Blood transcriptomes of anti-SARS-CoV2 antibody positive healthy individuals with prior asymptomatic versus clinical infection

TL;DR: In this paper, the authors performed genome-wide transcriptional whole-blood profiling to test the hypothesis that immune response-related gene signatures may differ between healthy individuals with prior entirely asymptomatic versus clinical SARS-CoV-2 infection, all of which developed an equally robust antibody response.

Abstract: Despite tremendous efforts by the international research community to understand the pathophysiology of SARS-CoV-2 infection, the reasons behind the clinical variability, ranging from asymptomatic infection to lethal disease, are still unclear. Existing inter-individual variations of the immune responses, due to environmental exposures and genetic factors, may be critical to the development or not of symptomatic disease after infection with SARS-CoV-2, and transcriptomic differences marking such responses may be observed even later, after convalescence. Herein, we performed genome-wide transcriptional whole-blood profiling to test the hypothesis that immune response-related gene signatures may differ between healthy individuals with prior entirely asymptomatic versus clinical SARS-CoV-2 infection, all of which developed an equally robust antibody response. Among 12.789 protein-coding genes analyzed, there were only six and nine genes with significantly decreased or increased expression, respectively, in those with prior asymptomatic infection (n=17, mean age 34 years) relatively to those with clinical infection (n=15, mean age 37 years). All six genes with decreased expression (IFIT3, IFI44L, RSAD2, FOLR3, PI3, ALOX15), are involved in innate immune response while the first two are interferon-induced proteins. Among genes with increased expression six are involved in immune response (GZMH, CLEC1B, CLEC12A), viral mRNA translation (GCAT), energy metabolism (CACNA2D2) and oxidative stress response (ENC1). Notably, 8/15 differentially expressed genes are regulated by interferons. Our results suggest that an intrinsically weaker expression of some innate immunity- related genes may be associated with an asymptomatic disease course in SARS-CoV-2 infection. Whether a certain gene signature predicts, or not, those who will develop a more efficient immune response upon exposure to SARS-CoV-2, with implications for prioritization for vaccination, warrant further study.

Summary

Introduction

• Since December 2019 the SARS-CoV-2 has spread throughout the world infecting dozens of millions of people and resulting in over 2.8 million deaths, as of April 2021.
• Notably, the authors only know in retrospect who was indeed asymptomatic, since individuals without symptoms at the time of a positive molecular test should be followed for 14 days to determine the clinical picture, being "pre-symptomatic” if they develop symptoms later.
• The proportion of asymptomatic individuals varies widely in viral infections.
• Most of this inter-individual immune variation is explained by environmental exposures early in life [16] but genetic factors are clearly also involved.
• Since variations in the strength and/or extent of the immune response may be critical for the clinical picture and progress after infection with SARS-CoV-2, existing inter-individual differences at the transcriptome level may be observed even later, after convalescence.

Methods

• Blood collection and anti-SARS-CoV-2 antibody testing Blood samples were collected from members of the NKUA, Athens, Greece in June–November 2020.
• The protocol was approved by the Ethics and Bioethics Committee of the School of Medicine, NKUA (protocol #312/02-06-2020) and study participants provided written informed consent.

3’ mRNA sequencing, mapping, quality control, and quantifications

• Total RNA was isolated from whole blood, stored in paxgene, using the ExtractionMonarch® Total RNA Miniprep Kit (NEB #T2010).
• Addition of the protection reagent and the following RNA isolation was performed as described in the Kit's manual for Total RNA Purification from Mammalian Whole Blood Samples.
• The gene counts table was normalized for inherent systematic or experimental biases (e.g., sequencing depth, gene length, GC content bias) using the Bioconductor package EDASeq [25].
• The Mann-Whitney U test was applied in order to calculate the significance of the difference in distributions between the asymptomatic and clinical groups.

