key: cord-0976127-ujf07dh3
authors: Asashima, H.; Mohanty, S.; Comi, M.; Ruff, W. E.; Hoehn, K. B.; Wong, P.; Cohen, I.; Coffey, S.; Raddassi, K.; Chaudhary, O.; Unterman, A.; Emu, B.; Kleinstein, S. H.; Montgomery, R. R.; Iwasaki, A.; Dela Cruz, C. S.; Kaminski, N.; Shaw, A. C.; Hafler, D. A.; Sumida, T. S.
title: PD-1highCXCR5-CD4+ Peripheral Helper T (Tph) cells Promote Tissue-Homing Plasmablasts in COVID-19
date: 2021-03-17
journal: nan
DOI: 10.1101/2021.03.13.21253527
sha: 05fbbda2b0089de812615e7f47ea83f71affecbb
doc_id: 976127
cord_uid: ujf07dh3

A dysregulated immune response against coronavirus-2 (SARS-CoV-2) plays a critical role in the outcome of patients with coronavirus disease 2019 (COVID-19). A significant increase in circulating plasmablasts is characteristic of COVID-19 though the underlying mechanisms and its prognostic implications are not known. Here, we demonstrate that in the acute phase of COVID-19, activated PD-1highCXCR5-CD4+ T cells, peripheral helper T cells, (Tph) are significantly increased and promote inflammatory tissue-homing plasmablasts in patients with stable but not severe COVID-19. Analysis of scRNA-seq data revealed that plasmablasts in stable patients express higher levels of tissue-homing receptors including CXCR3. The increased Tph cells exhibited "B cell help" signatures similar to that of circulating T follicular helper (cTfh) cells and promoted B cell differentiation in vitro. Compared with cTfh cells, Tph cells produced more IFN{gamma}, inducing tissue-homing chemokine receptors on plasmablasts. Finally, expansion of activated Tph cells was correlated with the frequency of CXCR3+ plasmablasts in the acute phase of patients with stable disease. Our results demonstrate a novel role for Tph cells in acute viral immunity by inducing ectopic, antibody secreting plasmablasts.

To examine the mechanism of T cell regulation of plasmablast differentiation in COVID- 79 19, we investigated the characteristics of B and T cells in patients by single cell RNA-seq 80 (scRNA-seq) and flow cytometry datasets. Here, we report that PD-1 high CXCR5 -CD4 + T cells, 81 -4 -so called peripheral helper T (Tph) cells 19, 20 , are significantly increased and positively 82 correlated with the frequency of plasmablasts in peripheral blood in patients with COVID-19. 83 These Tph cells exhibit "B cell help" signatures to a similar degree as cTfh cells, but also 84 express more inflammatory chemokine receptors including CCR2 and CCR5. In vitro 85 experiments indicate that PD-1 high CXCR5 -Tph cells have higher IFNγ production, which 86 promotes CXCR3 expression and differentiation of plasmablasts. Finally, we demonstrate that 87 CXCR3 + tissue-homing plasmablasts are significantly increased in patients with stable 88 COVID-19 while they are decreased in patients with severe disease. These findings provide 89 a mechanism for the increase of plasmablasts apart from Tfh cells in the acute phase of 

Expanded plasmablasts express higher tissue-homing molecules in patients with 95 stable COVID-19 96 We first confirmed that the frequency of plasmablasts is significantly increased in the is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint PD-1 high CXCR5 -Tph cells promote B cell differentiation, and produce more IFNγ than 164 PD-1 high CXCR5 + Tfh cells 165 We examined whether PD-1 high CXCR5 -Tph cells promote B cell differentiation in 166 vitro. As with PD-1 high CXCR5 + Tfh cells, these cells induce memory B cells into plasmablasts, 167 and initiated IgG production (Figure 3a, 3b). Stimulation with anti-CD3/CD28 antibodies 168 induced greater IFNγ and IL-10 production from PD-1 high CXCR5 -Tph cells (Figure 3c ), which 169 is of interest as IFNγ is known to upregulate CXCR3 expression during B cell differentiation 33 . 170 We observed that CXCR3 expression on plasmablasts was upregulated in a dose dependent 171 manner by IFNγ while IL-10 was not (Figure 3d, 3e) . Moreover, besides CXCR3, other 172 inflammatory tissue-homing receptors such as CCR2 were upregulated, but CXCR4 was not 173 ( Figure 3f ). These data indicate that PD-1 high CXCR5 -Tph cells promote B cell differentiation 174 and have a capacity to induce plasmablasts which express homing-molecules for 175 inflammatory tissues via IFNγ production.

