key: cord-0710333-ry97bngd
authors: Carsetti, R.; Zaffina, S.; Piano Mortari, E.; Terreri, S.; Corrente, F.; Capponi, C.; Palomba, P.; Mirabella, M.; Cascioli, S.; Palange, P.; Cuccaro, I.; Milito, C.; Zumla, A.; Maeurer, M.; Camisa, V.; Vinci, M. R.; Santoro, A.; Cimini, E.; Marchioni, L.; Nicastri, E.; Palmieri, F.; Agrati, C.; Ippolito, G.; Porzio, O.; Concato, C.; Onetti Muda, A.; Raponi, M.; Quintarelli, C.; Quinti, I.; Locatelli, F.
title: Spectrum of innate and adaptive immune response to SARS CoV 2 infection across asymptomatic, mild and severe cases; a longitudinal cohort study
date: 2020-06-23
journal: nan
DOI: 10.1101/2020.06.22.20137141
sha: 850fbb2fdd709c7cb3cd0767d1bb42289095fe88
doc_id: 710333
cord_uid: ry97bngd

SARS-CoV-2 is a novel coronavirus, not encountered before by humans. The wide spectrum of clinical expression of SARS-CoV-2 illness suggests that individual immune responses to SARS-CoV-2 play a crucial role in determining the clinical course after first infection. Immunological studies have focussed on patients with moderate to severe disease, demonstrating excessive inflammation in tissues and organ damage. In order to understand the basis of the protective immune response in COVID-19, we performed a longitudinal follow-up, flow-cytometric and serological analysis of innate and adaptive immunity in 64 adults with a spectrum of clinical presentations: 28 healthy SARS-CoV-2-negative contacts of COVID-19 cases; 20 asymptomatic SARS-CoV-2-infected cases; 8 patients with Mild COVID-19 disease and 8 cases of Severe COVID-19 disease. Our data show that high frequency of NK cells and early and transient increase of specific IgA, IgM and, to a lower extent, IgG are associated to asymptomatic SARS-CoV-2 infection. By contrast, monocyte expansion and high and persistent levels of IgA and IgG, produced relatively late in the course of the infection, characterize severe disease. Modest increase of monocytes and different kinetics of antibodies are detected in mild COVID-19. The importance of innate NK cells and the short-lived antibody response of asymptomatic individuals and patients with mild disease suggest that only severe COVID-19 may result in protective memory established by the adaptive immune response.

Antibodies to SARS-CoV-2 are produced in large amounts in patients with severe disease, two-116 three weeks after the occurrence of first symptoms [6] . The role of antibodies in viral elimination is 117 supported by the successful use of convalescent plasma in patients with severe COVID-19 [7] . 118

Neutralizing antibodies are directed against the Receptor Binding Domain contained in the S1 119 subunit of the Spike protein [8, 9] . Whilst immune responses to novel antigens encountered for the 120 first time, are first dominated by antibodies of IgM isotype, followed by IgG[10], the kinetics and 121 protective or deleterious nature of the antibody responses to SARS-CoV-2 remains to be defined. A 122 recent study indicated that the IgA response to SARS-CoV-2 may be rapid, strong and 123 persistent [11] . The observation that patients with severe COVID-19 disease have very high 124 antibody levels led to the suggestion that antibodies to SARS-CoV-2 may be damaging or 125 ineffective rather than protective [12, 13] , as was reported from very sick patients with Middle East 126 respiratory syndrome (MERS) [14] . 127

Whilst immunological studies to date have focussed on patients with moderate to severe COVID-19 128 disease [15] , studies from across the clinical spectrum are required to better understand immune 129 responses to SARS-CoV-2 infection. We performed a longitudinal study of innate and adaptive 130 immune populations in the peripheral blood of adults with asymptomatic SARS-CoV-2 infection, 131 and those with mild and severe COVID-19 disease and their healthy contacts. 132

133

Methods: 134 135

Ethical approval 136 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. 