Differential gene expression

• The resulting gene counts table was subjected to differential expression analysis (DEA) to compare individuals with a history of asymptomatic versus clinical (“symptomatic”) infection using the Bioconductor packages DESeq [27], edgeR [28], NOISeq[29], limma [30], NBPSeq [31], baySeq [32].
• Multidimensional scaling was also applied through metaseqR2.
• DAVID analysis [33] was performed for the increased and decreased genes, both for enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and for biological processes [Gene Ontology (GO)].
• For the prediction of enriched regulons in asymptomatic disease the authors used the TRRUST (v2) reference transcription factor (TF)–target interaction database [34] and enrichR [35] focusing on the ChEA prediction with the increased genes in asymptomatic disease as input.
• For the identification of interferon-regulated genes the inteferome database (v2) [36] was used.

Results

• Whole blood transcriptional profiling and determination of immune cell subsets in seropositive asymptomatic versus clinical infection Clinical infections were in their majority of low to moderate severity.
• A multidimensional scaling (MDS) plot generated using all 16.737 expressed genes, in order to avoid gene-type biases, revealed no clear separation of the two sample groups .
• The values for all samples (17 asymptomatic on the left and 15 clinical on the right) is plotted.
• The genes characterized as differentially expressed in those with prior asymptomatic infection relatively to those with clinical SARS-CoV-2 infection were queried in the Interferome database;.

Discussion

• Genome-wide transcriptome analyses studies using next generation sequencing technology in patients infected with SARS-CoV-2 provide evidence that transcriptome-wide changes may serve as predictors of morbidity and possibly of response to specific therapies [48].
• Variations in innate immune system responses and cytokine networks could explain, at least in part, the wide heterogeneity in clinical presentation of SARS-CoV-2 infection [51].
• Decomposition of many type I interferon genes [55] and partial loss of function in stimulator of interferon genes is observed in bats [56].
• It should be highlighted that the transcriptome analysis was not performed at the time of active infection; thus certain potential differential responses may have been blunted during assessment after infection.
• In contrast to these findings, increased levels of interferons and interferon-stimulated genes have been observed in severe and life-threatening infections in many other studies [63, 64, 65].

Figures

Figure 3. Venn diagram of differentially expressed genes in seropositive individuals with prior asymptomatic or clinical SARS-CoV-2 infection with respect to interferon activity.

Table 1: Demographics and antibody measurements.

Figure 2. Differential gene expression analysis in seropositive individuals with prior asymptomatic or clinical SARS-CoV-2 infection.