The proportion of activated PD-1 high CXCR5 -Tph cells are significantly increased in 177 stable as compared to severely ill patients with COVID-19 and are positively 178 correlated with CXCR3 + plasmablasts 179 To investigate the characteristics of PD-1 high CXCR5 -Tph cells in vivo, we evaluated the 180 specific gene signatures of this subset (Supplemental Table 4 ). Compared to the other five 181 subsets in memory CD4 + T cells, CXCR6, LAG3, and PRR5L were significantly upregulated, 182 while CHD7, ZBTB20, ZNF251, GRK25, and GPRASP1 were significantly downregulated in 183 PD-1 high CXCR5 -Tph cells (Figure 4a, 4b) . Flow cytometry analysis also showed the same 184 trend of upregulation of LAG3 and CXCR6 expressions (Supplemental Figure 5a, 5b) . 185 Embedding these gene lists with CD4 + T cell clusters from our scRNA-seq data set 5 , we found 186 that dividing CD4 + T cells best fit with these signatures (Figure 4c ). We previously reported 187 that dividing CD4 + T cells share the characteristics of HLA-DR + CD38 + activated T cells 5 , and 188 indeed, more than half of the activated CD4 + T cells were in the PD-1 high CXCR5 -Tph cell is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 17, 2021. ; https://doi.org/10.1101/2021.03.13.21253527 doi: medRxiv preprint -9 -highly overlapping signatures with dividing CD4 + T cells that are positively correlated with the 237 proportion of Ki67 + plasmablasts 40 , which had lower SHM frequencies in our scRNA-seq data. 238 These findings suggest that circulating PD-1 high CXCR5 -Tph cells may be the counterpart of T is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Table 1 Table 2 ). All the experiments were 291 performed on fresh peripheral blood mononuclear cells (PBMCs), and samples were drawn 292 on the average of 11.8 days after first symptoms. 92 patients with COVID-19 and 64 COVID-293 19-uninfected healthcare workers (HCs) were enrolled. COVID-19 patients who required 294 admission to the ICU had been classified as "progressive" and 27.8% of them were expired. 295 The other patients classified as "stable" were all discharged without ICU admission. For the 296 patients who are 90 years-old or older, their ages were protected health information, and '90' 297 was put as the surrogate value for the analyses. HCs were all negative in both PCR and 298 serology tests. 299 Single cell RNA-seq (scRNA-seq) was performed on cryopreserved PBMC samples of 300 10 patients with COVID-19 following the same criteria as above and 13 age-and sex-matched 301 controls. All the samples were drawn on the average of 11.7 days after first symptoms. Control 302 samples were already collected before the first report of COVID-19 in 2018. From eight of ten 303 patients with COVID-19, PBMC samples from two different time points had been analyzed. 304 Four patients had been classified as "progressive", who required admission to the ICU, and 305 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 17, 2021. ; https://doi.org/10.1101/2021.03.13.21253527 doi: medRxiv preprint the other six patients classified as "stable" who were hospitalized and all discharged, and the 306 same criteria as flow data. We have described the full cohort elsewhere 5 . 307 All the other experiments which include in vitro experiments and bulk RNA-seq were 308 performed with fresh PBMCs at the baseline. All the patients were admitted between 21 st July 309 and 22 nd Jan 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint identification and cell type annotation. We have described the detailed methods elsewhere 5 . 384 The annotated R object was used for sub-clustering of B cells. 385 The three cell populations, "Memory B cells", "Naïve B cells " and "Plasma cells" in 386 total PBMCs were re-clustered to obtain a finer cell type granularity. To remove batch-and 387 single-donor effects, we integrated all 31 samples of these populations into one dataset using is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint (a, b) . c, The proportion of PD-1 high CXCR5 -Tph cells between male (n=45) 572 and female (n=47) COVID-19 patients were evaluated by two-tailed unpaired Student's t-test. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint We would like to thank all the hospital staff who helped care for the patients and obtain 620 samples. We are also grateful to all the members of YALE IMPACT research team who 621 obtained data. We also thank Yale Environmental Health and Safety (EHS) office, is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 17, 2021. ; https://doi.org/10.1101/2021.03.13.21253527 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 17, 2021. ; https://doi.org/10.1101/2021.03.13.21253527 doi: medRxiv preprint TIGIT  SH2D1A  ICOS  FYN  IL21  TOX2  PDCD1  SLAMF6  CD200  BTLA  MAF  CXCR6  CXCL13  IL6ST  CDK5R1  CTSB  CXCR5  NFATC1  IL6R PDE1B  TLR3  KLHL34  CDK5R1  PAWR  ZNF845  CXCR5  ZNF251  SULF2  CENPV  AP004609.1  KIF3B  ZBTB40  PRSS36  MAGI3  SMCR8  DIRC2  ZBTB20  WWC2  GRK5  ATAD2B  ZNF398  ARL13B  PASK  PAPOLG  ACTN1  IPCEF1  MMS22L  KIF21B  CHD7  CXCL8  NR2C1  RBM15   ANXA1  TIMP1  GLIPR2   CEP78  FZD6  APOBEC3H  SAMD3  CAPG  C12orf75  ACTN4  LGALS3  LGALS1  PTPRM  LAG3  DOPEY2  MCM9  CTSH  F5  PAM  TSPAN32  ID2  CYTH3  FAM84B  CHN1  VCL  PRR5L  ALKBH8  TNFRSF18  FAM129A  CX3CR1  LAPTM4B  CXCR6  DPP4  RRM2  GZMH  RARG  CSGALNACT1  LINC002481  SPDL1  UGGT1  HBA1  CPT1B  CCR5  HBB  PLEKHG3  EDARADD  SYTL3  HDAC9  HOPX  CCR2  NEFL  PI16  SERAC1  HSD11B1  TRMT5  SPAG5  ENPP2  SCRN1  CCT6B  SPIN3  GALNT8  CD274  ADGRG5 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The proportions of activated PD-1 high CXCR5 -Tph cells were evaluated (stable; n= 56, progressive; n= 36) by two-tailed unpaired Student' s t-test (down). f, Correlation between activated PD-1 high CXCR5 -Tph cells (percentage of CD3 + CD4 + CD45RAmemory T cells) and CXCR3 + plasmablasts (percentage of CD19 + CD27 + CD38 + plasmablasts) (both stable and progressive, n=51). Linear regression is shown with 95% confidence interval (gray area). Correlation statistics is two-tailed Spearman' s rank correlation test. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint C27  C29  C30  C31  C32  C33  C34  C35  C36  C37  C38  C39  C40  NS0A  NS0B  NS1A   NS1B  TS2A  TS2B  TS3A  TS3B  TS4A  TS4B  TS5A  TS5B  TP6A  TP6B  TP7A  TP7B  TP8B  TP9B   1  3 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Supplemental Figure 3 The characteristics of PD-1 high CXCR5 -Tph cells. a-b, Correlation between the proportion of PD-1 high CXCR5 -Tph cells (percentage of CD3 + CD4 + CD45RAmemory T cells) and each clinical background in COVID-19 patients (both stable and progressive, n=92) (a, age; b, BMI). Linear regression is shown with 95% confidence interval (gray area). Correlation statistics by two-tailed Spearman' s rank correlation test (a, b) . c, The proportion of PD-1 high CXCR5 -Tph cells between male (n=45) and female (n=47) COVID-19 patients were evaluated by two-tailed unpaired Student' s t-test. d, Representative flow data of CX3CR1 expression on PD-1 high CXCR5 -Tph cells compared with PD-1 high CXCR5 + Tfh cells. e, Heatmap of T cell lineage genes among six subsets of memory CD4 + T cells. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint , and COVID (n=109; all at baseline samples) groups. One-way ANOVA with Dunn' s multiple comparisons tests were performed to evaluate differences. c, Correlation between PD-1 high CXCR5 -Tph cells (percentage of CD3 + CD4 + CD45RAnon-naive T cells) and plasmablasts (percentage of CD19 + B cells) in COVID-19 patients (n=109). Linear regression is shown with 95% confidence interval (gray area). Correlation statistics is two-tailed Spearman' s rank correlation test. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint COVID-19 patients) . c-d, Correlation between the proportion of activated HLA-DR + CD38 + PD-1 high CXCR5 -Tph cells (percentage of PD-1 high CXCR5 -Tph cells) and each clinical background in COVID-19 patients (both stable and progressive, n=92)(c, age; d, BMI). Linear regression is shown with 95% confidence interval (gray area). Correlation statistics is two-tailed Spearman' s rank correlation test (c, d) . e, The proportion of activated HLA-DR + CD38 + PD-1 high CXCR5 -Tph cells between male (n=45) and female (n=47) COVID-19 patients were evaluated by two-tailed unpaired Student' s t-test. f-g, Longitudinal frequencies of PD-1 high CXCR5 -Tph cells (f) and activated PD-1 high CXCR5 -Tph cells (g) after hospitalization. Only the samples which could follow blood collection (hospitalization, week1 of day1-7, week2 of day8-14) were analyzed (Stable n=23, Progressive n=16). At each time point, Two-tailed unpaired Student' s t-test were performed (*p<0.05). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