. CC-BY-NC-ND 4.0 International license It 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 preprint this version posted June 23, 2020. . https://doi.org/10.1101/2020.06.22.20137141 doi: medRxiv preprint Four leukocyte profiling panels computing 7-to 9-surface marker antigens for monitoring the major 172 leukocyte subsets as well as characteristics of T-cell, B-cell, monocytes and NK cells subsets were 173 designed (Supplementary Table S2 contacts, fifty-one from SARS-Cov-2 asymptomatic patients, forty-one from COVID-19 mild 190 patients and fifteen from COVID-19 severe patients. All sera were kept on ice after collection and 191 then stored at -80°C. 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 preprint this version posted June 23, 2020. We performed the unpaired, two-tailed Mann-Whitney U-tests. A p£0.05 was considered to be 206 statistically significant. 207 208

Innate immunity 210

The PBMCs of patients with asymptomatic infection, mild and severe disease and their healthy 211 contacts were compared. We correlated the immunological findings with the clinical course and 212 studied the dynamic changes of cells of innate and adaptive immune response. 213

By flow-cytometry, we confirmed that the Monocyte-Lymphocyte ratio (MLR) increases in 214 advanced COVID-19 cases, when T cells, normally representing the major lymphocyte population 215 in the peripheral blood, are reduced[15] (Figure 1a We found that the frequency of NK cells was reduced and that of monocytes increased in patients 227 with severe COVID-19. Similar alterations of NK and monocytes were observed in the group of 228 patients with mild disease (Figure 1d and e) . For this reason, we calculated the Monocyte to NK 229 ratio (MNKR), which appeared to be altered not only in the severe cases, but also in patients with 230 mild disease (Figure 1e ). The differences between asymptomatic, mild and severe COVID-19 231 remained significant when we included all the samples collected from the patients at different time 232 points (Figure 2a) , suggesting that the reduction of NK and monocytes and the high MNKR was 233 not an incidental finding observed in a particular moment of the infection, but rather is a 234 characteristic of the disease. Each individual maintained his typical NK and Monocyte frequency 235 throughout the time of follow-up (Figure 2b) . We confirmed the importance of the frequency of 236 NK cells by the retrospective analysis of 77 patients with severe COVID-19. Cases who did not 237 need ICU treatment had a significantly higher number of NK cells (CD56 + cells calculated in CD3 -238 lympho-monocyte gate) than ICU patients (Figure 2c ). In addition, the percentage of NK cells 239 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. . https://doi.org/10.1101/2020.06.22.20137141 doi: medRxiv preprint COVID-19, not only the number of CD8 + T cell declines but also their function is impaired, in 275 association with the increase of pro-inflammatory cytokines [18] . 276 We observed an increase of activated, HLA-DR + CD4 T cells in patients with mild and severe 277 disease (Figure 3a and b) . By contrast, HLA-DR + CD8 T cells were increased only in patients with 278 severe COVID-19 (Figure 3 a and b) . This finding was confirmed when we included all samples in 279 the analysis (Supplementary Figure 4) . 280

In a separate staining, we identified naïve and memory T cells, including central, effector and 281 terminally differentiated (TEMRA) memory T cells (gating strategy in supplementary Figure 5) . 282

In the CD4 + population of patients with mild and severe COVID-19, we found a reduction of recent 283 thymic emigrants (CD45RA + CCR7 + CD31 + ) and a consequent relative increase of CD31 -284 Figure 7) . The most significant findings were the reduction of 301 total B cells and the increase of plasmablasts in the severe cases (Figure 5b) . Among MBCs, we 302 found an increase of IgM + and a reduction of switched MBCs in asymptomatic and mild cases. In 303 patients with severe disease, IgM + MBCs were reduced and switched MBCs increased (Figure 5c) , 304 with a significant expansion of the IgGswitched population. IgG + MBCs were reduced in the blood 305 of patients with mild disease (Figure 5d) . For this reason, the ratio between IgG + and IgGswitched 306

MBCs, that is >1 in the SARS-Cov-2 negative contacts and asymptomatic individuals, is 307 significantly lower in patients with mild and severe COVID-19 indicating the expansion of MBCs 308 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. . https://doi.org/10.1101/2020.06.22.20137141 doi: medRxiv preprint that do not express IgG and most probably express IgA (Supplementary Figure 7) . All the 309 findings were confirmed by the cumulative analysis of all samples (Supplementary Figure 8) . IgG and IgA antibodies directed against the S1 domain of the SARS-CoV-2 Spike protein were 316 measured in the entire study cohort. We found that antibodies are produced by all COVID-19 317 patients and also by SARS-CoV-2 positive asymptomatic individuals, with higher levels of IgG and 318

