key: cord-0800604-9ht3b1ce
authors: Lai, Yang-dian; Chen, Ying-ying; Sun, Ji-ping; Ling, Yun; Xu, Jie; Ye, Youqiong; Shen, Lei; Lu, Hong-zhou; Su, Bing; Wang, Ying
title: Immune profiles of a COVID-19 adolescent with mild symptoms and anti-viral antibody deficiency
date: 2021-02-17
journal: nan
DOI: 10.1016/j.fmre.2021.02.004
sha: d810ecc4562f390c44613be900f1a48de04a0950
doc_id: 800604
cord_uid: 9ht3b1ce

nan

Severe acute respiratory syndrome corona virus 2(SARS-CoV-2) infection has led to the outbreak of COVID-19 syndrome since December 2019, becoming prevalent worldwide afterward [1, 2] .

Unlike severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS), which cause severe symptoms and high mortality[1],morbidity from COVID-19 is relatively low, especially in young people [3] . In China, the mortality rate was 1023 of 44,672 confirmed cases by the end of February11, 2020 [4] . However, there were no cases of death among 416 patients aged 0-9 years, and only 1 in 549 patients aged 10-19 years [4] . The exact reasons for this clinical phenomenon remain unclear.

It is widely accepted that viral infection triggers innate and adaptive immunity sequentially [5] .

While innate immune systems such as monocytes, dendritic cells, and NK cells constitute the first line of the protection against the invading virus, viral-specific antibody and T cell responses are more efficient in eliminating virus and virus-infected host cells, combined with the induction of memory responses for long-term protection. It is well documented that antibodies with neutralizing activity targeting SARS-CoV-2 spike (S) protein exist in convalescent COVID-19 patients [6] [7] [8] , which could be helpful for resistance to viral re-infection in the long run. These antibodies can last for at least 6 months in the human body [9] . Cellular immune responses are also reported in recovered COVID-19 patients targeting viral spike (S), nucleocapsid (N), and membrane (M) proteins [10] . Herein, we report a 15-year-old COVID-19 patient with mild symptoms and a deficiency in anti-viral protein antibody responses at the rehabilitation stage. We defined immunological profiles in peripheral blood mononuclear cells using single-cell RNA sequencing for a better understanding of potential protective mechanisms against SARS-CoV-2 infection.

The COVID-19 adolescent was an inpatient at the Shanghai Public Health Clinical Center (Shanghai, China). Clinical manifestations were recorded at several time points. These included body temperature, treatments, total white blood cell count, neutrophil count, lymphocyte count, monocyte count, platelet count, and C-reactive protein level. Computed tomography (CT) images were taken at different time points as needed. Reverse transcription polymerase chain reaction (RT-PCR) results for viral RNA from oropharyngeal swabs, feces, and urine were collected retrospectively. This study was approved by the Shanghai Ethics Committee for Clinical Research (SECCR/2020-04-01). Written informed consent was obtained from the parents of the included child.

A magnetic bead-based nucleic acid extraction kit was used in a fully automated nucleic acid extraction instrument (Master Biotechnology, China). Total RNA was extracted from 200 μL samples, and dual fluorescence PCR was performed according to the manufacturer's instructions (Zhijiang Co., Shanghai, China) using Applied Biosystems 7500 Real-Time PCR System (Foster City, CA, USA). A Ct value < 37 was defined as a positive result, and Ct > 40 was defined as a negative result.

The FIA assay was performed using detection cards coated with fluorescence-labeled S and N protein (Sino Biological, Beijing, China) for IgG and IgM detection according to the manufacturer's instructions (Dialab ZJG Biotech Co, Suzhou, China) [11] . Briefly, 10 μL plasma was mixed with 990 μL dilution buffer. 80 μL diluted plasma was added to the sampling well of the detection cards. The fluorescence signal was captured by a DL300 Quantitative 

Whole blood was collected in a tube containing ethylene diamine tetraacetic acid (EDTA). 

