key: cord-0882663-h00mbgsq
authors: Georg, Philipp; Astaburuaga-García, Rosario; Bonaguro, Lorenzo; Brumhard, Sophia; Michalick, Laura; Lippert, Lena J.; Kostevc, Tomislav; Gäbel, Christiane; Schneider, Maria; Streitz, Mathias; Demichev, Vadim; Gemünd, Ioanna; Barone, Matthias; Tober-Lau, Pinkus; Helbig, Elisa T.; Hillus, David; Petrov, Lev; Stein, Julia; Dey, Hannah-Philine; Paclik, Daniela; Iwert, Christina; Mülleder, Michael; Aulakh, Simran Kaur; Djudjaj, Sonja; Bülow, Roman D.; Mei, Henrik E.; Schulz, Axel R.; Thiel, Andreas; Hippenstiel, Stefan; Saliba, Antoine-Emmanuel; Eils, Roland; Lehmann, Irina; Mall, Marcus A.; Stricker, Sebastian; Röhmel, Jobst; Corman, Victor M.; Beule, Dieter; Wyler, Emanuel; Landthaler, Markus; Obermayer, Benedikt; von Stillfried, Saskia; Boor, Peter; Demir, Münevver; Wesselmann, Hans; Suttorp, Norbert; Uhrig, Alexander; Müller-Redetzky, Holger; Nattermann, Jacob; Kuebler, Wolfgang M.; Meisel, Christian; Ralser, Markus; Schultze, Joachim L.; Aschenbrenner, Anna C.; Thibeault, Charlotte; Kurth, Florian; Sander, Leif E.; Blüthgen, Nils; Sawitzki, Birgit
title: Complement activation induces excessive T cell cytotoxicity in severe COVID-19
date: 2021-12-28
journal: Cell
DOI: 10.1016/j.cell.2021.12.040
sha: 0efde14de85a5beac6f10a5c99d573f41ca5ac65
doc_id: 882663
cord_uid: h00mbgsq

Severe COVID-19 is linked to both dysfunctional immune response and unrestrained immunopathology, and it remains unclear whether T cells contribute to disease pathology. Here, we combined single-cell transcriptomics and single-cell proteomics with mechanistic studies to assess pathogenic T cell functions and inducing signals. We identified highly activated, CD16+ T cells with increased cytotoxic functions in severe COVID-19. CD16 expression enabled immune complex-mediated, T cell receptor-independent degranulation and cytotoxicity not found in other diseases. CD16+ T cells from COVID-19 patients promoted microvascular endothelial cell injury and release of neutrophil and monocyte chemoattractants. CD16+ T cell clones persisted beyond acute disease maintaining their cytotoxic phenotype. Increased generation of C3a in severe COVID-19 induced activated CD16+ cytotoxic T cells. Proportions of activated CD16+ T cells and plasma levels of complement proteins upstream of C3a were associated with fatal outcome of COVID-19, supporting a pathological role of exacerbated cytotoxicity and complement activation in COVID-19.

Severe acute respiratory distress syndrome Coronavirus 2 (SARS-CoV-2) infection in humans 88 causes a diverse spectrum of clinical manifestations, ranging from asymptomatic disease to 89 acute respiratory distress syndrome (ARDS) and multi-organ failure (Miyazawa, 2020) . 90

In addition to direct virus-induced injury to the respiratory system and other organs, increasing 91 evidence suggests that the immune response evoked by SARS-CoV-2 infection contributes to 92 the pathophysiology of Coronavirus disease (COVID-19), particularly during severe disease 93

courses ( A higher state of T cell activation in all T cell compartments (CD4 + , CD8 + , double-negative 106 (DN)) in patients progressing to severe COVID-19 was reported (Zenarruzabeitia et al., 2021) . 107

Interstitial T cell infiltration is observed in pathological specimens of COVID-19 pneumonia 108 along with macrophage accumulation in the alveolar space, and it has been hypothesized that 109

infiltrating T cells also contribute to alveolar wall damage and endothelial cell injury known as 110 lymphocytic endotheliitis (Miyazawa, 2020; Varga et al., 2020) . 111

All this argues for a complex relationship between T cell immune responses and disease 112 outcome during COVID-19 beyond a mere quantitative influence. It is likely that additional 113 factors present in the microenvironment will shape the quality of T cell responses and 114 consequently impact pathology. It is therefore important to identify whether and which T cell 115 subsets have a pathogenic role. Also, mechanisms by which potentially pathogenic T cells are 116 induced need to be revealed, as studies on this matter are currently lacking (Yan et al., 2021) . 117

Here, we combined single-cell proteomics and transcriptomics with mechanistic studies to 118 reveal alterations in the T cell compartment, their upstream signals, and functional relevance, 119

which explain important immunopathological features observed in severe COVID-19. Mass 120 cytometry (CyTOF) and single-cell RNA-seq (scRNA-seq) combined with VDJ-seq-based T 121 cell clonotype identification were used to determine COVID-19-and severity-specific 122

alterations in the T cell compartment. In addition to the severity-independent formation of 123 highly activated HLA-DR hi CD38 hi CD137 + Ki67 + T follicular helper (TFH) -like cells and CD8 + 124 Samples of COVID-19 patients collected during the first three weeks after symptom onset 146 independent of the severity were characterized by increased proportions of CD4 + T cell cluster 147 7 (CD38 hi HLA-DR + Ki67 + ICOS + ) compared to other infections or controls, whereas the 148 abundance of cluster 18 (CD27 -CD25 + ) was lower ( Figure 1D , E). Cluster 7 T cells are 149 characterized by high expression of activation markers such as HLA-DR, CD38, CD137, 150 CD69, and Ki67. Furthermore, the T cells in this cluster express CXCR5, ICOS, and PD-1, 151

resembling TFH-like cells. The increase of TFH-like cells was not seen for FLI, neither HIV, 152 nor HBV. T helper cells in severe COVID-19 patients showed higher proportions of cluster 8 153 T cells (CD4 + , highly activated natural killer T cell (NKT-like), which in addition to the 154 expression of CD38, HLA-DR, CD137, CD69, Ki67 and CXCR3, were characterized by high 155 levels of CCR6 and CD16. Those T cells belong to the CD62L + CD45RO + central memory 156

metacluster, reflecting a more recent differentiation. In contrast, in patients suffering from 157 chronic HIV or HBV infection we observed a tendency towards a higher proportion of cluster 158 19 (TIGIT + CD38 + CD16 + ). T cells from this cluster were classified as terminally differentiated 159 RA + cells, expressed low levels of CD16, and showed no signs of recent activation. 160

