistent with the modulatory effect of corticosteroids on type I IFN expression³¹. These data therefore support that clinically severe SARS-CoV-2 infection in infants is associated with a robust ISG response.

Increased serum concentrations of inflammatory cytokines in SARS-CoV-2 infected infants.

To complement our transcriptional profiling studies, we measured serum cytokines in 34 children with COVID-19 (7 G1, 20 G2 and 7 G3) and 20 age-matched HCs using the Olink inflammatory panel (n = 92 analytes). This cohort included 24 of the 26 infants profiled with scRNAseq (Extended Data Fig. 1a, Supplementary Table. 1e). Overall, 72 cytokines were consistently detected in serum samples (Extended Data Fig. 6e); and 33 of them showed significantly different concentrations between disease severity groups (Fig. 6c). IFNg concentrations were increased in infants with COVID-19 across all disease groups. Cytokines and inflammatory proteins such as IL6, IL8, IL17C, IL18R1 and CXCL10 were particularly increased in G3 infants (Fig. 6d), while soluble CD6 and TNFS12/TWEAK were markedly decreased in this severe group compared with HCs. Patients in the Wimmers cohort did not show significant increases in plasma inflammatory cytokines possibly due to differences in time since infection and/or clinical disease severity. In summary, SARS-CoV-2 infected infants displayed increased serum concentrations of inflammatory cytokines, especially in those with severe disease (G3).

Anti-SARS-CoV-2 and anti-IFN antibody responses

To assess the infants’ humoral response to SARS-CoV-2 and seasonal coronaviruses, we measured antibody levels against: (i) SARS-CoV-2 full-length spike, RBD and S2 subunit, (ii) seasonal coronaviruses full-length spike, (iii) S1 subunit of beta-coronaviruses OC43 and HKU1, and (iv) full-length spike of alpha-coronavirus 229E. Samples from 16 infants were obtained at the time of hospitalization, and 13 of those had paired follow-up samples four weeks later. The 13 infants with paired samples, who were also included in the scRNAseq cohort (Extended Data Fig. 1a, Supplementary Table. 1f), predominantly (12/13) showed a significant increase in IgG antibody titers against SARS-CoV-2 antigens at follow-up, including spike, RBD and S2 subunit (Extended Data Fig. 7a,b). While 10 infants increased their anti-S2 and anti RB levels by 20 and ~ 8 fold, respectively, no increase in antibody titers against any of the seasonal alpha- or beta- coronaviruses was observed, despite the presence of detectable IgG levels at baseline (T0); (Extended Data Fig. 7c). Given the young age of this cohort (median 3.2 months), we hypothesized that the detected antibodies against seasonal coronaviruses are likely of maternal origin.

Finally, although anti-IFN autoantibodies have been identified in adults with severe COVID-19³², none of the patients (n = 16) or HCs (n = 6) in this pediatric cohort demonstrated anti-IFN activity in either the acute or convalescent serum samples (Extended Data Fig. 7d).

Lymphocytes from infants express a broader ISG signature compared to adults

We next compared our infant data with a scRNA-seq PBMC dataset (GSE161918¹⁶) from a reported cohort of 28 adults with COVID-19 and controls. For this comparison, we included data from the adult patients’ first time-point (T0), as well as from 11 healthy matched controls (aHC)¹⁶. After concatenating adult (n = 39) and infant (n = 40) datasets, we applied our pipeline (see methods for details) on doublet-cleaned data, which included pre-processing, batch correction (using Harmony³³), and unsupervised clustering. Based on the criteria we used for the infant cohort (excluding viral loads, which were not available for the adult dataset), we categorized adult patients into three clinical groups, from lower to higher severity (aG1, aG2 and aG3; Fig. 7a). The first round of clustering generated 28 clusters and seven cell-types (Fig. 7b-d) that were then analyzed separately.

CD14⁺ monocytes included four subclusters (SCs), which were categorized into: (i) ISG^low (SC0), (ii) ISG^hi (SC1), (iii) ISG^hi inflam⁺ (SC2) and (iv) CD163⁺ IL1R2⁺ (SC3) (Fig. 7e & Extended Data Fig. 8a,b). Cell composition analysis of CD14⁺ monocytes showed a decrease of SC0 (ISG^lo) and a switch to SC1 (ISG^hi) and/or SC2 (IL1B⁺ ISG^hi) in pG2/pG3 and aG2/G3 groups compared to their respective controls. Interestingly, we reproduced the increase in SC3 (IL1R2⁺ CD163⁺) in patients receiving steroid therapy (Fig. 7f). The analysis of CD16⁺ monocytes showed a switch to an ISG^hi state in both G2/G3 infants and G3 adults (Extended Data Fig. 8c-e).

Subclustering CD4⁺ T cells generated six SCs (Fig. 7g), annotated as: (i) naïve (SC0 and SC1; “SELL” and “CCR7”), with SC0 being ISG^low and SC1 ISG^hi, (ii) memory (SC2 and SC4, “S100A4”), (iii) SOX4⁺CD4⁺ T cells (SC3; “SOX4”) and (iv) regulatory T cells (SC5, “FOXP3”, “IL2RA”). SC4 (CTLs) exhibited the cytotoxic cell-associated chemokine “CCL5” (Extended Data Fig. 8f). Cell composition analysis showed: (i) an expansion of naïve cells in pediatric healthy controls (pHC) compared to adult healthy controls (aHC), (ii) a contraction of naïve cells in pG2/pG3 compared to pHC/pG1, (iii) expansion of memory cells (SC2) in aHC compared to pHC, (iv) expansion of CTLs (SC4) in aHC, and (v) an expansion of SOX4⁺ CD4⁺ T cells in pHC. Tregs (SC5) did not significantly change upon infection in either infants or adults. The ISG^hi CD4⁺ T cell SC (SC1) was present in both infected infants (2/10 pG1, 8/11 pG2 and 5/5 pG3) and adults (6/21 aG3 patients), although their proportion was higher in infants (Fig. 7h).