Under healthy conditions, we can detect few plasmablasts and PD-1 high CXCR5 -Tph cells. In the acute phase, PD-1 high CXCR5 -Tph cells increased in COVID-19 patients and are related to the promotion of plasmablasts. Among patients with COVID-19, stable groups can increase activated PD-1 high CXCR5 -Tph cells quickly, which can produce IFNγ more than PD-1 high CXCR5 + Tfh cells, and promote inflammatory-tissue homing plasmablasts at the proper timing. Progressive COVID-19 patients can also increase activated PD-1 high CXCR5 -Tph cells but delayed, which lead to insufficient promotion of plasmablasts at proper timing and worse clinical outcome.

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 17, 2021. ; https://doi.org/10.1101/2021.03.13.21253527 doi: medRxiv preprint

Characteristics of and Important Lessons from the 671

COVID-19) Outbreak in China: Summary of a Report of 672 72314 Cases from the Chinese Center for Disease Control and Prevention

A Novel Coronavirus from Patients with Pneumonia in China

Sex differences in immune responses that underlie COVID-19 677 disease outcomes

Complex Immune Dysregulation in COVID-19

Patients with Severe Respiratory Failure

Single-Cell Omics Reveals Dyssynchrony of the Innate and 681

Adaptive Immune System in Progressive

Deep immune profiling of COVID-19 patients reveals patient 684 heterogeneity and distinct immunotypes with implications for therapeutic interventions

Immunological imprint of COVID-19 on human peripheral blood 687 leukocyte populations

Single-cell landscape of bronchoalveolar immune cells in patients with 689 COVID-19

Human Ebola virus infection results in substantial immune 691 activation

Rapid and Massive Virus-Specific Plasmablast Responses during 693

Acute Dengue Virus Infection in Humans

Leukocyte trafficking to the lungs and beyond: lessons from influenza 695 for COVID-19