This activation pattern was not restricted to the CD4 + T cell compartment, but also detectable 161 in CD8 + CTLs and TCRgd + T cells. Both mild and severe COVID-19 patients displayed 162

increased proportions of cluster 25 T cells (CD8 + CD38 hi HLA-DR + Ki67 + ) in comparison to the 163 other groups ( Figure 1D , E). Severe COVID-19 was further characterized by increased 164 abundance of cluster 26 T cells (CD8 + , highly activated NKT-like), also expressing high levels 165 of CCR6 and CD16. We observed that severity explained on average 96% of the variance in 166 cluster frequencies, whereas age only 4% ( Figure S1A ), indicating that the increased 167

proportions of activated CD16 + T cells are mainly due to disease severity and not to age. 168 J o u r n a l P r e -p r o o f 8 We confirmed the existence of a CD3 + CD16 + HLA-DR + T cell population by flow cytometry in 169 a second cohort of COVID-19 patients ( Figure S1B ). 170

Time-dependent analysis of the CD4 + and CD8 + T cell clusters revealed a trend to faster 171 accumulation of the activated CD38 hi HLA-DR + Ki67 + expressing clusters 7 and 25 during the 172 first week after SARS-CoV-2 infection in severe versus mild COVID-19 ( Figure S1C , D). In 173 contrast, the highly activated CD16 + NKT-like CD4 + and CD8 + T clusters appeared only during 174 the second week of infection. 175

Our findings concerning the identification of highly activated and proliferating T helper cells 176 and CTLs after SARS-CoV-2 infection is in line with recent reports (Mathew et 

188 To obtain functional information on the COVID-19-and severity-specific T cell clusters, we 189 performed scRNA-seq analysis of peripheral blood mononuclear cell (PBMC) samples as well 190

as purified CD38 expressing T cells from acutely infected and convalescent mild and severe 191 COVID-19, as well as FLI, HBV and controls. To align the CyTOF and scRNA-seq T cell 192 clusters, we applied a feature-based cluster annotation approach on the whole T cell space 193 from PBMC and CD38 + T cell libraries (STAR Methods) which resulted in 17 clusters ( Figure  194 2A, B). The proportion of T cells belonging to clusters 7, 8 and 10 was higher in COVID-19 195 patients compared to FLI or HBV patients as well as controls ( Figure 2D , E). We observed 196 other T cell clusters with FCGR3A expression (clusters 9, 11, 12, 13). Of these, only cluster 9 197

transcribed SELL (CD62L) and displayed a central memory phenotype, whereas clusters 11, 198 12 and 13 T cells displayed a more advanced differentiation profile. Overall, cluster 7 contains 199 TFH-like cells similar to the CyTOF cluster 7, whereas cluster 8, 9 and 10 most likely contain 200 a mixture of highly activated and CD16 + NKT-like cells with similarities to the CyTOF clusters 201 8, 24 and 25. Next, we performed a Gene Ontology (GO) enrichment analysis across clusters 202 7, 8, 9 and 10, comparing mild and severe COVID-19 T cells ( Figure 2F ). Interestingly, 203

whereas T cells from mild patients showed an enrichment for cellular responses to type I 204 interferon and antiviral defense, we observed a specific enrichment of genes involved in 205 degranulation in severe COVID-19 T cells. This selective enrichment was validated by a Gene 206

Set Enrichment Analysis (GSEA) for genes belonging to the GO terms 'response to type I 207 interferon' and 'defense response to virus', which were enriched in mild, and for genes 208 9 belonging to a cytotoxicity signature, which were enriched in severe COVID-19 T cells (all 209 p_adj<0.005, Figure 2G ). Indeed, several genes that are known to promote degranulation 210 (LAMP1, STX11) or exert cytotoxic potential (PRF1, GZMB, GZMH, GZMK) were increased 211 in severe COVID-19 across clusters 7, 8, 9 & 10 T cells ( Figure 2H+S2B ) or the whole T cell 212 space ( Figure S2C ). We validated these findings in scRNA-seq data from our second cohort 213 ( Figure S3 ). Here, clusters 6 and 7 contained CD4 + T cells transcribing ICOS, CD40LG, 214 PDCD1 and CXCR5 and thus resembled cluster 7 T cells from cohort 1, although the 215 expression level of CD38 and HLA-DR genes seemed to be lower. Clusters 10 & 13 from the 216 second cohort resembled clusters 10 & 8 in cohort 1, respectively. Furthermore, GSEA 217 similarly showed increased expression of genes mediating cytotoxicity in severe COVID-19 T 218 cells across these clusters ( Figure S3E , F). 219

Thus, scRNA-seq analysis of samples from two independent cohorts supported our finding of 220 a subset of activated CD16 + T cells across the major T cell compartments in severe COVID-221

19 and identified an increase in cytotoxicity-associated transcriptional programs. 222 223 CD16-mediated degranulation of CD8 + T cells causes chemokine 224 release by endothelial cells 225 CyTOF and scRNA-seq analyses identified two main T cell activation features: i) formation of 226 highly activated, proliferating TFH-like CD4 + cells and CXCR3 expressing CTLs independent 227 of disease severity, and ii) activated CD16 + T cells specific for severe COVID-19. Since TFH 228 cells promote B cell help (Crotty, 2014 ) and since we observed a trend for a faster formation 229 of activated TFH-like cells in severe COVID-19 patient samples, we tested whether SARS-230

CoV-2-specific antibody responses were more pronounced in those patients. Serum 231

concentrations of SARS-CoV-2-specific IgA, but particular IgG levels were higher in severe 232 COVID-19 ( Figure 3A ), which is in line with previous reports (Garcia-Beltran et al., 2021). 233

Furthermore, maximal antibody levels determined during the second week post-symptom 234 onset correlated positively with the cell proportions of the TFH-like CyTOF cluster 7 ( Figure  235 3B), whereas the CD16 + CD4 + T cell cluster 8 did not correlate with antibody levels (IgA: 236 R 2 =0.11, p=0.19; IgG: R 2 =0.18, p=0.08; data not shown). Although severe COVID-19 patients 237 developed a faster antibody response, all study cohort patients were reactive (OD ratios>1.1) 238

at later time points. 239

Next, we investigated the functional properties related to the CD16 + CD4 + and CD8 + clusters. 240

As indicated by scRNA-seq, samples from patients with severe COVID-19 contained 241 significantly more Granzyme B expressing CD8 + T cells compared to controls ( Figure 3C ). 242

Since CD16 is known to mediate antibody-mediated degranulation of NK cells (Moretta et al.,  243 2008), we tested whether T cells from patients with severe COVID-19 display enhanced CD16-244 dependent degranulation potential. As a surrogate for immune complex-mediated stimulation, 245

we assessed cell surface CD107a in PBMCs from mild or severe patients as well as controls 246

after 6-hour incubation with anti-CD16 antibody or isotype antibody-coated beads ( Figure 3D ). 247