B cell subclustering generated seven SCs (Fig. 7i), annotated as: (i) naïve (SC0, “CCR7”), (ii) activated (SC1, “CD69”) (iii) memory (SC2, “S100A4”) and (iv) transitional (SC4; “CD9”, “MME (CD10)”). SC1 encompassed ISG^hi B cells and SC5 upregulated “CD83” (Extended Data Fig. 8g). Cell composition analysis showed: (i) an expansion of naïve B cells in pHC compared to aHC, (ii) an expansion of memory B cells in aHC compared to pHC, and (iii) a subtle ISG^hi expansion in infected adults (aG3) contrasting with a near-complete ISG^hi status in the infected infants with severe disease (pG2/pG3) (Fig. 7j).

CD8⁺ T cell subclustering generated five SCs (Extended Data Fig. 9a), classified as: (i) naïve (SC0; “RPLs”), (ii) TEMRA (SC1; “GZMB”), (iii) Gzk (SC2, “GZMK”), (iv) MAIT (SC3, “KLRB1” and “ZBTB16”) and (v) proliferative CD8 T cells (SC4, “MKI67”) (Extended Data Fig. 9b). As expected, cell composition analysis showed expanded naïve CD8⁺ T cells in pHC and TEMRA/MAIT in aHC (Extended Data Fig. 9c). NK cell subclustering generated two SCs (Extended Data Fig. 9d). SC0 upregulated “FCGR3A” (CD16⁺ NK), and SC1 was XCL1⁺ and no disease severity-associated differences were detected (Extended Data Fig. 9e,f).

Overall, these comparative analyses (Extended Data Fig. 10a-d) showed that T and B cells from infants with severe disease almost entirely switch to an ISG^hi state – while only a fraction of T and B cells in adults with severe disease become ISG^hi.

In this study, we characterized the immune response to SARS-CoV-2 infection during the first weeks of life. By assembling a cohort of young infants with varying disease severity – ranging from subacute/mildly symptomatic (G1), moderate (G2), and severe (G3) – we were able to perform an in-depth analysis of disease severity-associated changes in cell composition and transcriptional profiles at the single cell level, serum cytokine concentrations, and antibody responses. Unlike previous studies²⁰, our cohort focused on very young infants, covering an age range of 3 days to 19 months. We found disease severity-associated cell composition and transcriptional changes in almost all immune cell types investigated, along with increased concentrations of inflammatory cytokines, IFNg, and IFN-inducible cytokines. Importantly, these infants mounted consistent antibody responses to SARS-CoV-2 antigens in the absence of anti-IFN autoantibodies.

During the initial stages of the pandemic, it was believed that infants and children were resistant to COVID-19. However, subsequent studies showed that they not only get infected but can also develop moderate to severe disease³⁴. Despite this, these patient populations have been largely understudied, therefore there is limited information on the immune responses of infants who develop severe versus mild disease. A recent study²⁰ analyzed the mucosal and systemic immune response to SARS-CoV-2 infection in infants and young children with mild disease not requiring hospitalization. Children in that cohort showed durable antibody responses, and a robust mucosal immune response characterized by the production of inflammatory cytokines, type I IFN, as well as detection of T helper (Th) 17 and neutrophil-derived proteins (e.g., IL-17, IL-8, and CXCL1). In the systemic compartment, these children had increased chemokine and IFNa concentrations that correlated with viral loads and with marked ISG expression in myeloid cells²⁰. However, no inflammatory cytokines were detected systematically.

Our study revealed that infants with moderate/severe disease have a distinct immune response, including a prominent ISG transcriptional response across all PBMCs. These transcriptional differences were partially anticipated, as bulk RNA-seq analyses have shown a similar correlation between disease severity and transcriptional IFN responses in infants with other respiratory viral infections^35,36.

The prominent ISG signature in our cohort was expressed in both the myeloid and lymphoid compartments. Among CD14⁺ monocytes, three distinct cell states were identified according to the co-expression of ISGs and pro-inflammatory IL-1B transcripts: (i) ISG^low + IL1-B, which was expanded in HC and G1, (ii) ISG^high + IL1-B, which was expanded in G2 patients, and (iii) ISG^high only, which was expanded in G3 patients. Several studies suggest a cross-regulation between IL-1B and IFN³⁷, but monocytes co-expressing ISGs and IL-1 were reported in SARS-CoV-2 infected adults³⁸, as well as in pediatric patients with systemic lupus erythematosus (SLE)^39,40. This suggests that SARS-CoV-2 may, directly or indirectly, alter this cross-regulation, a finding that needs further mechanistic exploration.

Lymphocytes from COVID-19 infants showed an ISG signature far more prominent than that observed in adults. This suggests that infants might either produce more type I IFN than adults or respond more robustly to it. Such an enhanced IFN response might represent an innate mechanism of protection for infants against viral infections, particularly while their adaptive immune system is still developing. Although the ISG^hi CD4⁺ T cell subset (CD4_SC1) was mostly contributed by moderate to severe (G2/G3) patients, it was also present at low frequency in every HC, perhaps indicating a constitutive activation of the IFN pathway in the steady state of young infants (Extended Data Fig. 4a-b).

Circulating CD14⁺ monocytes with macrophage markers (“CD163”, “IL1R2” and “FKBP5”) were another notable feature observed in G3 infants. IL-1 Receptor antagonist (IL1R2) is a secreted decoy receptor for IL1A, IL1B and IL1RN that modulates IL1 activities⁴¹ and reduces IL-1-mediated inflammation. A similar CD163⁺ IL1R2⁺ population was expanded in 7 out of 16 adult patients receiving corticosteroid treatment¹⁶. Indeed, two infants on our cohort who were treated with systemic steroids (dexamethasone) exhibited this CD163⁺ IL1R2⁺ CD14⁺ monocyte subset. It is tempting to speculate that these cells could contribute to a “return to homeostasis”. COVID-19 infected infants also displayed a reduced frequency of pDCs, as found in adult patients. However, unlike adult pDCs, which expressed a signature of apoptosis¹⁶, the pDCs in infants exhibited and ISG signature.