IgA being detected in the serum of patients with severe disease (Figure 6a ) as reported 319 before [13, 32] . 320

In order to study the dynamic nature of the antibody response, we analysed its kinetics in 9 321 asymptomatic individuals, 8 patients with mild and 4 with severe disease. We found that 322 asymptomatic patients secrete high levels of S1-specific IgA early after diagnosis. IgA rapidly 323 declines and becomes undetectable after 5-7 weeks. IgG is low at most time points (Figure 6b) . 324

The response of patients with mild disease has different kinetics, with IgG and IgA increasing later 325 and remaining relatively low, with few exceptions (Figure 6c) . IgG and IgA are produced late in 326 patients with severe COVID-19, but in higher amounts [13, 32] . IgA was always more abundant than 327 IgG (Figure 6d) . 328

Antigen-specific IgA and IgG were undetectable in 2 patients with severe disease and in 1 with 329 mild disease. The inability to produce antibodies in isolated cases with severe and mild disease may 330 be due to individual genetic variation or age: pt34 was 83-year-old with a severe disease (1 single 331 sample was obtained), and pt14 was young but had a rapidly evolving severe disease (3 samples 332 were analysed at different time points), that improved with treatment. Pt11 (data not shown) with 333 mild disease was otherwise healthy, but had neurological symptoms and a positive nasopharyngeal 334 swab PCR lasting for 8 weeks (6 samples were evaluated). 335

The most interesting and intriguing aspect of the immunoglobulin study was the observation that 336 among the samples of asymptomatic individuals 17% (9 of 53 samples) were negative for IgG and 337 22.6% (12 of 53) were negative for IgA. This high percentage of negative samples cannot be 338 explained by individual variations. As serum antibody titers rapidly decline in asymptomatic 339 patients (Figure 6b) , we correlated the level of SARS-CoV-2 specific antibody to the time from the 340 first positive nasopharyngeal swab. We found that six serum samples with non-detectable specific 341

IgA and IgG had been obtained from individuals diagnosed as positive for SARS-CoV-2 eight to 342 sixteen weeks prior to serum analysis (Figure 6e-empty circles) . Of the other asymptomatic 343 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Our data show that the balance between NK cells and monocytes is a sensitive indicator of the 375 individual reaction to the virus and is related to the clinical course of the disease. We found that 376 when NK cells are reduced, monocytes increase. Since each individual included in our study 377 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. The particular antigen-specific IgA/IgG profile associated with clinical outcome may reflect 408 increased TGF-beta production induced by coronavirus-species [58] . Increased viral load may 409 increase TGF-beta production, that -if produced locally in the lung, facilitates neutrophil attraction 410 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. 19 recovered patient cohort and their implications. medRxiv 2020; 481 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. 

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 23, 2020. 

Graphs show the percentage of naïve T cells (CD3 + CCR7 + CD45RA + ) that either expressed or lack CD31 (CD31 + are 661 . CC-BY-NC-ND 4.0 International license It 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 preprint this version posted June 23, 2020. . https://doi.org/10.1101/2020.06.22.20137141 doi: medRxiv preprint recent thymic emigrants and CD31are revertant T cells). Effector memory T cells (CD3 + CCR7 -CD45RA -) can be 662 separated in EM1, EM2, EM3 and EM4. TEMRA T cells are CD3 + CCR7 -CD45RA + . Midlines indicate mean. Statistical 663 significances were determined using unpaired, two-tailed Mann-Whitney U-tests. *p£0.05, **p<0.01, ***p<0.001.

. CC-BY-NC-ND 4.0 International license It 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 preprint this version posted June 23, 2020. CC-BY-NC-ND 4.0 International license It 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 preprint this version posted June 23, 2020. 