Antigen-specific IFN-γ release was detected using an enzyme-linked immunospot (ELISpot) assay according to the manufacturer's instructions (U-CyTech, Utrecht, Netherlands). Briefly, 96-well PVDF plates (Millipore) were coated with anti-human IFN-γ coating antibody overnight at 4°C. PBMCs (0.25×10 6 ) were added to each well and stimulated with the recombinant receptor-binding domain of S protein (S-RBD), N (nucleocapsid protein), envelope protein (E) (Novoprotein, Shanghai, China) (both at 20g/mL), or tuberculin purified protein derivative (PPD) (20g/mL) (Statens Serum Institut, SSI, Copenhagen, Denmark) for 20 h at 37°C. Untreated RPMI 1640 culture medium served as a negative control, and treatment with 2.5 μg/mL phytohemagglutinin (PHA) (Sigma-Aldrich) was used as a positive control. After incubation for 20 h at 37°C, the plates were incubated with biotin-labeled detection antibody at 37°C for 1 h, and subsequently HRP-conjugated streptavidin working solution for an additional 1 h. The AEC substrate solution was added to each well for 30 min in the dark at room temperature (RT). Color development was stopped by thoroughly rinsing both sides of the PVDF membrane with demineralized water. Plates were dried in the dark at RT. The spots were counted using a C.T.L.

ImmunoSpot® S6 Ultra Analyzer (Cellular Technology Limited, OH, USA). The number of antigen-specific IFN-γ-producing cells was calculated based on the number of spot-forming units (SFUs) per 2.5 × 10 5 PBMCs after deducting the background SFUs of the paired negative control wells.

Isolated PBMCs were diluted as a single-cell suspension with cell viability exceeding 90%.Sequencing libraries of PBMCs were prepared following the Drop-seq methodology.

Single-cell RNA (sc-RNA) libraries were prepared according to the guidelines of the Chromium SingleCell 5'v3 chemistry (10x Genomics, Cat No. 120237). Libraries were sequenced using an Illumina NovaSeqS6000 device (Genergy Biotechnology, Shanghai, China).

scRNA sequencing reads were subjected to quality control based on FastQC software v0.11.9

(https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Cell Ranger software (version 3.1.0) was used to process, align, and summarize sequencing data. De-multiplexed FASTQ reads were aligned to the human reference genome (GRCh38-3.0.0). Reads withlow base-calling quality scores and assigned cell barcodes were filtered out. Unique molecular identifier (UMI) counts of each transcript were quantified. The UMI matrix was converted to Seurat objects using the R package Seurat v3. Further analysis was performed using the R package. Raw UMI count matrices were filtered to remove genes expressed in fewer than three cells, cells with either fewer than 200 or more than 6000 genes, or high percentages of mitochondrial genes (more than 15%).

A total of 9340 cells were used for further analysis. To account for differences in sequencing depth across cells, UMI count was normalized by a global-scaling method, converted with a scaling factor (10,000 by default), and log-transformed with the LogNormalize function in Seurat for downstream analysis.

FindVariableFeaturesfunction in R package Seurat v3was used to scale data with the top 2000 most variable genes. Principal component analysis (PCA) was subsequently performed based on these variable genes. FindNeighbors in Seurat was used to obtain nearest neighbors for graph clustering based on PCs, and FindCluster in Seurat to divide cell subsets. Cells were divided into myeloid, NK/T, and B cells based on their type-specific gene signatures, then visualized with the uniform manifold approximation and projection (UMAP) algorithm. Signature scores were calculated as the mean log 2 (LogNormalizedUMI+1) across related genes in the signatures. Each cluster was assigned to the compartment of its maximal score, and all cluster assignments were manually checked to ensure the accurate partition of cells.

To identify differentially expressed genes (DEGs) across different immune cell types, we used the FindAllMarkers function in Seurat based on normalized data. Non-parametric Wilcoxon rank-sum tests were used to obtain p-values for comparisons. P-value adjustment was performed using Bonferroni correction based on the total number of genes in the dataset. DEGs with adjusted p-values < 0.05 and an average expression log 2 -fold-change > 0.25 were kept for further analysis.