Analysis of the infant’s lymphoid cell populations showed that, as expected given their young age, the majority displayed a naïve phenotype. However, within CD4⁺ T cells, we could detect a memory compartment, including Tregs and CD4⁺ T cells expressing cytotoxicity-associated markers. The distribution of GZMs transcripts was particularly intriguing, with broad GZMM expression, contrasting with the absence of GZMB. Contrary to previous reports ⁴², we did not detect an increase in the frequency of CXCR5⁺ CD4⁺ Tfh cells in infants with moderate to severe disease. While this discrepancy might be due to methodological variations, our data suggest an enhanced role of the extrafollicular B cell differentiation pathways, similar to what was reported in adults^28,43.

Regarding CD8⁺ T cells, the most remarkable finding was the expansion of cycling MKI67 CX3CR1⁺ CD8⁺ T cells in G2/G3 infants, a cell population that has, until now, only been identified in children with multisystem inflammatory syndrome in children (MIS-C)²⁵. While in children with MIS-C the presence of these cells was reported to correlate with the use of vasopressor drugs²⁵, only one infant in our cohort received such medication, suggesting alternative induction pathways. Further, studies assessing the SARS-CoV-2-specificity of these unique T cells could provide valuable insights into the origin and role of this population.

The B cell compartment, as expected given the young age of the infants, was predominantly composed of naïve B cells with limited numbers of memory cells. Yet, the magnitude of the B cell ISG signature was remarkable. ISG^hi subclusters were identified in 65% (17/26) of the infected infants, though also detected at low frequency in every HC, perhaps indicating a tonic activation of the IFN pathway in the steady state (Extended Data Fig. 4c-d). In addition, markers restricted to extra-follicular B cells and plasma cell precursors were identified within these predominantly naïve populations. Overall, despite the lack of well-defined blood T helper subsets, infants with COVID-19 showed a robust antibody response to the different SARS-CoV-2 antigens, regardless of disease severity. We also detected antibodies against beta coronaviruses, likely of maternal origin, however, contrary to previous hypotheses, these pre-existing antibodies from endemic coronaviruses did not appear to prevent the infection in these young infants. In addition, none of the infants in our cohort displayed anti-interferon antibodies. In conclusion, our analysis of infants with severe disease revealed a robust ISG signature across most immune cell types, as well as a strong systemic inflammatory response, evidenced by increased transcription and high serum concentrations of inflammatory cytokines. Notably, the co-expression of ISGs and inflammasome-related genes, especially in monocytes, was remarkable, especially in infants with severe disease²⁰.

Taken together, our findings highlight distinct immune alterations triggered by SARS-CoV-2 infection in young infants and identify important differences compared with infants with mild disease and with adults. The magnitude and expansion of the ISG signature across most immune cell populations, coupled with the capacity to mount a significant antibody response to the virus despite a predominantly naïve adaptive immune compartment, are among the most remarkable and unexpected findings. Whether these differences are generalizable to other viruses warrants future studies. These insights may have important implications for developing more effective preventive and therapeutic strategies for this vulnerable population.

Limitations of the study when performing

Our study has the limitations associated with the difficulty of studying infants longitudinally, especially those severely ill. Other studies have described biphasic early and late immune responses to SARS-CoV-2, pointing out to the importance of timing in immune studies¹⁶. For this reason, we accounted for duration of viral exposure and/or symptoms, which allowed us to classify infants in three groups and investigate the differences in immune responses by multidimensional analyses. Infants in G1 were enrolled later after SARS-CoV-2 exposure compared with infants in G2 and G3, which could partly explain the more prominent ISG response in these two latter groups. Nevertheless, accounting for differences in the duration of infection allowed us to document that ISG responses are influenced by time, and not only by disease severity. Except for follow-up serologic studies, samples for scRNAseq and serum cytokine analyses were obtained at a single time point during the hospitalization. Finally, we did not measure the impact of SARS-CoV-2 infection at the mucosal level, the primary site of infection.

In summary, our multidimensional approach leveraging small volume blood samples from infants with COVID-19 and variable disease severity provides valuable data to better understand the complex immune alterations that take place during SARS-CoV-2 infection in this age group. Future studies including correlations with mucosal responses are warranted.

Study design

A convenience sample of children <2 years of age hospitalized with COVID-19 were prospectively enrolled at Nationwide Children’s Hospital (NCH) in Columbus, Ohio, USA from March 2020 through January 2021. Blood (2.5-5 mL) and nasopharyngeal samples were obtained within a median (25%-75% interquartile range [IQR]) of 31 [12–57] hours of hospitalization for multiomic analyses and SARS-CoV-2 quantitation by real-time polymerase chain reaction (PCR)⁴⁶. Children with multisystem inflammatory syndrome (MIS-C), or those in whom confirmatory nasopharyngeal SARS-CoV-2 PCR was negative were excluded from the study. At enrollment, we collected demographic and clinical information using a standardized questionnaire designed for the study, and information transferred to a secure database (REDCap). The information collected included duration of illness or time since exposure to a COVID-19 case, and standard parameters of disease severity including oxygen administration, pediatric intensive care unit (PICU) admission, duration of hospitalization, and administration of COVID-19 directed therapies including remdesivir or systemic steroids. During the study period all children hospitalized at NCH, irrespective of the presence of symptoms, underwent nasopharyngeal SARS-CoV-2 testing using a PCR assay per standard of care as described⁴⁶. As a refence for all immune assays, we also included in the study a cohort of age-matched healthy control infants with no respiratory symptoms or treated with antibiotics within two weeks of enrollment. All healthy controls were enrolled pre-pandemic. Healthy controls were typically enrolled in the operating room while undergoing minor scheduled surgical procedures not involving the respiratory tract, or at the primary care offices during well-child visits. The study was approved by the Nationwide Children’s Hospital (NCH) IRB (18-00591). Written informed consent was obtained from all children’s guardians before study participation. Further clinical details of all participants are summarized in Supplementary Table 1.