Stimulation with anti-CD16 elicited strong degranulation of CD8 + T cells from patients with 248 10 severe COVID-19, compared to T cells from non-infected controls ( Figure 3D ) . T cells from  249  patients with mild COVID-19 showed intermediate degranulation potential. As a second  250  approach to assessing immune complex-mediated degranulation, we stimulated T cells with  251  SARS-CoV-2 spike protein-coated beads complexed with patient-derived serum. Serum from  252  COVID-19 patients induced similar degranulation as anti-CD16 antibody-coated beads, which  253 was not observed with serum from uninfected controls ( Figure 3E ). 254

We also investigated SARS-CoV-2 specificity of activated CD16 + T cells. PBMCs presence of anti-CD16 antibodies. Subsequently, we analyzed the release of inflammatory 274 mediators ( Figure 3G ). Anti-CD16-triggered severe COVID-19 T cells elicited enhanced 275 CXCL8 (IL-8) and CCL2 (MCP-1) release by co-cultured endothelial cells. Chemokines were 276 produced by endothelial cells, as we did not observe chemokine release by anti-CD16-277 triggered T cells alone (below <0.4pg/ml for CXCL8 and <1.9pg/ml for CCL2) . T cells from  278  COVID-19 patients amplified Concanavalin A-induced loss of transendothelial electrical  279 resistance, indicating endothelial barrier disruption, as compared to control T cells, but this 280 effect was only significant for T cells from patients with severe COVID-19 ( Figure 3H ). 281

Complementing our findings in peripheral blood, we investigated the tissue localization of 282 CD16 + T cells in the lungs of patients with COVID-19. We co-stained CD3 and CD16 in 283 autopsy lung tissues from COVID-19 patients and from three different cohorts of non-COVID 284 patients, i.e. patients without inflammatory or fibrotic lung disease (control), patients with non-285 infectious acute respiratory distress syndrome (ARDS), and influenza pneumonia (either 286 H1N1 or seasonal influenza) positive patients (FLU). In autopsy lung tissue from COVID-19 287 patients, we found an increased number of CD3 + CD16 + T lymphocytes in comparison to lung 288 tissue from the different control autopsy groups ( Figure 3I+J ). The pulmonary accumulation of 289 J o u r n a l P r e -p r o o f 11 CD3 + /CD16 + cells was most prominent in autopsy samples from COVID-19 patients between 290 7 and 14 days and declined at late time points of death ( Figure 3J ). 291

We also made use of a published scRNA-seq data set of bronchoalveolar lavage (BAL)  292  samples from COVID-19 and non-COVID-19 T cells from severe COVID-19 patients display enhanced cytotoxic  309 properties potentially contributing to organ damage, we analyzed their persistence after 310 clearance of the acute infection. We obtained VDJ sequence information in addition to the 311 gene expression data of T cells from acute and convalescent samples of mild and severe 312 COVID-19 as well as FLI, HBV and controls, allowing us to study the fate of early expanded 313 T cell clones during convalescence at months 3 to 8 post-symptom onset (Figure 4 A-E). First, 314

we analysed whether the individual COVID-19 T cell clusters differ in their clonal enrichment. 315

The FCGR3A expressing scRNA-seq clusters enriched in patients with COVID-19 (8, 9 and 316 10) showed a high level of clonal enrichment during acute COVID-19 infection, while it was 317 rather low for cluster 7, containing CD4 + TFH-like cells ( Figure 4A ). Consequently, only clones 318 belonging to clusters 8, 9 and 10, composed of mixed CD4/CD8A or CD8A/TCRgd T cell 319 clusters, displayed a high degree of persistence with up to 50% of the TCR clones being 320 recovered in convalescent samples ( Figure 4B ). Furthermore, clones expanded in the highly 321

proliferating cluster 9 showed a higher persistence in severe COVID-19 patients ( Figure 4B ). 322

Following the clone-specific VDJ sequences also allowed us to track their differentiation 323 trajectory during convalescence ( Figure 4C+D ). T cells expanded during the acute infection 324 belonging to clusters 7-10 dropped dramatically during convalescence ( Figure 4D ), the 325 respective T cell clones evolved into T cells mainly identified in clusters 11 and 12, which 326 retained FCGR3A expression ( Figure 4C+D ). GSEA further revealed the high cytotoxic 327 potential of these clones, among clusters 6, 11 and 12 T cells of COVID-19 patients versus  328  12   controls, with an evident difference even when comparing severe versus mild COVID-19  329 (padj<0.01, Figure 4E ). 330

Using the VDJ sequence information allowed us also to define whether the activated 331 FCGR3A + clusters are mainly composed of NKT cells known to show innate immune cell 332

functions (Krovi and Gapin, 2018) . With the vast majority of iNKT cells expressing an identical 333

TCRα chain (TRAV10-TRAJ18) paired to a restricted set of TCRβ chains (TRBV25), we 334 determined the proportion of TRAV10-TRAJ18-TRBV25 pairing T cell clones across all T cell 335 clusters. The median proportion of all T cells in our dataset expressing the TCR alpha-beta 336 pair was 0.04%. We did not observe a major iNKT cell enrichment in the described COVID-19 337

T cell clusters 7-10, with the highest frequency of 0.03% in cluster 7). 338 Furthermore, we labelled cells from samples during COVID-19 convalescence with the CyTOF 339 clusters previously found via k-nearest neighbour ( Figure 4F , Figure S4A ). Similar to our 340 scRNA-seq findings, we observed alterations in T cell cluster abundances during 341

convalescence. For CD8 + T cells, we observed a strong increase in T cell proportions for 342 cluster 29, which belongs to the effector memory meta cluster and is characterized by PD-1, 343

TIGIT as well as CD137 expression. To corroborate a potential link between aging and complement activity, we analyzed the 370 correlation of the complement component complement factor D (CFD) with age in the 371 generation Scotland study ( Figure 5B , (Messner et al., 2020) ), which revealed a clear age-372 dependent increase of CFD. We then extended these findings to COVID-19 patients and 373

showed that particularly severe COVID-19 at early time points was characterized by 374 significantly elevated CFD plasma concentrations ( Figure 5C , left panel). 375

CFD is a serine protease that catalyses the formation of the active C3 convertase in the 376 alternative pathway (Noris and Remuzzi, 2013) . C3a plasma levels detected in severe COVID-377

19 patients during the first three weeks after symptom onset exceeded C3a levels detected in 378 mild COVID-19, or other acute respiratory infections ( Figure 5C , middle panel). C5a levels 379

peaked in week three post-onset of symptoms but did not reach significance ( Figure 5C , right 380 panel). Both, CD4 + and CD8 + T cells from COVID-19 patients regardless of disease severity 381 displayed higher C3a binding potential than cells from controls ( Figure 5D ). Finally, plasma 382