Freshly isolated PBMCs were washed with phosphate-buffered saline(PBS) containing 2% FBS. A 15- year-oldmale adolescent developed symptoms on February 4, 2020, including fever with a body temperature of 37.7°C the next day, accompanied by nasal congestion, runny nose, cough, and white sputum. However, no other typical COVID-19 symptoms (headache, dizziness, chest tightness, chest pain, hemoptysis, nausea, vomiting, abdominal pain, diarrhea, or bloody stool) were observed. The patient was diagnosed at Huashan Hospital (Shanghai, China). In terms of epidemiological history, the patient denied travel to Hubei province (including Wuhan), contact with the Huanan seafood market, and a history of contact with wild animals. However, he had contacted a confirmed COVID-19 patient displaying fever, cough, and other symptoms. Nucleic acid detection of SARS-CoV-2 using oropharyngeal swabs was negative on February5 and positive after the second RT-PCR detection on February 6 ( Figure 1A) . The patient was then admitted to the Shanghai Public Health Clinical Center (SPHCC) for further treatment. A typical pneumonia lesion was observed in the dorsal segment of the right lower lobe and the outer basal segment of the right lung on the same day based on chest CT scans ( Figure 1B) . Blood C-reaction protein (CRP) was 11.4 mg/L (normal range: 0-5 mg/L). The adolescent was diagnosed with COVID-19. He had received recombinant human Interferon-α2b spray at admission and LianhuaQingwen granules on February 10. Vitamin C tablets were administered daily. No antibiotics, steroids, or antiviral agents were administered during hospitalization.

SARS-CoV-2 detection was performed using oropharyngeal swabbing, as well as stool and urine samples after admission. However, all detection results were negative ( Figure 1A) . On February 10, the inflammation lesion in the lung began to be absorbed. With negligible lesion in the right lower lobe (Figure 1B) , normal CRP value (Figure 1C) , and two successive negative results for SARS-CoV-2 virus (with a 24 h sampling interval), he was discharged on February 15. During hospitalization, the counts of peripheral white blood cells (5.24-6.63×10 9 cells/L (range: 3.5-9.5 × 10 9 /L)), neutrophils (2.49-3.59 × 10 9 /L (range: 1.8-6.3 × 10 9 /L)), lymphocytes (2.29-3.17 × 10 9 /L (range: 1.10-3.20 × 10 9 /L)), monocytes (0.35-0.57×10 9 cells/L (range: 0.10-0.6 ×10 9 /L)), and platelets (230-286 ×10 9 /L (range: 125-350 × 10 9 /L)) were within normal ranges (Extended Data Figure 1 ).

Since anti-S antibody is most commonly detected after recovery and likely to play critical roles in neutralizingSARS-CoV-2 virus [6, 7] ,we performed anti-S antibody detection in the serum using an FIA assay at the first follow-up visit. Surprisingly, both anti-S IgG and IgM were below the limit of detection (LOD) of the assay when compared to samples from convalescent COVID-19adults (Figure 1D and 1E) . Anti-S IgG and IgM were also absent at the second follow-up visit (Day 53). We traced both back to the day of admission, and neither anti-S IgG and IgM were detected.

A scRNA-seq was further performed using freshly isolated PBMCs from the recovered COVID-19 adolescent to dissect the immune cell distribution with gene signatures. Three lineages, including myeloid, NK/T, and B cells, were clustered in 9340 PBMCs using t-distributed stochastic neighbor embedding (t-SNE) analysis (Extended Figure 2A and 2B) according to gene signatures (Extended Figure 2C and 2D) . B cells were defined by signatures of CD19 andCD20 (MS4A1), and were 10.7% of the total ( Figure 1F ). Based onimmunoglobulin heavy chain (IGH) gene expression in B cells, it was found that the percentages of IGHD-and IGHM-positive B cells were the most frequent, whereas IGHA1-, IGHG1, IGHG2-, IGHG3-, and IGHG4-expressing B cells were very few, with a proportion of 6.35% for IGHG (including IGHG1, IGHG2, IGHG3 and IGHG4), and 4.99% for IGHA expression in total B cells ( Figure   1G ).

Nevertheless, virus-specific cellular responses were apparent at the first follow-up visit. IFN-γ release in PBMCs upon in vitro stimulation of SARS-CoV-2 S-RBD, N, and E proteins separately was obvious, together with the positivity of PPD-specific cellular response ( Figure 1H) . These data indicate that anti-SARS-CoV-2 humoral responses are absent, whereas SARS-CoV-2-specific T cell responses are still detectable in this COVID-19 recovered adolescent.