PCR assays for SARS-CoV-2 viral loads

Nasopharyngeal (NP) swabs were collected at enrollment, placed in viral transport media, transported immediately to the laboratory, aliquoted and stored at –80°C. Similarly, blood samples were collected in EDTA tubes (BD Vacutainer, Franklin Lakes, NJ, USA), centrifuged, and aliquots of plasma were stored at –80°C until processed in batches. Viral RNA was extracted from 200 microliters of NP or plasma samples using the QIAcube HT instrument (Qiagen Inc, Germantown, MD, USA) and eluted into 100 microliter volume. In brief, SARS-CoV-2 viral loads were measured using a two-step reverse-transcription (RT) quantitative PCR assay targeting the conserved region of the N1 gene using described primers and probe⁴⁷ A standard curve was generated from a dilution series of a known concentration(10⁷copies/mL) of the N1 gene of SARS-CoV-2. Standards and negative controls were included and tested with each PCR assay. The lower limit of detection of the assay was 1-2 copies/reaction (~200 copies/mL).

Cytokine assays

Serum cytokines were measured in duplicate using two different platforms. For interferon concentrations we used the Biolegend (San Diego CA) Legend plex human antivirus 13-plex cytokine panel which includes IL1b, IL6, TNFa, IP10, IFNa1, CXCL8, IL12p70, IFNa2, IFN l2/3, GMCSF, IFNb, IL10, IFNg. Samples were analyzed at Nationwide Children’s Hospital, following the instructions of the manufacturer as described⁴⁸. In addition, we applied the Olink (Boston, MA) inflammation panel that includes 92 proteins. Samples were analyzed in the Olink Boston Laboratory.

Enzyme-Linked Immunosorbent Assay (ELISA)

Ninety-six-well microtiter plates (Thermo Fisher) were coated at a concentration of 2 ug/mL of recombinant protein overnight (4°C): SARS-CoV-2 full-length spike, RBD and S2 subunit. For seasonal coronaviruses, full-length and S1 subunit was used for beta- OC43 and HKU1; and full-length spike for alpha- 229E. The next day, plates were washed three times with PBS (phosphate-buffered saline; Gibco) containing 0.1% Tween-20 (T-PBS, Fisher Scientific) using an automatic plate washer (BioTek). After washing, the plates were blocked for 1h at room temperature with 200 ml blocking solution (PBS-T with 3% (w/v) milk powder (American Bio)) per well. The blocking solution was removed, and serum samples diluted to a starting concentration of 1:80, serially diluted 1:3 in PBS-T supplemented with 1% (w/v) milk powder and incubated at room temperature for 2 h. The plates were washed three times with PBS-T and 50 ml anti-human IgG (Fc-specific) horseradish peroxidase antibody (HRP, Sigma, #A0170) diluted 1:20,000 in PBS-T containing 1% milk powder was added to all wells and incubated for 1 h at room temperature. After washing, 100 μL of peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine, TMB, Rockland) was added and incubated at room temperature for 30 min. The reaction was stopped with 1 N sulfuric acid solution (Fisher Science) and the plates read at a wavelength of 450 nm with a plate reader (BioTek). The results were recorded in Microsoft Excel and AUC values were computed by plotting normalized optical density (OD) values against the reciprocal serum sample dilutions for ELISAs in GraphPad Prism.

Functional evaluation of anti-IFN-I auto-antibodies

A549 cells (ATCC; CRM-CCL-185) were cultured in Dulbecco’s modified Eagle medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; PEAK) and penicillin-streptomycin (Gibco), at 37°C, under an atmosphere containing 5% CO2. Cells were checked periodically and tested negative for mycoplasma contamination. A549 cells were seeded in 96-well plates at a density of 3x10³ cells/well. The next day, plasma samples or a commercial anti-IFNα2 antibody (catalog number 21100-1; R&D systems) were serially diluted (10-fold) and incubated with 50 pg/ml recombinant IFN-α2 (catalog number 11101-2; R&D systems) for 1 h at 37°C (starting concentration: plasma samples=1/100 and anti-IFN-α2 antibody=1/1000). Following this incubation period, the cell culture medium in the 96-well plates was removed by aspiration and replaced with the plasma/antibody-IFN-α2 mixture. Each sample was tested once, in triplicate. The plates were incubated overnight and the plasma/antibody-IFN-α2 mixture was removed by aspiration. The cells were then washed three times with PBS to remove potential anti-influenza neutralizing antibodies and infected with a recombinant Cal/09 virus expressing NS1-mCherry (CalNSmCherry) at MOI=0.5. One day after infection, cells were fixed with 4% formaldehyde, washed twice with PBS, and stained with DAPI. The percentage of infected cells was quantified using the Celigo (Nexcelcom) imaging cytometer. All antibody assays were performed at Mt. Sinai School of Medicine.

Blood preparation for single cell RNA sequencing (scRNA-seq)

PBMCs were thawed quickly at 37^oC and into DMEM supplemented with 10% FBS. Cells were quickly spun down at 400 g, for 10 min. Cells were washed once with 1 x PBS supplemented with 0.04% BSA and finally re-suspended in 1 x PBS with 0.04% BSA. Viability was determined using trypan blue staining and measured on a Countess FLII. Briefly, 12,000 cells were loaded for capture onto the Chromium System using the v2 single cell reagent kit (10X Genomics). Following capture and lysis, cDNA was synthesized and amplified (12 cycles) as per manufacturer's protocol (10X Genomics). The amplified cDNA was used to construct an Illumina sequencing library and sequenced on a single lane of a HiSeq 4000.

Single-cell Raw data processing and data combining

Illumina basecall files (*.bcl) were converted to fastqs using cellranger v3.0.2, which uses bcl2fastq v2.17.1.14. FASTQ files were then aligned to hg19 genome and transcriptome using the cellranger v3.0.2 pipeline, which generates a gene - cell expression matrix. The samples were combined using cellranger aggr from cellranger, which aggregates outputs from multiple runs, normalizing them to the same sequencing depth (normalize=mapped) and then re-computing the gene-barcode matrices and analysis on the combined data. The sequencing information for individuals included in the study (e.g., number of reads per cell or UMI counts per cell) are shown in Supplementary Table 2.