Full text

Blood transcriptomes of anti-SARS-CoV2 antibody
positive healthy individuals with prior
asymptomatic versus clinical infection
Petros P. Sfikakis
1,2*
, Kleio-Maria Verrou
1,2
, Ourania Tsitsilonis
3
, Dimitrios Paraskevis
4
, Efstathios
Kastritis
5
, Evi Lianidou
6
, Paraskevi Moutsatsou
7
, Evangelos Terpos
5
, Ioannis Trougakos
8
, Vasiliki
Chini
1
, Menelaos Manoloukos
1
, Panagiotis Moulos
1,9
, Georgios A. Pavlopoulos
1,9
, George
Kollias
1,2,10
, Giannis Ampatziadis-Michailidis
1
, Pantelis Hatzis
1,9
, Meletios A Dimopoulos
1,5
1. Center of New Biotechnologies & Precision Medicine, National and Kapodistrian University of Athens
Medical School, Athens, Greece
2. Joint Rheumatology Program, National and Kapodistrian University of Athens Medical School, Athens,
Greece
3. Department of Biology, National and Kapodistrian University of Athens (NKUA), 15784 Athens, Greece
4. Department of Hygiene, Epidemiology and Medical Statistics, Medical School, National and Kapodistrian
University of Athens, Athens, Greece
5. Department of Clinical Therapeutics, School of Medicine, National and Kapodistrian University of
Athens, Athens, Greece
6. Department of Chemistry, NKUA, Athens 15771, Greece
7. Department of Clinical Biochemistry, School of Medicine, University General Hospital Attikon, NKUA,
12462 Haidari, Greece
8. Department of Cell Biology and Biophysics, Faculty of Biology, National and Kapodistrian University of
Athens, Athens, 15784, Greece
9. Institute for Fundamental Biomedical Research, BSRC Alexander Fleming, Vari, Greece
10. Institute for Bioinnovation, BSRC Alexander Fleming, Vari, Greece
* Corresponding Author: Petros P. Sfikakis, School of Medicine, National and Kapodistrian University of
Athens, 17 Agiou Thoma Street, 11527 Athens, Greece. E-mail: psfikakis@med.uoa.gr
Short title: Blood transcriptomes in prior asymptomatic SARS-CoV2!infection !
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprintthis version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.19.21255748doi: medRxiv preprint
NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Abstract
Despite tremendous efforts by the international research community to understand the
pathophysiology of SARS-CoV-2 infection, the reasons behind the clinical variability, ranging from
asymptomatic infection to lethal disease, are still unclear. Existing inter-individual variations of the
immune responses, due to environmental exposures and genetic factors, may be critical to the
development or not of symptomatic disease after infection with SARS-CoV-2, and transcriptomic
differences marking such responses may be observed even later, after convalescence. Herein, we
performed genome-wide transcriptional whole-blood profiling to test the hypothesis that immune
response-related gene signatures may differ between healthy individuals with prior entirely
asymptomatic versus clinical SARS-CoV-2 infection, all of which developed an equally robust
antibody response. Among 12.789 protein-coding genes analyzed, there were only six and nine
genes with significantly decreased or increased expression, respectively, in those with prior
asymptomatic infection (n=17, mean age 34 years) relatively to those with clinical infection (n=15,
mean age 37 years). All six genes with decreased expression (IFIT3, IFI44L, RSAD2, FOLR3, PI3,
ALOX15), are involved in innate immune response while the first two are interferon-induced
proteins. Among genes with increased expression six are involved in immune response (GZMH,
CLEC1B, CLEC12A), viral mRNA translation (GCAT), energy metabolism (CACNA2D2) and
oxidative stress response (ENC1). Notably, 8/15 differentially expressed genes are regulated by
interferons. Our results suggest that an intrinsically weaker expression of some innate immunity-
related genes may be associated with an asymptomatic disease course in SARS-CoV-2 infection.
Whether a certain gene signature predicts, or not, those who will develop a more efficient immune
response upon exposure to SARS-CoV-2, with implications for prioritization for vaccination,
warrant further study.
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprintthis version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.19.21255748doi: medRxiv preprint

Introduction
Since December 2019 the SARS-CoV-2 has spread throughout the world infecting dozens of
millions of people and resulting in over 2.8 million deaths, as of April 2021. Although the case
fatality rate in hospitalized patients may exceed 10% [1, 2], 35-50% of infected adults do not
develop, perceive and report any clinical symptom [3, 4]. Asymptomatic infected persons may be
responsible for viral transmission for more days than aware self-isolated cases, which may also
explain, at least partially, the exponential increase in the number of infections globally [5, 6, 7].
Notably, we only know in retrospect who was indeed asymptomatic, since individuals without
symptoms at the time of a positive molecular test should be followed for 14 days to determine the
clinical picture, being "pre-symptomatic” if they develop symptoms later.
The proportion of asymptomatic individuals varies widely in viral infections. For example, a
significant fraction of cytomegalovirus infections, similarly to SARS-CoV-2, are asymptomatic and
unsuspected [8]. In contrast, an asymptomatic carrier state has not been documented for measles
virus infection [9]. The reasons why certain individuals, including people living with HIV [10] or
other immunodeficiencies [11], do not develop clinical symptoms during SARS-CoV-2 infection
are essentially unknown [12]. So far, studies assessing the immune response in asymptomatic
infection are few. In an elegant study, Long et al. showed that asymptomatic individuals presented
with significantly longer duration of viral shedding compared to symptomatic patients, lower levels
of IgG antibodies to SARS-CoV-2, and lower serum levels of 18/48 cytokines, including interferon-
gamma levels, suggesting that asymptomatic individuals indeed displayed a weaker anti-virus-
reactive immune response to SARS-CoV-2 [13].
While the role of genetics in determining immune and clinical response to the SARS-CoV-2 virus is
currently under investigation [14], it is well established that individual human immune systems are
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprintthis version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.19.21255748doi: medRxiv preprint