C3a levels in COVID-19 patients measured at week 2 post-symptom onset correlated with 383

proportions of COVID-19-specific activated CD16 expressing CD4 + and CD8 + CyTOF clusters 384 (cluster 8+26) ( Figure 5E ). 385

Next, we stimulated enriched CD3 + cells from healthy unexposed controls with plate-bound 386

anti-CD3/CD28 antibodies in the presence of recombinant IL-2 and serum from mild or severe 387 COVID-19 patients or control serum. The addition of serum from severe COVID-19 patients 388 resulted in a 10-and 20-fold increase of CD16 + T cells, which was higher compared to the 389 increase observed when adding serum from mild COVID-19 patients ( Figure 5F ). Furthermore, 390 the addition of recombinant C3a to cells stimulated in the presence of control serum enhanced 391 the formation of CD16 + T cells ( Figure 5F ). The in vitro-generated T cells phenotypically and 392 functionally resembled the T cells identified by CyTOF in severe COVID-19 patients displaying 393 a higher degranulation potential ( Figure 5G) . 394

Finally, we tested whether C3a is responsible for the altered T cell differentiation potential of 395 serum from patients with severe COVID-19. Neutralisation of C3a reduced CD16 + T cells in 396 most T cell differentiation cultures ( Figure 5H ). 397

In summary, complement split products such as C3a produced at high levels in severe COVID-398

19 generate an inflammatory milieu that promotes differentiation of CD16 + , highly cytotoxic T 399 cells. from COVID-19 and those who survived. We observed significantly higher percentages of 405 activated CD16 + TCRab + cells among all CD4 + and CD8 + T cells in samples from severe 406 COVID-19 patients who died (non-survivor) compared to those who survived (survivor) ( Figure  407 6A, right panel). Furthermore, proportions of CD3 + CD16 + HLA-DR + T cells measured by 408 multicolor flow cytometry ( Figure S4B ) and activated FCGR3A expressing scRNA-seq clusters 409

( Figure S4C ) in samples from cohort 2 showed a trend to be higher in severe COVID-19 410 patients who deceased during follow-up. 411

Next, we tested in a larger cohort whether plasma levels of complement proteins upstream of 412

C3a generation are associated with patient disease course and outcome. Levels of positive 413 regulators of the classical and alternative pathway, such as C1QA, C1QB, C1QC, C1R, and 414 CFD, were higher in plasma samples of patients with severe compared to mild COVID-19 415 ( Figure 6B ). We also analyzed complement protein levels in relation to disease trajectory, 416 specifically subsequent worsening of disease severity. C1R and CFD were elevated in 417 samples from patients who showed a clinical deterioration, whereas the abundance of CFI, 418 which inhibits the classical and lectin-dependent complement pathway, was lower in samples 419 from patients with subsequent disease progression ( Figure 6C ). Finally, amounts of C1QA, 420 C1QB, C1QC and CFD were not only higher in severe COVID-19 but were also associated 421 with fatal outcome ( Figure 6D ). 422

Altogether these data further support the pathological role of complement and of activated Figure S1B&5A ). We identified an elevated and activated T cell population 453 expressing CD16 across the three major T cell compartments. Activated CD16 + T cells showed 454

increased TCR-independent pathogenic potential. 455

The activated CD16 + CD4 + and CD8 + T cells enriched in severe COVID-19 expressed high 456 levels of chemokine receptors such as CXCR3 and CCR6 ( Figure 1D ). can link here to unexpected phenotypic and functional properties. Immunofluorescence co-462 staining of CD3 and CD16 in lung samples of an autopsy cohort showed enrichment of 463 CD3 + CD16 + T cells in COVID-19 compared to influenza pneumonia or other causes of ARDS 464 ( Figure 3I+J ). Although strong T cell activation is a feature of both severe COVID-19 and 465 influenza pneumonia, specific differences have been described between both diseases 466 (Youngs et al., 2021) . 467

We found that approximately 5% of the activated CD16 + CD8 + T cells respond to stimulation 468 with SARS-CoV-2 peptides. This is in line with the previously described positive correlation 469

between ex vivo-determined HLA-DR + CD38 hi Ki67 + CD8 + T cells and SARS-CoV-2-specific 470 CD8 + T cells (Rydyznski Moderbacher et al., 2020). Activated CD16 + T cells show significant 471 enrichment of SARS-CoV-2 specific T cells compared to activated CD16 -T cells ( Figure 5F ). 472

The remaining, non-responding activated CD16 + T cells may recognize other SARS-CoV-2 473 epitopes or may be driven by bystander activation and/or homeostatic proliferation 474 ( As such, the distinct serological profile observed in severe COVID-19 with afucosylated, spike-507 directed IgG, and an inherently increased inflammatory capacity could further enhance the 508 pathogenic potential of CD16 + T cells. 509

In a search for important environmental signals driving differentiation of activated CD16 + T 510 cells, we detected a positive correlation between high serum C3a levels and proportions of 511 CD16 + T cell clusters ( Figure 5E ). It has been reported that serum C3 hyperactivation is a risk 512 factor for COVID-19 mortality (Sinkovits et al., 2021) and widespread complement activation 513 by all three pathways and thus generation of C3a has also been described in patients with 514

severe The limited number of convalescent samples did not allow us to perform correlations with 534 patient recovery. Our major focus was to reveal immunopathogenic functions of severity-535 associated T cell populations during acute COVID-19 and to identify driving signals. 536

In this context, it would be of great interest to see whether application of the C3 inhibitor AMY-537 101 in COVID-19 patients with ARDS will ameliorate differentiation of CD16 + , cytotoxic T cells 538

and thus endothelial cell injury and ultimately improved patient outcome. 539

It is very likely that the role of complement activation in COVID-19 goes beyond the here-540 described mechanisms, which should be investigated in future studies. Furthermore, although 541

C3a neutralisation significantly reduced the CD16 + differentiation potential of severe COVID-542 19 serum, we cannot exclude those other pro-inflammatory mechanisms could induce this 543 activated CD16 + T cell phenotype. Certainly, here also a more detailed comparative 544 investigation of T cell responses between COVID-19 and influenza infection on a larger and 545 more stratified cohort are needed. 546

Also, It will be interesting to investigate whether the described inflammatory circuit is also 547 active in other immune pathologies for which complement activation and immune complex 548

formation have been described. 549

Taken together, particularly severe COVID-19 leads to an elevated number of activated CD16 + 550