Based on scRNA-seqdata, NK/T cell lineages (75.7% in total PBMCs) were further sub-grouped into CD4 + T cells (34.1%, co-expressing CD3E and CD4 genes), CD8 + T cells (21.9%, co-expressing CD3E and CD8 genes), and NK cells (44.4%, co-expressing NCAM1 and FCGR3A genes) (Figure 2A and 2B) . The proportion of NK cells in the NK/T cell subsets was extremely high (Figure 2A) , which was verified by flow cytometry (Extended Figure 3) . The percentage of NK cells in total lymphocytes was 29.9%, which was higher than that of CD4 + T cells (20.9%) or CD8 + T cells (20.45%). UMAP and violin plot analysis indicated that IFNG, TNF, and GZMB were highly expressed in NK cells and CD8 + T cells (Figure 2C and 2D) .We also determined cytokine secretion in CD8 + T cells and NK cells by flow cytometry. Upon PMA/ionomycin stimulation in vitro, it was obvious that CD8 + T cells produced robust IFN-γ (23.1%), TNF-α (14.1%), and GzmB (38.7%). The percentages of IFN-γ-, TNF-α-, and GzmB-positive NK cells were 36.6%, 18.1%, and 79.1%, respectively, higher than corresponding values for CD8 + T cells ( Figure 2F and 2G) . Furthermore, the UMAP plot showed that a portion of NK cells highly expressed KLRC2 (Figure 2H) , an important memory-like marker [12] , suggesting the existence of memory-like NK cells in the periphery of the recovered patient. Our results thus indicate that there exist functional CD8 + T cells and NK cells in the periphery of the recovered adolescent, with greater expansion of NK cells.

Myeloid cells, including monocytes and dendritic cells, are crucial for virus clearance at the early stage and engaged in the antigen-presenting process to initiate adaptive immunity. According to scRNA-seq analysis, myeloid cells were divided into five populations, including classical CD14 ++ CD16monocytes (CD14-high and FCGR3A-negative), intermediate CD14 ++ CD16 + monocytes (CD14-high and FCGR3A-positive), non-classical CD14 + CD16 ++ monocytes (CD14and FCGR3A-positive), CD11C + myeloid dendritic cells (mDCs) (expressing CD11C), and CD123 + plasmacytoid dendritic cells (pDCs) (expressing IL3RA) (Figure 3A and 3B) . The UMAP plot showed that TNF was expressed in all cell subsets, while IL-1β was highly expressed in monocytes and mDCs. In addition, IFNG was mainly expressed in CD14 ++ CD16 + monocytes.

However, IL-6 and CSF2 were rarely expressed in all subsets. Type I interferon (IFNI) plays an important role in inhibiting viral replication in both infected and non-infected cells. Although IFNA expression was not detectable, some monocytes and DCs expressed IFNAR1 and IFNAR2, two genes encoding IFN-α receptors (Figure 3C and 3D) . Upon LPS stimulation, CD14 hi monocytes expressed high levels of IL-1β, TNF-α, and IL-6, whereas CD14 medium monocytes produced fewer cytokines (Figure 3E and 3F) . In particular, CD14 hi monocytes co-expressed IL-1β andIL-6 more apparently, whileIL-1β and TNFα were co-expressed to a less extent ( Figure   3G and 3H). Our results thus illustrate that peripheral CD14 ++ monocytes secrete high levels of pro-inflammatory cytokines in the recovered COVID-19 adolescent and may function in a more extensive manner.

The immune system plays critical roles in defending against SARS-CoV-2 virus infection by inducing innate immunity and antigen-specific T cell and antibody responses. Occasionally, a cytokine storm is induced in certain cases and causes acute respiratory distress syndrome (ARDS), leading to multiple organ failure and mortality [13] . However, the vast majority of infected children do not develop ARDS and do not require respiratory support or intensive care [3] . Overall, SARS-CoV, MERS-CoV, and SARS-CoV-2 infections have less fatal effects on children than adults, with mild clinical symptoms, low mortality, and a good prognosis [14] [15] [16] [17] .