highly variable [15]. Most of this inter-individual immune variation is explained by environmental
exposures early in life [16] but genetic factors are clearly also involved. For example, a gene
expression signature dominated by interferon-inducible genes in the blood is prominent in systemic
lupus erythematosus [17], whereas interferon-α is increased not only in the serum of these patients
but also in their healthy first-degree relatives [18] pointing to genetic influences on the interferon-
mediated immune interactions.
Clearly, the most successful response against SARS-CoV-2 occurs in those individuals who, while
remaining asymptomatic, develop a robust adaptive immune response. We have recently examined
the humoral immune response to SARS-CoV-2 in members of the National and Kapodistrian
University of Athens, Greece [19]. Overall, among 4.996 people the unweighted seroprevalence of
SARS-CoV-2 antibodies was 1.58%, whereas 49% of the seropositive individuals denied having
had any clinical symptom compatible with previous SARS-CoV-2 infection, which was also
unsuspected for 33% of them. Interestingly, in our study, the mean levels of antibodies to both the
nucleocapsid (N) protein and the receptor-binding-domain (RBD) of the spike (S) protein were
comparable between asymptomatic and clinical infection cases and not associated with age or sex
[4]. Others have also reported that IgG antibodies are commonly observed in both asymptomatic
and clinical infections (85% versus 94% of patients, respectively) [20], whereas durable B cell-
mediated immunity against SARS-CoV-2 after mild or severe disease occurs in most individuals
[21].
Since variations in the strength and/or extent of the immune response may be critical for the clinical
picture and progress after infection with SARS-CoV-2, existing inter-individual differences at the
transcriptome level may be observed even later, after convalescence. Therefore, we performed 3’
mRNA next generation sequencing-based genome-wide transcriptional whole blood profiling to test
the hypothesis that immune response-related genes are differentially expressed between healthy
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprintthis version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.19.21255748doi: medRxiv preprint

individuals who developed an equally robust antibody response following either an entirely
asymptomatic or clinical SARS-CoV-2 infection.
Methods
Blood collection and anti-SARS-CoV-2 antibody testing
Blood samples were collected from members of the NKUA, Athens, Greece in June–November
2020. The protocol was approved by the Ethics and Bioethics Committee of the School of
Medicine, NKUA (protocol #312/02-06-2020) and study participants provided written informed
consent. All plasma samples were analyzed as previously described [4] using, a) the CE-IVD Roche
Elecsys® Anti-SARS-CoV-2 test, an electrochemiluminescence immunoassay (ECLIA) for the
detection of total antibodies (IgG, IgM, and IgA; pan-Ig) to SARS-CoV-2 N-protein (Roche
Diagnostics GmbH, Mannheim, Germany), and b) the CE-IVD Roche Elecsys® Anti-SARS-CoV-2
S, an ECLIA for the quantitative determination of antibodies (including IgGs) to the SARS-CoV-2
S-protein RBD (Roche Diagnostics).
3’ mRNA sequencing, mapping, quality control, and quantifications
Total RNA was isolated from whole blood, stored in paxgene, using the ExtractionMonarch® Total
RNA Miniprep Kit (NEB #T2010). Upon blood isolation, Monarch DNA/RNA Protection Reagent
(supplied as a 2x concentrate) was added undiluted to an equal volume of blood. Addition of the
protection reagent and the following RNA isolation was performed as described in the Kit's manual
for Total RNA Purification from Mammalian Whole Blood Samples.
After quantification on a NanoDrop ND-1000 (Thermofisher) and Bioanalyzer RNA 6000 Nano
assay (Agilent), 140-300ng of total RNA from samples passing quality control were processed
using the QuantSeq 3’ mRNA-Seq Library Prep Kit FWD (Lexogen, 015.96) for library
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprintthis version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.19.21255748doi: medRxiv preprint

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Blood transcriptomes of anti-sars-cov2 antibody positive healthy individuals with prior asymptomatic versus clinical infection" ?