Scrublet for multiplet prediction and removal

Generally, we expected about 2 to 8% of the cells to be hybrid transcriptomes or multiplets, occurring when two or more cells are captured within the same microfluidic droplet and are tagged with the same barcode. Such artifactual multiplets can confound downstream analyses. We applied Scrublet¹¹ python package to remove the putative multiplets. Scrublet assigns each measured transcriptome a ‘multiplet score’, which indicates the probability of being a hybrid transcriptome. Multiplet scores were determined for each individual (using the raw data). Before/after multiplet removal, the number of cells was 313,076/300,708.

Single-cell preprocessing, dimension reduction, graph-based clustering, and cluster annotation

The cleaned (after multiplet removal using scrublet¹¹) aggregated matrices were fed into the Python-based Scanpy⁴⁴ workflow (https://scanpy.readthedocs.io/en/stable/), which includes preprocessing, visualization, clustering and differential expression testing. The pipeline we applied was inspired by the Seurat R package workflow and can be found here: https://github.com/dnehar/Infants_Cov19.

Quality control and cell filtering

We applied the following filtering parameters: (i) all genes that were not detected in ≥ 3 cells were discarded, using pp.filter_genes function, (ii) cells with less than 400 total unique transcripts were removed prior to downstream analysis using pp.filter_cells function, (iii) cells in which > 25% of the transcripts mapped to the mitochondrial genes were filtered out, as this can be a marker of poor-quality cells and (iv) cells displaying a unique gene counts > 4,500 genes were considered outliers and discarded.

Data normalization, log transformation and scaling

We normalized the data using the pp.normalize_per_cell function. Thus, library-size normalization was performed based on gene expression for each barcode by scaling the total number of reads per cell to 10,000. We log-transformed the data using the pp.log1p function and scaled to unit variance using pp.scale function (with the following parameters: max_value=10). The 1,362 highly variable genes (HVG) were identified using filter_genes_dispersion function (with the following parameters: min_mean=0.0125, max_mean=3, min_disp=0.5).

Linear dimensional reduction using PCA

To reduce the dimensionality of the data, we ran principal component analysis (PCA) using tl.pca function, which reveals the main axes of variation and denoises the data. The contribution of each PCs to the total variance was assessed using pl.pca_variance_ratio function.

Neighborhood graph computing, embedding and clustering

The neighborhood graph of cells was computed based on the PCA representation of the data matrix, using pp.neighbors function (with the following parameters: n_neighbors=10, n_pcs=40). The neighborhood graph was then embedded using UMAP⁴⁹ (tl.umap function) and visualized using pl.umap function. We finally used the Leiden graph-based clustering using tl.leiden function with resolution=1.2 (stored in ‘Res1_2_BC’ column).

Batch effect correction

To account for technical source of variation, such as ‘10X runs’ or ‘Years’, we applied a batch effect correction using BBKNN (github.com/Teichlab/bbknn).

bbknn.bbknn function was ran using the following parameters: metric="angular", approx=True, neighbors_within_batch=5, n_pcs=20, trim=50 and copy=True.

All the analyses presented in this study were done on the BBKNN corrected data. UMAP visualization and Leiden clustering were then computed on the BBKNN corrected using the following parameters:

sc.tl.umap(BBKNN_corrected_object, min_dist=0.3, n_components=3)
sc.tl.leiden(BBKNN_corrected_object, resolution=1.2, key_added='Res1_2_AC')

To account for batch effects in the pediatric-adult (paCov) dataset, we used harmony³³.

IFN score calculation

ISGs (from previously described interferon-related modules⁵⁰, Supplementary Table 3) were used to score IFN expression in each cluster/subcluster generated in the pCov40 dataset. To do so, we calculated the mean expression for each cell, within each cluster/subcluster using the h5ad object (adata), as follow: adata.obs['ISG_score’] = adata.X[:,IFN_markers].mean(1).

Finding marker genes/evaluation of cluster identity

To annotate the clusters generated from the BBKNN corrected object, we used both differential expression analysis between clusters and classification based on putative marker gene expression. We applied tl.rank_genes_groups function to compute a ranking for the differential genes in each cluster/Subcluster, comparing each cluster to the rest of the cell using Wilcoxon test. We only considerate clusters/Subclusters that showed distinct transcriptomic programs. The top 100 marker genes for cluster and subclusters, in pCoV are included in Supplementary Table. 4 & 5, respectively. The top 10 marker genes were visualized using the sc.pl.rank_genes_groups_matrixplot() function.

Subclustering parameters and data cleaning.

A script showing the subclustering process can be found here: https://github.com/dnehar/Infants_Cov19. We only considered SC defined by distinct gene sets, by merging similar ones. Based on the number of cells and to avoid over-clustering, we used different clustering resolutions. The following parameters of resolution were used: 0.4 (CD14⁺ and CD16⁺ monocytes, NK cells, CD4⁺ T cells, MGK), 0.6 (B cells, CD8⁺ T cells and cDCs), 0.8 allowed the separation of PC and pDC (C24_PCs_pDC in Fig.1d). We discarded two small SC of CD14⁺ monocytes (one PF4⁺ and the other was IGKC⁺). cDC (n=504 cells) which initially mingled with CD14⁺ monocytes were separated after the subclustering step. Suspected doublets in NK cells (PF4⁺ SC), B cells (NKG7⁺, GNLY⁺ SC), MGK (T and B markers), Eryth (T, B, NK markers) were discarded from the analysis. Overall, the number of cells after multiplet removal (n=300,708), the filtration (n=256,766), and visual inspection steps was 203,402 cells. A similar strategy was applied to the pediatric and adult combined (paCoV) cohort. Number of cells before filtration was 499,996 and after filtration 440,494 cells.

Adult COVID-19 data Integration (paCoV dataset)

We first downloaded scRNA-seq raw counts (GSE161918¹⁶) from an adult cohort, which included 28 patients (aCov; we only selected the time 0), as well as 11 healthy controls (aHC). After converting the Seurat object into an anndata object (.h5ad), we concatenated adult (28 aCov collected at time 0 and 11 aHC) and pediatric (14 pHC and 26 pCov) raw counts, resulting in an object including 79 samples. We then cleaned the data from doublets using Scrublet¹¹ and applied a computational pipeline, including pre-processing, batch correction (using Harmony³³) and non-supervised clustering.