T cells that link triggering of the complement cascade via TCR-independent cytotoxic T cell 551 functionality to endothelial damage and patient survival. This functionally links the innate and 552 the adaptive immune system with endothelial injury, which might constitute an important 553 molecular axis explaining the vast spectrum of organ damage observed in COVID-19. 554 J o u r n a l P r e -p r o o f 19 Acknowledgements 555 We thank Katrin Vogt, Christine Appelt, Claudia Conrad, Anja Freiwald, Daniela Ludwig, and 556

Vadim Farztdinov for technical support and Jonas Schulte-Schrepping, Elena de Dominico, 557

Nico Reusch, Kristian Händler, Heidi Theis, Michael Kraut and Kevin Baßler for generating 558 the scRNA-seq dataset of cohort 2. In addition, we thank Franziska Scheibe and Il-Kang Na 559

for providing human antibodies. 560 We also thank Desireé Kunkel and Jacqueline Keye from the BIHFlow and Mass Cytometry 561

Core facility for cell sorting and the BIH / MDC Genomics Platform for sequencing 562

We thank all members of the Pa-COVID-19 collaborative study group ( information included in all reported assays can be found in Table S1 . The proportions of T cells between severity groups were compared using the Wilcoxon test 898

with Benjamini-Hochberg correction. 899 D, Gating of CD3 + CD45 + CD19 -CD15 -T cells and the three T cell compartments for a 900

representative CyTOF data set of cohort 1 prior to clustering, as shown in Figure 1E , Figure  901 S1C. 902 903 Supplemental Figure 5 . Representative plots of gating strategy, related to Figure   904 5A.

A, Gating of CD3 + CD4 + CD16 + and CD3 + CD8 + CD16 + T cells, as shown in Figure 5A 

A, Gating of CD137 + CD69 + of CD8 + CD16 + CD38 + T cells, also discriminationg between CD8 low 918 & CD8 high cells, as shown in Figure 3F . 2019 from patients with acute respiratory distress syndrome ("ARDS") without acute 997 inflammation of the selected ARDS samples (see Table S1 for cohort details). In order to test whether the age of the patients in cohort 1 was confounding the abundance of 1042 the activated CD16 + T cell CyTOF clusters C8 and C26, we did a multivariate model for the 1043 frequencies of these clusters, with severity and age as predictors using the R function lm (stats 1044 package, version 4.0.3). In this analysis, we tested the cluster frequencies based on age and 1045 severity for healthy controls and acute COVID-19 patients selected for the non-weekly analysis 1046

( Figure S1C-D) . 1047

For the survival analysis ( Figure 6A ), we selected mild and severe COVID-19, samples 1048 acquired between 7 and 28 days post-symptom onset. In case of repeated measurements per 1049 patient, the first sample was selected. The proportions of T cells between severity groups were 1050 compared using the Wilcoxon test with Benjamini-Hochberg correction. In order to define an 1051 objective way to select the sum of all activated CD16 + TCRab + clusters of our CyTOF data set 1052 (cohort 1), we computed an activation value for each cluster in the TCRab + space, defined as 1053 the mean of the average z-scored expression of activation markers (CD25, HLA-DR, CD38, 1054 CD137, CD69, and Ki67). Then, clusters with an activation value higher than the average, and 1055

with an average z-scored CD16 expression higher than 1, were considered activated CD16 + 1056

TCRab + T cell clusters (CyTOF cluster 8, 13, 26). 1057 1058 Blood processing and data analysis for multi-color flow cytometry-1059 COVID-19 cohort (cohort 2) 1060 Whole blood was prepared by treatment of 1ml peripheral blood with 10ml of RBC lysis buffer 1061 (Biolegend, USA) . After one wash in DPBS, cells were directly processed for scRNA-seq (BD 1062

Rhapsody) or multi-color flow cytometry (MCFC). After RBC lysis, cells were washed with 1063 DPBS and from each sample 1-2 million cells were stained for flow cytometric analysis (Table  1064 S3). Antibody staining was performed in DPBS with the addition of BD Horizon Brilliant Stain 1065

Buffer (Becton Dickinson, USA) for 30min at 4°C as described before Dickinson, USA) configured with 5 lasers (UV, violet, blue, yellow-green, red). Flow cytometry 1071 data analysis was performed with FlowJo V10.7.1, CD16/HLA-DR double-positive cells were 1072 gated from the total T lymphocytes (living/CD45 + /CD66b -/CD19 -/CD3 + ), STAR Methods, Data 1073

and Code Availability reports the detailed gating strategy and representative plots. With this 1074 analysis we confirmed the existence of a CD3 + CD16 + HLA-DR + T cell population. 1075 1076 Age-dependent control cohorts 1077 We used our flow cytometry data sets (fcs files) described previously (Kverneland et al., 2016) 1078

to report the proportions of CD16 expressing CD4+ and CD8+ T cells. Briefly, whole blood 1079 samples collected from healthy controls spanning an age range between 20 and 84 years with 1080 equal distribution of females and males in each 10-year age bin were stained using the ONE 1081

Study antibody panels as previously described ( Controller for partitioning single-cells into nanoliter-scale Gel Bead-In-Emulsions (GEMs). 1104

The remaining PBMCs were subjected to flow-cytometric sorting based on DAPI , CD3 (clone 1105 UCHT1), CD4 (clone RPA-T4), CD8 (RPA-T8) and CD38 (clone HB7) antibody staining and 1106 J o u r n a l P r e -p r o o f 37 simultaneously hashtagged as described above. We did not include CD45RA or CD45RO as 1107 we did not want to exclude activated naive T cells from our analysis. Live (DAPI-) 1108 CD3 + CD4 + CD38 + , as well as CD3 + CD8 + CD38 + cells, were sorted using the FACS Aria II (BD 1109

Biosciences, USA). Afterwards, CD4 + CD38 + and CD8 + CD38 + T cells from each donor were 1110 pooled equally and CD38 + T cells from up to four samples were pooled to equal proportions. 1111

The resulting cell pool was resuspended in DPBS, filtered through a 40 µm mesh (Flowmi™ 1112

Cell Strainer, Merck, Germany) and counted using the C-Chip hemocytometer (NanoEntek, 1113

South Korea). The cell suspension was super-loaded with 40,000 -50,000 cells per lane, in 1114 the Chromium™ Controller for partitioning single-cells into nanoliter-scale Gel Bead-In-1115

Emulsions (GEMs). 1116

The Chromium Next GEM Single Cell 5′ Kit v1.1 was used for reverse transcription, cDNA 1117 amplification and library construction of the gene expression libraries (10x Genomics, USA). 1118

For additional VDJ and hashtag libraries the Chromium Single Cell V(D)J Enrichment Kit, 1119

Human T Cell (10x Genomics, USA) and the Chromium Single Cell 5' Feature Barcode Library 1120