Herein, the authors performed genome-wide transcriptional whole-blood profiling to test the hypothesis that immune response-related gene signatures may differ between healthy individuals with prior entirely asymptomatic versus clinical SARS-CoV-2 infection, all of which developed an equally robust antibody response. Whether a certain gene signature predicts, or not, those who will develop a more efficient immune response upon exposure to SARS-CoV-2, with implications for prioritization for vaccination, warrant further study. It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Their results suggest that an intrinsically weaker expression of some innate immunityrelated genes may be associated with an asymptomatic disease course in SARS-CoV-2 infection. 

Q2. What have the authors stated for future works in "Blood transcriptomes of anti-sars-cov2 antibody positive healthy individuals with prior asymptomatic versus clinical infection" ?

The described association of a subtle immune response to SARS-CoV-2 with a lack of clinical symptoms needs further investigation, which hopefully will be performed in the near future by established consortia [ 14 ] or other groups. Whether a certain innate immunity signature predicts, or not, those who will develop a more successful immune response upon contact with SARS-CoV-2, with possible implications for prioritization of vaccination, warrant further study. 

Q3. What is the role of ACE2 in the immune response to SARS-CoV-2?

the SARS-CoV-2 receptor ACE2, which is expressed in specific cell subsets across tissues is an interferon-stimulated gene in human airway epithelial cells [67], suggesting that a weaker individual interferon response may be protective. 

Q4. What is the characterization of IFIT3 as an antiviral gene?

Identification of alpha interferon-induced genes associated with antiviral activity in Daudi cells and characterization of IFIT3 as a novel antiviral gene. 

Q5. What was the metaseq method used to combine the data?

In order to combine the statistical significance from multiple algorithms and perform metaanalysis, the PANDORA weighted P-value across results method was applied through metaseqR2. 

Q6. What was used to estimate the proportion of blood immune cells?

Blood immune cell subsets deconvolutionCIBERSORTx [26] was used to estimate the proportion of blood immune cell subsets for each individual. 

Q7. What is the role of IFIT3 in the antiviral activity of Daudi cells?

Identification of alpha interferon-induced genes associated with antiviral activity in Daudi cells and characterization of IFIT3 as a novel antiviral gene. 

Q8. How many genes were found with significantly different expression in the asymptomatic patients?

Their results provide evidence that among 12.789 genes, there were only 15 with significantly different expression when comparing healthy, relatively young individuals after convalescence from a previous entirely asymptomatic SARS-CoV-2 infection to those with a clinical infection history. 

Q9. What is the significance of the differentially expressed genes in patients with asymptomatic SARS?

8 of the 15 differentially expressed genes in those with prior asymptomatic infection relatively to those with clinical SARS-CoV-2 infection can be found in datasets that include genes which have been implicated in interferon related signaling pathways in vitro [36]. 

Q10. What is the role of the apoptosis gene in NSCLC cells?

2021 Feb;13(2):275.46 Carboni GL, Gao B, Nishizaki M, Xu K, Minna JD, Roth JA, Ji L. CACNA2D2-mediated apoptosis in NSCLC cells is associated with alterations of the intracellular calcium signaling and disruption of mitochondria membrane integrity. 

Q11. How many libraries were assessed for molarity and median library size?

Libraries were assessed for molarity and median library size using Bioanalyzer High Sensitivity DNA Analysis (Agilent, 5067-4626). 

Q12. How long should the authors follow asymptomatic individuals?

the authors only know in retrospect who was indeed asymptomatic, since individuals without symptoms at the time of a positive molecular test should be followed for 14 days to determine the clinical picture, being "pre-symptomatic” if they develop symptoms later. 

Q13. What was the significance of the difference in distributions between the asymptomatic and clinical groups?

The Mann-Whitney U test was applied in order to calculate the significance of the difference in distributions between the asymptomatic and clinical groups. 

Q14. What is the reason for the lack of differentially expressed genes?

such differential responses should be more robust at the time of infection and more genes and immune networks may be differentially expressed.