. CC-BY-NC-ND 4.0 International license It 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 preprint this version posted June 23, 2020. . https://doi.org/10.1101/2020.06.22.20137141 doi: medRxiv preprint

Clinical course and risk factors for mortality of adult inpatients 461 with COVID-19 in Wuhan, China: a retrospective cohort study

COVID-19: unanswered questions on immune response 463 and pathogenesis

Lymphopenia is associated with severe coronavirus 466 disease 2019 (COVID-19) infections: A systemic review and meta-analysis

Reduction and Functional Exhaustion of T Cells in Patients 469 With Coronavirus Disease 2019 (COVID-19)

Phenotype of SARS-CoV-2-specific T-cells in 471 COVID-19 patients with acute respiratory distress syndrome

Antibody responses to SARS-CoV-2 in patients with 474 COVID-19

Effectiveness of convalescent plasma therapy in severe COVID-476 19 patients

SARS-CoV-2 specific antibody responses in COVID-478 19 patients

Memory B cells

IgA-Ab response to spike glycoprotein of SARS-484

CoV-2 in patients with COVID-19: A longitudinal study

Immune phenotyping based on neutrophil-to-lymphocyte 486 ratio and IgG predicts disease severity and outcome for patients with COVID-19

Antibody responses to SARS-CoV-2 in patients of novel 489 coronavirus disease 2019

Molecular Mechanism for Antibody-Dependent Enhancement 491 of Coronavirus Entry

The clinical course and its correlated immune status in COVID-493 19 pneumonia

Asymptomatic Cases with SARS-CoV-2 Infection

Middle east respiratory syndrome

An inflammatory profile correlates with decreased 499 frequency of cytotoxic cells in COVID-19

The fate and lifespan of human monocyte subsets in 502 steady state and systemic inflammation

The 'intermediate' CD14++ CD16+ 504 monocyte subset increases in severe peripheral artery disease in humans

Expanded CD14+ CD16+ Monocyte subpopulation in patients 506 with acute and chronic infections undergoing hemodialysis

Nonclassical Monocytes in 508 Health and Disease

The multifaceted role of CD4+ T cells in CD8+ T cell 510 memory

Germinal center dynamics revealed by 512 multiphoton microscopy with a photoactivatable fluorescent reporter

Epidemiological and clinical characteristics of 99 cases of 515 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

Functional exhaustion of antiviral lymphocytes in COVID-517 19 patients

Four Functionally Distinct Populations of Human 519 Effector-Memory CD8 + T Lymphocytes

Lack of Gut Secretory Immunoglobulin A in 602

Analysis of a SARS-CoV-2 infected individual 605 reveals development of potent neutralizing antibodies to distinct epitopes with limited 606 somatic mutation

The Egyptian Rousette Genome Reveals 609 Unexpected Features of Bat Antiviral Immunity

Immunoglobulin heavy chain diversity in Pteropid bats: 611 Evidence for a diverse and highly specific antigen binding repertoire

Novel Insights Into 613 Immune Systems of Bats

Key to successful treatment of COVID-19: accurate 615 identification of severe risks and early intervention of disease progression

Dysregulation of immune response in patients with COVID-19 in 617 Wuhan

Waiting for Certainty on Covid-19

Antibody Tests -At What Cost?

We identified transitional (CD24 + CD38 ++ ), naïve (CD24 + CD27 -), memory (CD24 + CD27 + )

atypical MBCs (CD24 -CD38 -) and plasmablasts (CD24 -CD27 ++ CD38 ++ ). In the CD27 + memory B-cell population based 671 on IgM expression, we show IgM and switched (IgM -) MBCs. MBCs were also gated as IgM + , IgG + and IgG -IgM -672

Plots indicates the percentage of B cells, MBCs and plasmablasts

The ratio between IgG + and IgG -IgM -is also 674 shown. Midlines indicate mean. Statistical significances were determined using unpaired, two-tailed Mann-Whitney U-675 tests

**p<0.01, ***p<0.001. (b-d) Graphs show the levels (AU) of IgA (solid line) and IgG

during the course of the disease in three asymptomatic (b), three mild (c) and three severe (d) disease patients

Time is indicated in weeks from the first positive nasopharyngeal 685 swab. Empty circles represent patients of which we had only one sample and antibodies were undetectable