Comparison with Wimmers et al. cohort²⁰.

We downloaded the single cell Multiome dataset (cell ranger outputs) from GEO (GSE239799²⁰). We then concatenated the raw counts (n=43) with our pediatric cohort (n=40), resulting in anndata object (.h5ad) including 83 samples. We then cleaned the data from doublets using Scrublet¹¹ and applied a computational pipeline, including pre-processing, batch correction (using Harmony³³) and non-supervised clustering. The Wimmers et al. cohort included 9 longitudinal samples (pre-infection or “pre” – acute non-Omicron or “acute” and convalescent or “conv”), 1 acute non-omicron, 8 acute Omicron and 7 healthy controls²⁰.

Statistical analysis

Statistical analysis was performed using R/4.0.2. Tests were used to determine data distribution and depending on the normality of the data, comparisons were performed using the Student t test (for two groups, parametric) or the non-parametric the Wilcoxon signed rank test (for two groups, paired) with two-tailed P values unless otherwise stated. Differences were considered significant when P <0.05 (*), P < 0.01 (**), P < 0.001 (***) and P < 0.0001 (****).

Python module versions

scanpy==1.7.1 anndata==0.7.5 umap==0.4.6 numpy==1.19.2 scipy==1.5.2 pandas==1.1.3 scikit-learn==0.24.1 statsmodels==0.12.2 python-igraph==0.8.3 leidenalg==0.8.3

Acknowledgments

We are grateful to the patients, their families, the healthy subjects who participated in our study. We thank the Nationwide Children’s Hospital Infectious Diseases clinical research team for study subject recruitment and management; M. Collet, L. Choquette for help with IRB and uploading the data to dbGAP. We gratefully acknowledge the contribution of the Single Cell Biology service, the Genome Technologies service, and Cyberinfrastructure high performance computing resources at The Jackson Laboratory for expert assistance with the work described herein. These shared services are supported in part by the JAX Cancer Center (P30 CA034196). This work was supported by grants U01 AI 131386 (to O.R. and J.F.B) and start-up funds from the Jackson Laboratory. This work is partly supported by CRIPT (Center for Research on Influenza Pathogenesis and Transmission), a NIAID funded Center of Excellence for Influenza Research and Response (CEIRR, contract # 75N93021C00014) to A.G.-S, O.R. and T.A. This work is also partly supported by grants U19AI135972, U19AI142733 and U19AI168631 to A.G.-S.

Author contributions

D.N-B. designed the experiments, processed, analyzed, interpreted the data, generated the figures, and wrote the first draft of the manuscript. A.M. collected the samples, clinically characterized the patients, interpreted data, and reviewed the manuscript. R.Y. and G.C. helped performing analysis under the supervision of D.N-B and J.F.B. G.C. helped collecting, processing, and analyzing adult data under the supervision of D.N-B and J.F.B. Z.X. analyzed cytokine data, interpreted the data, generated figures, and reviewed the manuscript. T.A. analyzed and interpreted the data, generated figures, and reviewed the manuscript. L.M. analyzed, interpreted the data, generated the figures, and reviewed the manuscript. A.C. helped generating the IFN auto-antibodies data under the supervision of L.M. T.A. generated antibody data. A. G-S. interpreted the data and reviewed the manuscript. V.P. interpreted the data and reviewed the manuscript. O.R. conceived, supervised the study, interpreted the data, and reviewed the manuscript. J.F.B. conceived and supervised the study, interpreted the data, and reviewed the manuscript. All authors reviewed and contributed to the final manuscript.

Competing interests statement

At the time of the study, J.F.B. was a member of the BOD and SAB of Neovacs and Ascend Biopharma, and a SAB member of Cue Biopharma. J.F.B is now an employee of Immunai. A.M. has received research grants from Janssen and Merck, fees for participation in advisory boards from Janssen, Merck, Sanofi-Pasteur, and fees for lectures from Sanofi-Pasteur and Astra-Zeneca. O.R. has received research grants from the Bill & Melinda Gates Foundation, Merck and Janssen; and fees for participation in advisory boards from Merck, Sanofi-Pasteur, Lilly, Pfizer and Adagio; and fees for lectures from Pfizer, AstraZeneca, and Sanofi-Pasteur. None of these fees were related to the research described in this manuscript. The A.G.-S. laboratory has received research support from GSK, Pfizer, Senhwa Biosciences, Kenall Manufacturing, Blade Therapeutics, Avimex, Johnson & Johnson, Dynavax, 7Hills Pharma, Pharmamar, ImmunityBio, Accurius, Nanocomposix, Hexamer, N-fold LLC, Model Medicines, Atea Pharma, Applied Biological Laboratories and Merck, outside of the reported work. A.G.-S. has consulting agreements for the following companies involving cash and/or stock: Castlevax, Amovir, Vivaldi Biosciences, Contrafect, 7Hills Pharma, Avimex, Pagoda, Accurius, Esperovax, Farmak, Applied Biological Laboratories, Pharmamar, CureLab Oncology, CureLab Veterinary, Synairgen, Paratus and Pfizer, outside of the reported work. A.G.-S. has been an invited speaker in meeting events organized by Seqirus, Janssen, Abbott and AstraZeneca. A.G.-S. is inventor on patents and patent applications on the use of antivirals and vaccines for the treatment and prevention of virus infections and cancer, owned by the Icahn School of Medicine at Mount Sinai, New York, outside of the reported work.

Data availability

Raw counts (cell ranger outputs) and processed data (.h5ad objects) will be publicly available in Gene Expression Omnibus under: GSE206289. Fastq files will be available on dbGAP (phs002655.v1.p1) upon acceptance.