In the present study, we have reported a COVID-19adolescent with no humoral responses after recovery. He typically had mild symptoms, such as a transient increase in body temperature and CRP, no respiratory failure or acute respiratory distress syndrome, and a short period of hospitalization. The unexpected observation is that anti-S protein as well as anti-N protein (Extended Figure 4 ) IgG and IgM were absent from admission to discharge, implying that the adolescent exhibited a deficiency in humoral immune response upon SARS-CoV-2 infection.

Nevertheless, based on the ELISpot assay, an antigen-specific cellular immune response is still detectable in the recovered COVID-19 child, implying that COVID-19 adolescents displays unique immune profiles different from those of adults. Interestingly, results from the ELISpot assay targeting S, N, and E proteins indicated that the immunoreactivity to E protein was the strongest ( Figure 1H ). Due to their location and biological function, S and N proteins have been extensively investigated for their humoral and cellular immunogenicity [18] [19] [20] , as well as for their applications in diagnosis and antibody development [21, 22] . E protein is a structural protein that exists in both monomeric and homo-pentameric forms. It is reported to be involved in the onset of viral infection, replication, and dissemination within host cells [23, 24] . Its immunogenic peptides have been computationally predicted [25, 26] , but not clinically validated. Our data thus provide direct evidence of its capacity to induce cellular immune response. The higher cellular immunoreactivity to E protein we observed in our case also implies the possibility that viral proteins display distinct immunogenicity related to their protein properties. The humoral responses targeting S protein prevent interaction with its receptor (ACE2), blocking entry of the virus into host cells, while cellular immune responses targeting other viral proteins are more inclined to eliminate virus-infected cells. Our data present the possibility that E protein exerts long-term protection against SARS-CoV-2 re-infection.

In addition to apparent T cell responses, the most striking immunological properties in this case are the exaggerations of innate immunity even at the rehabilitation stage. Several studies on scRNA-seq analysis of adult COVID-19 patients have been reported [27] [28] [29] , and in both the acute and convalescent stages, innate lymphocyte signatures are well defined, including cytokine and chemokine enrichment. In our case, NK cells accounted for nearly 30% of total lymphocytes with CD56 dim CD16 bright NK cells as the main group with high cytolytic ability. In the myeloid proportion, 90% of monocytes are CD14 ++ CD16classical monocytes. The remaining monocytes consist of CD14 ++ CD16 + intermediate monocytes and CD14 + CD16 ++ non-classical monocytes [30] .

Intermediate monocytes play an important role in viral infections, such as dengue fever [30] .

Interestingly, the proportion of intermediate monocytes and non-classical monocytes increased in the COVID-19 child. It was previously reported that the "loss" of intermediate monocytes and non-classical monocytes is the result of down-regulation after stimulation [31] , which is consistent with our results. Moreover, peripheral monocytes are able to secrete inflammatory cytokines, including IL-1β, TNF-α, and IL-6, even after recovery, among which CD14 ++ monocytes co-expressed multiple cytokines. Therefore, we deduce that in the process of SARS-CoV-2 infection, innate immune responses are likely to play important roles when humoral immune responses are attenuated.

In conclusion, the case we report here indicates that although antibody response is absent after viral infection, CD8 + T cells and innate immune cells, including NK cells and monocytes, exhibit functional properties with cytokine secretion. The induction of strong cellular immunity also implies the possibility of rapid virus clearance after infection, and likely contributes to the mild symptoms and good prognosis of younger COVID-19 patients.

The funders had no role in study design, data collection, analysis, decision to publish or preparation of the manuscript. All the authors declare no conflict of interest.

Immune responses in COVID-19 and potential vaccines: Lessons learned from SARS and MERS epidemic

Identification of a novel coronavirus causing severe pneumonia in human: a descriptive study

Characteristics of pediatric SARS-CoV-2 infection and potential evidence for persistent fecal viral shedding

The epidemiological characteristics of an outbreak of 2019 novel coronavirus diseases (COVID-19) in China

Innate and adaptive immune responses to viral infection and vaccination

Potent binding of 2019 novel coronavirus spike protein by a SARS coronavirus-specific human monoclonal antibody

COVID-19, an emerging coronavirus infection: advances and prospects in designing and developing vaccines, immunotherapeutics, and therapeutics