Plasma-cell homing

The generation of 699 antibody-secreting plasma cells

T follicular helper cells in human autoimmunity

Helper Cell Biology : A Decade of Discovery and 703 Diseases

Human blood CXCR5(+)CD4(+) T cells are counterparts of T follicular -26 -cells and contain specific subsets that differentially support antibody secretion

Humoral and circulating follicular helper T cell responses in 708 recovered patients with COVID-19

Loss of Bcl-6-Expressing T Follicular Helper Cells and Germinal 710 Centers in COVID-19

Pathologically expanded peripheral T helper cell subset drives B cells 712 in rheumatoid arthritis

Shared and distinct roles of T peripheral helper and T 714 follicular helper cells in human diseases

Longitudinal analyses reveal immunological misfiring in severe 717 COVID-19

Single-cell landscape of immunological responses in patients with 719 COVID-19

CD9 Is a Unique Marker for Marginal Zone B Cells

Cells, and Plasma Cells in Mice

CD9 is a novel marker for plasma cell precursors in human germinal 723 centers

B-cell memory in malaria: 728 Myths and realities

Distinct Effector B Cells Induced by Unregulated Toll-like Receptor 730

Contribute to Pathogenic Responses in Systemic Lupus Erythematosus

Discrete populations of isotype-switched memory B lymphocytes are 733 maintained in murine spleen and bone marrow

Challenges and Opportunities for Consistent Classification of Human B 735 Cell and Plasma Cell Populations

Extrafollicular 737 responses in humans and SLE

An Integrated Multi-omic Single-Cell Atlas of Human B Cell Identity

B cell subset alteration and the expression of tissue homing -27 -molecules in dengue infected patients

Regulation of CXCR3 and CXCR4 expression during terminal 743 differentiation of memory B cells into plasma cells

PD-1hiCXCR5-T peripheral helper cells promote B cell 745 responses in lupus via MAF and IL-21

The type I IFN system in rheumatoid arthritis

Type I interferon in rheumatic diseases

The cytokine TGF-β 2 co-opts signaling via STAT3-STAT4 to 751 promote the differentiation of human TFHcells

Type I Interferon Transcriptional Network Regulates Expression of 753

Coinhibitory Receptors in Human T cells. bioRxiv : the preprint server for biology

Extrafollicular B cell responses correlate with neutralizing 756 antibodies and morbidity in COVID-19

Post-infectious inflammatory disease in MIS-C features 758 elevated cytotoxicity signatures and autoreactivity that correlates with severity. 759 medRxiv : the preprint server for health sciences

The Multifaceted B Cell Response to Influenza Virus

Salmonella induces a switched antibody response without 767 germinal centers that impedes the extracellular spread of infection

Kinetics of antibody responses dictate COVID-19 outcome

A comprehensive, longitudinal analysis of humoral responses specific 772 to four recombinant antigens of SARS-CoV-2 in severe and non-severe COVID-19 773 patients

A bioinformatic approach to 778 investigating cytokine genes and their receptor variants in relation to COVID-19 779 progression

Effective chemical virus inactivation of patient serum compatible 781 with accurate serodiagnosis of infections

Integrating single-cell 784 transcriptomic data across different conditions, technologies, and species

Comprehensive Integration of Single-Cell Data

comprehensive 789 database for human and mouse immunoglobulin and T cell receptor genes. Nucleic 790 Acids Res

IgBLAST: an immunoglobulin variable 792 domain sequence analysis tool

Change-O: a toolkit for analyzing large-scale B cell immunoglobulin 794 repertoire sequencing data

Models of somatic hypermutation targeting and substitution based on 796 synonymous mutations from high-throughput immunoglobulin sequencing data

Human B Cell Clonal Expansion and Convergent Antibody 799 Responses to SARS-CoV-2

PRINSEQ++, a multi-threaded tool for fast 801 and efficient quality control and preprocessing of sequencing datasets

STAR: ultrafast universal RNA-seq aligner

RSEM: accurate transcript quantification from RNA-Seq data 806 with or without a reference genome

Moderated estimation of fold change and