Code availability

Scripts used to process the data and generate the figures will be available here:

https://github.com/dnehar/Infants_Cov19

COVID-NET Overview and Methods
Huang C et al (2020) Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 395:497–506
Tan L et al (2020) Lymphopenia predicts disease severity of COVID-19: a descriptive and predictive study. Signal Transduct Target Ther 5:33
Kuri-Cervantes L et al (2020) Comprehensive mapping of immune perturbations associated with severe COVID-19. Sci Immunol 5:eabd7114
Hadjadj J et al (2020) Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients. Science 369:718–724
Lucas C et al (2020) Longitudinal analyses reveal immunological misfiring in severe COVID-19. Nature 584:463–469
Galani I-E et al (2021) Untuned antiviral immunity in COVID-19 revealed by temporal type I/III interferon patterns and flu comparison. Nat Immunol 22:32–40
Yoshida M et al (2022) Local and systemic responses to SARS-CoV-2 infection in children and adults. Nature 602:321–327
Pierce CA et al (2021) Natural mucosal barriers and COVID-19 in children. JCI Insight 6, e148694, 148694
Loske J et al (2022) Pre-activated antiviral innate immunity in the upper airways controls early SARS-CoV-2 infection in children. Nat Biotechnol 40:319–324
Wolock SL, Lopez R, Klein AM, Scrublet (2019) Computational Identification of Cell Doublets in Single-Cell Transcriptomic Data. Cell Syst 8:281–291e9
Polański K et al (2019) BBKNN: Fast Batch Alignment of Single Cell Transcriptomes. Bioinformatics. 10.1093/bioinformatics/btz625
Chaussabel D et al (2008) A Modular Analysis Framework for Blood Genomics Studies: Application to Systemic Lupus Erythematosus. Immunity 29:150–164
Moratto D et al (2020) Flow Cytometry Identifies Risk Factors and Dynamic Changes in Patients with COVID-19. J Clin Immunol 40:970–973
Schulte-Schrepping J et al (2020) Severe COVID-19 Is Marked by a Dysregulated Myeloid Cell Compartment. Cell 182:1419–1440e23
Liu C et al (2021) Time-resolved systems immunology reveals a late juncture linked to fatal COVID-19. Cell 184:1836–1857e22
Sinha S et al (2022) Dexamethasone modulates immature neutrophils and interferon programming in severe COVID-19. Nat Med 28:201–211
Knoll R et al (2024) The life-saving benefit of dexamethasone in severe COVID-19 is linked to a reversal of monocyte dysregulation. Cell 187:4318–4335e20
Haschka D et al (2022) Alterations of blood monocyte subset distribution and surface phenotype are linked to infection severity in COVID-19 inpatients. Eur J Immunol 52:1285–1296
Wimmers F et al (2023) Multi-omics analysis of mucosal and systemic immunity to SARS-CoV-2 after birth. Cell 186:4632–4651e23
Maher AK et al (2022) Transcriptional reprogramming from innate immune functions to a pro-thrombotic signature by monocytes in COVID-19. Nat Commun 13:7947
Arunachalam PS et al (2020) Systems biological assessment of immunity to mild versus severe COVID-19 infection in humans. Science 369:1210–1220
Bernardes JP et al (2020) Longitudinal Multi-omics Analyses Identify Responses of Megakaryocytes, Erythroid Cells, and Plasmablasts as Hallmarks of Severe COVID-19. Immunity 53:1296–1314e9
Zheng H-Y et al (2020) Elevated exhaustion levels and reduced functional diversity of T cells in peripheral blood may predict severe progression in COVID-19 patients. Cell Mol Immunol 17:541–543
Vella LA et al (2021) Deep immune profiling of MIS-C demonstrates marked but transient immune activation compared to adult and pediatric COVID-19. Sci Immunol 6:eabf7570
Maucourant C et al (2020) Natural killer cell immunotypes related to COVID-19 disease severity. Sci Immunol 5:eabd6832
Laing AG et al (2020) A dynamic COVID-19 immune signature includes associations with poor prognosis. Nat Med 26:1623–1635
Woodruff MC et al (2020) Extrafollicular B cell responses correlate with neutralizing antibodies and morbidity in COVID-19. Nat Immunol 21:1506–1516
Jenks SA et al (2020) Distinct Effector B Cells Induced by Unregulated Toll-like Receptor 7 Contribute to Pathogenic Responses in Systemic Lupus Erythematosus. Immunity 52:203
Nguyen DC et al (2022) COVID-19 and plasma cells: Is there long-lived protection? Immunol Rev 309:40–63
Guiducci C et al (2010) TLR recognition of self nucleic acids hampers glucocorticoid activity in lupus. Nature 465:937–941
Bastard P et al (2020) Autoantibodies against type I IFNs in patients with life-threatening COVID-19. Science 370:eabd4585
Korsunsky I et al (2019) Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16:1289–1296
Hobbs CV et al (2022) Frequency, Characteristics and Complications of COVID-19 in Hospitalized Infants. Pediatr Infect Dis J 41:e81–e86
Heinonen S et al (2016) Rhinovirus Detection in Symptomatic and Asymptomatic Children: Value of Host Transcriptome Analysis. Am J Respir Crit Care Med 193:772–782
Heinonen S et al (2020) Immune profiles provide insights into respiratory syncytial virus disease severity in young children. Sci Transl Med 12:eaaw0268
Guarda G et al (2011) Type I Interferon Inhibits Interleukin-1 Production and Inflammasome Activation. Immunity 34:213–223
Lee JS et al (2020) Immunophenotyping of COVID-19 and influenza highlights the role of type I interferons in development of severe COVID-19. Sci Immunol 5:eabd1554
Nehar-Belaid D et al (2020) Mapping systemic lupus erythematosus heterogeneity at the single-cell level. Nat Immunol 21:1094–1106
Caielli S et al (2023) An unconventional mechanism of IL-1β secretion that requires Type I IFN in lupus monocytes. bioRxiv 2023.08.03.551696 10.1101/2023.08.03.551696
McMahan CJ et al (1991) A novel IL-1 receptor, cloned from B cells by mammalian expression, is expressed in many cell types. EMBO J 10:2821–2832
Chou J, Thomas PG, Randolph AG (2022) Immunology of SARS-CoV-2 infection in children. Nat Immunol 23:177–185
Kaneko N et al (2020) Loss of Bcl-6-Expressing T Follicular Helper Cells and Germinal Centers in COVID-19. Cell 183:143–157e13
Wolf FA, Angerer P, Theis FJ (2018) SCANPY: large-scale single-cell gene expression data analysis. Genome Biol 19:15
Alquicira-Hernandez J, Powell JE (2021) Nebulosa recovers single-cell gene expression signals by kernel density estimation. Bioinformatics 37:2485–2487
Mertz C et al (2022) Severe Acute Respiratory Syndrome Coronavirus 2 RNAemia and Clinical Outcomes in Children With Coronavirus Disease 2019. J Infect Dis 225:208–213
Fajnzylber J et al (2020) SARS-CoV-2 viral load is associated with increased disease severity and mortality. Nat Commun 11:5493
Taveras J et al (2022) Viral Loads, Age, and Disease Severity in Young Children With Respiratory Syncytial Virus Infection. J Infect Dis 227:61–70Type III Interferons
McInnes L, Healy J, Melville JUMAP (2018) Uniform Manifold Approximation and Projection for Dimension Reduction. arXiv:1802.03426 [cs, stat]
Chiche L et al (2014) Modular Transcriptional Repertoire Analyses of Adults With Systemic Lupus Erythematosus Reveal Distinct Type I and Type II Interferon Signatures: Modular Interferon Signatures and Systemic Lupus Erythematosus. Arthritis Rheumatol 66:1583–1595