Kit (10x Genomics, USA) were used respectively. All libraries were prepared following the 1121 detailed protocols provided by 10x Genomics, quantified by Qubit Flex Fluorometer (Thermo 1122

Fisher, USA) and quality was checked using 2100 Bioanalyzer with High Sensitivity DNA kit 1123 (Agilent, USA). Sequencing was performed in paired-end mode with a S1 and S2 flow cell 1124

using To align the CyTOF and scRNA-seq T cell clusters, we applied a feature-based cluster 1172 annotation approach. The selected feature list contained genes for all markers analyzed in 1173

CyTOF measurements plus additional genes such as T cell state-specific transcription 1174 markers (Data and Code Availability). The specific name was assigned according to the Z-1175 score standardized gene expression level of a particular set of genes (e.g. 1176

MKI67 ++ /FCGR3A ++ ) or well-established cellular states (e.g. CD4 naive). To annotate the 1177 clusters, we generated a heatmap ( Figure 2B ) to visualize the expression of selected genes 1178 that were also selected as variable features (in the previous step). We found 20 clusters, and 1179 some of them displayed nearly identical phenotypes and thus were merged, ending up with 1180 17 clusters (Figure 2A and iv) clusters of solely CD8 transcribing T cells (clusters 14 & 15). Cluster 8-10 were 1184 characterized by highly activated T cells with high transcription of CD38 and HLA-DR genes 1185

( Figure 2B , C). Cluster 7 showed the typical features of activated TFH cells, as in addition to 1186 CD38 and HLA-DR genes, they transcribed ICOS, CD40LG, PDCD1 (PD-1) and CXCR5 1187 ( Figure 2B ). Cluster 8 T cells transcribed the highest MKI67 (Ki67) levels, indicative of the 1188 highest proliferative potential ( Figure 2B, C) . In addition, they expressed FCGR3A (CD16A), 1189

which was even more pronounced for cluster 10 T cells ( Figure 2B, C) . 1190

The complete processing, dimensionality reduction, and clustering were performed with the 1191

Seurat The log2-fold change of differentially expressed genes (with high counts, i.e., "baseMean" > 1207 100) from DESeq2 was used to define the ranked gene list used for GSEA. We tested three 1208 signatures (listed in Table S2) Seurat's FindVariableFeatures function and UMAP dimensionality reduction was performed. 1223

In the newly defined cellular space, we identified unwanted cells (non-T cells) according to the 1224 expression of lineage markers. With this approach, we removed NK cells and a small 1225 proportion of monocytes found to contaminate the original lymphocytes cluster. We further 1226

optimized the selection of the T cells space by removing cells that overlap with a recent study 1227 on NK cells, which utilizes the same dataset (Krämer et al., 2021) . The cleaned T cell space 1228 was once more scaled and PCA, as well as UMAP dimensionality reduction, were computed 1229 as described above. Cell clusters were calculated on the first 20 principal components (PCs) 1230

using Figure S3F to the figure legends and the respective experimental procedures in the section "Methods 1501 details''. 1502

The study was not blinded and the sample size was calculated empirically prioritizing the 1503 inclusion of the highest number of COVID-19 samples and matching controls. Tables   1510   Table S1 . Cohort details, related to Figure 1-6. 1511 Table S2 . Gene lists for the signatures "Response to Type I Interferon", "Defense Response 1512

to virus" and "Cytotoxicity" used for GSEA, related to Figure 2 and Figure S3 . 1513 Table S3 . Antibody panel "MCFC cohort 2", related to STAR Methods and Figure S1 . 1514 Table S4 . Antibody panel "degranulation assays", related to Figure 3C -E. 1515 Table S5 . Antibody panel for "C3a binding capacity", related to Figure 5D . 1516 Table S6 . Antibody panel "SARS-CoV-2 reactivity", related to Figure 3F . 

Highlights  Severe COVID-19 is marked by activated, highly cytotoxic CD16 + T cells  Immune complex-mediated degranulation of CD16 + T cells causes endothelial cell injury  C3a-rich environment in severe COVID-19 promotes differentiation of CD16 + T cells  Activated CD16 + T cells and complement proteins are associated with fatal outcome In Brief Generation of the C3a complement protein fragment by SARS-CoV-2 infection drives differentiation of a CD16-expressing T cell population that associates with severe COVID-19 disease outcomes. CD4  CD8A  CD8B  TRAC  TRGC1  TRGC2  SELL  CD28  CD27  ICOS  PDCD1  LAG3  TIGIT  HAVCR2  CTLA4  IL2RA  IL3RA  IL7R  IL2RB 

Pierce 1337 Universal Nuclease, 250U/mL) seeded at 0.25x10 6 /well and rested overnight in a humidified 1338 incubator. Subsequent to overnight rest, PBMC were washed and resuspended in complete 1339 medium

Sodium Pyruvate (Gibco, Thermo Fisher Scientific, USA), 10% v/v heat-inactivated Fetal Calf 1342

MEM Non-essential amino Acid solution

Germany) coated with Biotin conjugated CD16 antibody (clone 1345 3G8, Biolegend, USA), the corresponding isotype control (clone MOPC21, Biolegend, USA) 1346 or MACSiBead-recombinant Spike protein-Serum-immune complexes at a ratio of 1:10 (cell 1347 to particle). Beads were loaded with indicated antibodies according to the protocol provided 1348 by the manufacturer

PBMC were cultured in a humidified incubator (37°C, 5% C02) for 6h in the presence of 1x 1350

final dilution 1:40) was added to 1352 the culture for the whole incubation period. Subsequent to 6h incubation cells were subjected 1353 to surface and intracellular staining. Fc receptor-mediated unspecific binding of antibodies was 1354 blocked by preincubation with human TruStain FcX Fc receptor blocking solution

CD4 1356 BV421 (clone OKT4, Biolegend), CD8 FITC (clone RPA-T8, Biolegend), CD16 BV605 (clone 1357 3G8, Biolegend), CD38-PEcy7 (clone HB7, Biolegend)

Biolegend) for 30 min at 4 C°. The fixable viability dye (ZombieRed, Biolegend) was 1359 incorporated in the surface staining mix. Surface antibody staining was performed in DPBS 1360

Cells were fixed for 20 min at 4°C with BD Cytofix/Cytoperm solution

USA) and intracellularly stained for 30 min at 4°C for Granzyme B APC/Fire 750 1363 (clone QA16A02

5 (clone GM26E7, Biolegend) or TNFα PerCP/Cyanine5.5 (clone Mab11

Intracellular staining was performed in 1x Perm/Wash permeabilization buffer

The antibody panel overview is provided in Table S4. Data analysis was 1367 performed in FlowJo™version 10.6.2 (BD Life Sciences, USA, Fig S6A includes detailed 1368 gating strategy and representative plots), statistical analysis and data visualization was done 1369 in Prism9