Sensitivity and specificity of SARS-CoV-2 S1 subunit in COVID-19 serology assays

Humoral Immune Response to SARS-CoV-2 in Iceland

Detection of SARS-CoV-2-Specific Humoral and Cellular Immunity in COVID-19 Convalescent Individuals

Dynamic anti-spike protein antibody profiles in COVID-19 patients

Epigenetic Modification and Antibody-Dependent Expansion of Memory-like NK Cells in Human Cytomegalovirus-Infected Individuals

COVID-19 infection: the perspectives on immune responses

Coronavirus Infections in Children Including COVID-19: An Overview of the Epidemiology, Clinical Features, Diagnosis, Treatment and Prevention Options in Children

Clinical characteristics of novel coronavirus disease 2019 (COVID-19) in newborns, infants and children

Middle East respiratory syndrome coronavirus disease in children

儿童新型冠状病毒感染诊断、治疗和预防专家共识(第二版)

A neutralizing human antibody binds to the N-terminal domain of the Spike protein of SARS-CoV-2

Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals

SARS-CoV-2-specific T cell immunity in cases of COVID-19 and SARS, and uninfected controls

Detection of COVID-19: A review of the current literature and future perspectives

COVID-19 diagnosis -A review of current methods

The OC43 human coronavirus envelope protein is critical for infectious virus production and propagation in neuronal cells and is a determinant of neurovirulence and CNS pathology

Subcellular location and topology of severe acute respiratory syndrome coronavirus envelope protein

Computational Immune Proteomics Approach to Target COVID-19

Design of a Multiepitope-Based Peptide Vaccine against the E Protein of Human COVID-19: An Immunoinformatics Approach

Immune cell profiling of COVID-19 patients in the recovery stage by single-cell sequencing

Single-cell landscape of immunological responses in patients with COVID-19

Single-cell analysis of two severe COVID-19 patients reveals a monocyte-associated and tocilizumab-responding cytokine storm

Pattern of human monocyte subpopulations in health and disease

ADAM17-Mediated Reduction in CD14(++)CD16(+) Monocytes ex vivo and Reduction in Intermediate Monocytes With Immune Paresis in Acute Pancreatitis and Acute Alcoholic Hepatitis

adolescent patientduring the disease process.(A) Timeline of the COVID-19 adolescent patient during the disease process

Red arrows indicate the inflammation lesion in the lung. (C) Dynamics of blood C-reactive protein (CRP) levels from admission to the follow-up visit. (D-E) Dynamics of SARS-CoV-2 spike (S) protein-specific IgG (D) and IgM (E) in the plasma of the adolescent patient. Limit of detection (LOD): anti-S IgG 15RU/mL, anti-S IgM 3.4 RU/mL. Three adult convalescent COVID-19 patients (blue triangles) and three healthy controls (black diamonds) were used as controls. (F) UAMP plot of B cell gene signatures, including CD19, CD27, MS4A1, CD38, IGHD, IGHM, IGHA1, IGHG1, IGHG2, IGHG3, IGHG4, and IGHE. (G) Distribution percentages of each isotype of immunoglobulin. (H) Antigen-specific IFN-γ releasing cells in PBMCs were determined by an ELISpot assay upon stimulation with tuberculin pure protein derivative (PPD), SARS-CoV-2 spike protein fragment (S-RBD)

A) UMAP plot showed three clusters, including NK cells, CD4 + and CD8 + T cells, according to the expression of marker genes. Cells were color-coded according to gene signatures. (B) The features of UMAP plot define NK/T cell types with specific genes, including CD3E, CD4, CD8A, NCAM1, and FCGR3A. (C-D) A UMAP plot (C) and violin plot (D) show the expression of cytokines across NK/T cell subtypes, including IFNG, TNF, GZMB, CSF2, IL2, and IL17RA. (E-G) Cytokine expression upon 12-myristate 13-acetate (PMA) and ionomycin stimulation in CD3 + CD8 -cells (E)

Figure 3. Characterization of myeloid cell subsets in the periphery of the COVID-19 child

Cells were color-coded according to defined myeloid subtypes. (B) The feature plot of UMAP shows subtype-specific gene expression for myeloid cells, including CD14

This work was supported by the Chinese National Mega Science and Technology Program on