Yes there is potential Competing Interest. At the time of the study, J.F.B. was a member of the BOD and SAB of Neovacs and Ascend Biopharma, and a SAB member of Cue Biopharma. J.F.B is now an employee of Immunai. A.M. has received research grants from Janssen and Merck, fees for participation in advisory boards from Janssen, Merck, Sanofi-Pasteur, and fees for lectures from Sanofi-Pasteur and Astra-Zeneca. O.R. has received research grants from the Bill & Melinda Gates Foundation, Merck and Janssen; and fees for participation in advisory boards from Merck, Sanofi-Pasteur, Lilly, Pfizer and Adagio; and fees for lectures from Pfizer, AstraZeneca, and Sanofi-Pasteur. None of these fees were related to the research described in this manuscript. The A.G.-S. laboratory has received research support from GSK, Pfizer, Senhwa Biosciences, Kenall Manufacturing, Blade Therapeutics, Avimex, Johnson & Johnson, Dynavax, 7Hills Pharma, Pharmamar, ImmunityBio, Accurius, Nanocomposix, Hexamer, N-fold LLC, Model Medicines, Atea Pharma, Applied Biological Laboratories and Merck, outside of the reported work. A.G.-S. has consulting agreements for the following companies involving cash and/or stock: Castlevax, Amovir, Vivaldi Biosciences, Contrafect, 7Hills Pharma, Avimex, Pagoda, Accurius, Esperovax, Farmak, Applied Biological Laboratories, Pharmamar, CureLab Oncology, CureLab Veterinary, Synairgen, Paratus and Pfizer, outside of the reported work. A.G.-S. has been an invited speaker in meeting events organized by Seqirus, Janssen, Abbott and AstraZeneca. A.G.-S. is inventor on patents and patent applications on the use of antivirals and vaccines for the treatment and prevention of virus infections and cancer, owned by the Icahn School of Medicine at Mount Sinai, New York, outside of the reported work.

ExtendeddataFigurespCov4009042024.pdf
Expansion of ISG IL-1B CD14 monocytes in moderate to severe disease
ST1Clinicaldata.xlsx
Supplementary Table 1. Characteristics of enrolled healthy controls and COVID-19 patients’ information.
ST2SequencingInfopCov40.xlsx
Supplementary Table 2. Sequencing information for each sample
ST3genelists.xlsx
Supplementary Table 3. Gene lists used for cell scoring
ST4Top100ClusterspCov40.xlsx
Supplementary Table 4. Differentially expressed genes defining each immune cluster – pediatric cohort (related to Figure 1c and Extended Data Figure.1).
ST5Top100DESCspCov40.xlsx
Supplementary Table 5. Differentially expressed genes defining each immune subcluster (SCs) - pediatric cohort (related to Figure 2 to Figure 5 and Extended Data Figure.2-3).

Download PDF

Version 1

posted

You are reading this latest preprint version

Immune perturbations induced by SARS-CoV2 in infants vary with disease severity and differ from adults’ responses

Status:

Version 1

Abstract

Figures

Introduction

Results

Clinical categorization of infected infants and changes in PBMC compositions

Expansion of ISG IL-1B monocyte In mild to severe disease

Contraction of cDCs in SARS-CoV-2 infected infants

ISG naïve CD4 T cells are expanded in SARS-CoV-2 infected infants

ISG CD8 T cells are expanded in infants with COVID-19

ISGhi NK cells during severe SARS-CoV-2 infection in infants

SARS-CoV-2 infection in infants is associated with increased ISG transitional and naïve B cells

Anti-SARS-CoV-2 and anti-IFN antibody responses

Lymphocytes from infants express a broader ISG signature compared to adults

Discussion

Limitations of the study when performing

Methods

Study design

PCR assays for SARS-CoV-2 viral loads

Cytokine assays

Enzyme-Linked Immunosorbent Assay (ELISA)

Functional evaluation of anti-IFN-I auto-antibodies

Blood preparation for single cell RNA sequencing (scRNA-seq)

Single-cell Raw data processing and data combining

Scrublet for multiplet prediction and removal

Single-cell preprocessing, dimension reduction, graph-based clustering, and cluster annotation

Quality control and cell filtering

Data normalization, log transformation and scaling

Linear dimensional reduction using PCA

Neighborhood graph computing, embedding and clustering

Batch effect correction

IFN score calculation

Finding marker genes/evaluation of cluster identity

Subclustering parameters and data cleaning.

Adult COVID-19 data Integration (paCoV dataset)

Comparison with Wimmers et al. cohort20.

Statistical analysis

Python module versions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

ISG^hi NK cells during severe SARS-CoV-2 infection in infants

Comparison with Wimmers et al. cohort²⁰.