Prior 1372 to surface marker staining, non-specific, Fc receptor-mediated staining was blocked by 10 min 1373 pre-incubation at 4°C with human TruStainFcX (Biolegend, USA) blocking solution or Fc 1374 receptor blocking solution

Zombie Red)) was performed for 1377 30 min at 4°C. Binding of complement split product hC3a was tested by incubation of cells for 1378 60 min at 4°C with 50nM synthetic human C3a labelled with AF647 (Almac, UK) subsequent 1379 to fixation and permeabilization with BD Cytofix/Cytoperm TM

Staining with C3a-AF647 was performed in 1x Perm/Wash permeabilization buffer

The antibody panel overview is provided in Table S5. Data analysis was 1382 performed in FlowJo™ version 10.6.2 (BD Life Sciences

Code Availability includes detailed gating strategy and representative plots), statistical 1384 analysis and data visualization was done in Prism9

Serum immune complex generation

Miltenyi Biotec, order # 130-127-685) 1388 at 30µg recombinant protein per 1x10 8 beads at 4°C with repeated resuspension. Subsequent 1389 to washing in PBS, spike coated beads were incubated with human serum derived from a 1390 COVID19 patient or serum from healthy male donors

MACSiBead-rSpike-Serum immune complexes were added at a 1:10 cell to bead ratio. 1393 Detection of SARS-CoV-2-specific T cells

Pierce 1395 Universal Nuclease, 250U/mL) seeded in 96 round well plate and rested for 3

10% v/v heat-inactivated human AB 1399 type Serum

USA) in a humidified 1401 incubator at 37°C. Prior to stimulation, PBMC were preincubated with CD40 blocking antibody 1402 for 15min

-309) was added to cultures at a concentration 1404 of 1µg/mL. After 24h stimulation, cells were washed and subsequently stained with CD3 1405 BV711 (clone UCHT-1

OX40 PE ( clone BER-ACT35, Biolegend), HLA-1407 DR BV785 (clone L243, Biolegend), CD137 BV421 (clone 4B4-1, Biolegend)

CD16 BV605 (clone 3G8, Biolegend), CD45RA PerCP-Cy5.5 (clone HI100

CCR7 Alexa Fluor 700 (clone G043H7, Biolegend) for 30 min at 4°C. The fixable 1411 viability dye (ZombieRed, Biolegend) and FC receptor block human TruStainFcX

USA) were incorporated in the surface staining mix. Finally, cells are fixed for 15 min at 4°C 1413 in 2% PFA in PBS. Surface antibody staining and washing steps are performed in DPBS 1414

The antibody panel overview is provided in Table S6. Data analysis was 1416 performed in FlowJo™ version 10.6.2 (BD Life Sciences, USA, Fig S7A), statistical analysis 1417 and data visualization was done in Prism9

SCENIC: single-cell 1522 regulatory network inference and clustering

1524 destiny: diffusion maps for large-scale single-cell data in R

Reference-based analysis of lung single-cell 1527 sequencing reveals a transitional profibrotic macrophage

Complement receptor CD46 co-stimulates 1530 optimal human CD8+ T cell effector function via fatty acid metabolism

Cellular interactions in the pathogenesis of interstitial 1533 lung diseases

Longitudinal analysis reveals that 1536 delayed bystander CD8+ T cell activation and early immune pathology distinguish severe 1537 COVID-19 from mild disease

Elevated numbers of Fc 1540 gamma RIIIA+ (CD16+) effector CD8 T cells with NK cell-like function in chronic hepatitis C 1541 virus infection

Integrating single-1546 cell transcriptomic data across different conditions, technologies, and species

Association of COVID-19 1550 inflammation with activation of the C5a-C5aR1 axis

Complement cascade in severe 1552 forms of COVID-19: Recent advances in therapy

FcγRIIIa (CD16) induction on human T lymphocytes and CD16pos T-lymphocyte 1555 amplification

T follicular helper cell differentiation, function, and roles in disease

Complement Alternative 1560 and Mannose-Binding Lectin Pathway Activation Is Associated With COVID-19 Mortality

Endothelial cell, myeloid, and 1563 adaptive immune responses in SARS-CoV-2 infection

DIA-NN: 1565 neural networks and interference correction enable deep proteome coverage in high 1566 throughput

A time-resolved proteomic and 1569 prognostic map of COVID-19

STAR: ultrafast universal RNA-seq aligner

The Novel Severe Acute Respiratory Syndrome Coronavirus

Directly Decimates Human Spleens and Lymph Nodes

Characterization of T lymphocytes in 1578 severe COVID-19 patients

Malaria inhibits surface expression of complement receptor 1 in 1581 monocytes/macrophages, causing decreased immune complex internalization

Unique carbohydrate-carbohydrate interactions 1585 are required for high affinity binding between FcgammaRIII and antibodies lacking core 1586 fucose

COVID-19-neutralizing 1589 antibodies predict disease severity and survival

Targeted data extraction of the MS/MS spectra generated by data-1592 independent acquisition: a new concept for consistent and accurate proteome analysis. Mol. 1593 Cell

Tissue-resident CD8+ T cells drive age-associated chronic lung 1596 sequelae after viral pneumonia

Complement activation contributes to 1599 severe acute respiratory syndrome coronavirus pathogenesis

Immunopathology of Hyperinflammation in COVID-19

Complex heatmaps reveal patterns and 1603 correlations in multidimensional genomic data

Normalization and variance stabilization of single-cell 1605 RNA-seq data using regularized negative binomial regression

Integrated analysis of multimodal single-cell data

Complement-Mediated Regulation of Metabolism and 1610 Basic Cellular Processes

High titers and low fucosylation of 1613 early human anti-SARS-CoV-2 IgG promote inflammation by alveolar macrophages

Simultaneous inference in general parametric 1616 models

Cellsnp-lite: an efficient tool for genotyping single cells

On the genetics and immunopathogenesis of COVID-19

Terminally differentiated effector memory 1623

CD8+ T cells identify kidney transplant recipients at high risk of graft failure

iRegulon: from a 1627 gene list to a gene regulatory network using large motif and track collections

phenoptr: inForm Helper Functions

Immune complex clearance by complement receptor type 1 in SLE

Fast, sensitive and accurate 1635 integration of single-cell data with Harmony

Web-based analysis and publication of 1637 flow cytometry experiments

Early IFN-α signatures 1640 and persistent dysfunction are distinguishing features of NK cells in severe COVID-19

Invariant Natural Killer T Cell Subsets-More Than Just 1643 Developmental Intermediates

Studying the 1646 pathophysiology of coronavirus disease 2019: a protocol for the Berlin prospective COVID-1647 19 patient cohort (Pa-COVID-19)

Age and gender leucocytes 1650 variances and references values generated using the standardized ONE-Study protocol

Decay of Fc-dependent antibody 1654 functions after mild to moderate COVID-19

Least-Squares Means: the R package lsmeans

Data-Driven Phenotypic Dissection of 1658 AML Reveals Progenitor-like Cells that Correlate with Prognosis

Highly functional virus-specific 1661 cellular immune response in asymptomatic SARS-CoV-2 infection

Single-cell landscape of bronchoalveolar immune cells in patients with COVID-1664 19

Moderated estimation of fold change 1666 and dispersion for RNA-seq data with DESeq2

Production of 1668 complement components by cells of the immune system

DirichletReg: Dirichlet Regression for Compositional Data in R (Institute for 1670 Statistics and Mathematics

Cutadapt removes adapter sequences from high-throughput sequencing 1672 reads

The first case of 1675 COVID-19 treated with the complement C3 inhibitor AMY-101

Complement C3 vs C5 inhibition in severe COVID-19: Early clinical findings reveal 1679 differential biological efficacy

Deep immune profiling of 1682 COVID-19 patients reveals distinct immunotypes with therapeutic implications

Increased complement activation is a 1685 distinctive feature of severe SARS-CoV-2 infection

The Innate Immune System: Fighting on the Front 1687 Lines or Fanning the Flames of COVID-19?

Immunopathogenesis of SARS-CoV-2-induced pneumonia: lessons 1692 from influenza virus infection

Comprehensive profiling of an 1695 aging immune system reveals clonal GZMK+ CD8+ T cells as conserved hallmark of 1696 inflammaging

NK cells at the 1698 interface between innate and adaptive immunity

COVID-19 in older adults

Overview of complement activation and regulation

CyTOF workflow: differential discovery in high-1705 throughput high-dimensional cytometry datasets

Trigger-happy resident 1709 memory CD4+ T cells inhabit the human lungs

Broad and strong memory CD4+ and CD8+ T cells induced 1712 by SARS-CoV-2 in UK convalescent individuals following COVID-19

Rationale for targeting complement in COVID-19

Extracting a cellular hierarchy from high-1719 dimensional cytometry data with SPADE

Unspecific post-mortem findings 1722 despite multiorgan viral spread in COVID-19 patients

The spatial landscape of lung pathology during 1725 COVID-19 progression

Antigen-Specific Adaptive 1728 Immunity to SARS-CoV-2 in Acute COVID-19 and Associations with Age and Disease 1729 Severity

Activation of classical and alternative complement pathways in the pathogenesis of lung 1732 injury in COVID-19

Regulatory cell therapy in kidney 1735 transplantation (The ONE Study): a harmonised design and analysis of seven non-1736 randomised, single-arm, phase 1/2A trials

Disease Severity, Fever, Age, and Sex Correlate 1739 With SARS-CoV-2 Neutralizing Antibody Responses

Severe COVID-19 Is 1742 Marked by a Dysregulated Myeloid Cell Compartment

Minimizing batch effects in mass cytometry 1745 data

Adaptive immunity to SARS-CoV-2 and COVID-19

Pulmonary Mycobacterium 1750 tuberculosis control associates with CXCR3-and CCR6-expressing antigen-specific Th1 and 1751 Th17 cell recruitment

Cytoscape: a software environment for integrated 1754 models of biomolecular interaction networks

Extracellular granzyme K mediates endothelial activation through the 1757 cleavage of protease-activated receptor-1

Alterations in T and B cell function persist 1760 in convalescent COVID-19 patients

Complement Overactivation and Consumption 1763 Predicts In-Hospital Mortality in SARS-CoV-2 Infection

Syntaxin 4 mediates endosome recycling for lytic granule exocytosis in cytotoxic T-1766 lymphocytes

Single-cell multi-omics 1769 analysis of the immune response in COVID-19

SARS-CoV-2 RNA screening 1772 in routine pathology specimens

Standardization of whole 1775 blood immune phenotype monitoring for clinical trials: panels and methods from the ONE 1776 study

Comprehensive Integration of Single-Cell 1779 Data

Gene set enrichment 1782 analysis: a knowledge-based approach for interpreting genome-wide expression profiles

Early induction of functional SARS-CoV-1786 2-specific T cells associates with rapid viral clearance and mild disease in COVID-19 1787 patients

Robust T Cell Response 1790 Toward Spike, Membrane, and Nucleocapsid SARS-CoV-2 Proteins Is Not Associated with 1791 Recovery in Critical COVID-19 Patients

Dysregulated Host Response in Severe Acute Respiratory Syndrome 1794 Coronavirus 2-Induced Critical Illness

Immunology of COVID-19: Current State of the 1797 Science

Endothelial cell 1800 infection and endotheliitis in COVID-19

Functions of natural 1802 killer cells

Discriminating mild from 1805 critical COVID-19 by innate and adaptive immune single-cell profiling of bronchoalveolar 1806 lavages

Complement and the regulation of T cell 1808 responses

Multisystemic Cellular Tropism of SARS-CoV-2 in Autopsies of COVID-19 Patients

SARS-CoV-2 drives JAK1/2-dependent local 1814 complement hyperactivation

Identification of immune correlates of fatal 1817 outcomes in critically ill COVID-19 patients

clusterProfiler: an R package for 1819 comparing biological themes among gene clusters

Antibody-dependent cellular cytotoxicity response to SARS-CoV-2 in COVID-19 1822 patients

T Cell Activation, Highly Armed Cytotoxic Cells and a Shift in Monocytes CD300 1826

Receptors Expression Is Characteristic of Patients With Severe COVID-19

Single-cell landscape of immunological responses in patients 1830 with COVID-19

Human pulmonary microvascular endothelial cells (HPMECs, Promocell, Germany, passage 1420 4-8) were plated on 96-well plates (96W10idf PET, Applied Biophysics Inc., USA) and grown 1421to confluency for 48-72 h. Cell impedance was quantified by electric cell impedance sensing 1422(ECIS® Z-Theta Applied Biophysics Inc., USA) at 4000Hz every 60 sec. For co-cultivation 1423 experiments, media was replaced by Opti-MEM TM (Gibco, Thermo Fisher Scientific, USA). 1424After 1 h stabilization phase, Convanacalin A (10 µM) from Canavalia ensiformis (Jack bean, 1425Sigma-Aldrich Chemie GmbH, Munich, Germany, anti-CD16 beads (5 beads / T cell), and 1426 20,000 flow-sorted, non-naive CD8 + T cells were added and cell barrier integrity was 1427 monitored continuously for 24 h. Resistance was normalized for each individual well to the 1428 baseline before treatment. Statistical analysis and data visualization was done in Prism9. 1429