Identification of soluble biomarkers that associate with distinct manifestations of long COVID

Categories: Disease & Virus

May 1, 2025

Study design

The objective of this study was to characterize the immunological and proteomic features of long COVID. SARS-CoV-2 spike-specific antibody titers were measured using an enzyme-linked immunosorbent assay. Antibody neutralization activity and ADNKA were quantified against the England-2 strain of SARS-CoV-2. Immune cell lineages were profiled via multidimensional flow cytometry. Antigen-specific CD4⁺ and CD8⁺ T cells were enumerated functionally using a flow cytometric AIM assay. Antigen-specific CD8⁺ T cells were further identified physically using peptide–HLA class I tetramers to enable detailed phenotypic analyses via multidimensional flow cytometry. Plasma proteomes were analyzed using a targeted affinity platform. Clinical symptoms were integrated with the frequencies and phenotypic attributes of immune cells to delineate plasma biomarkers and signaling pathways associated with long COVID.

Donors

The primary cohort included healthy convalescent individuals (controls; n = 70) and individuals with long COVID (cases; n = 70) recruited from University Hospital Llandough (Table 1 and Supplementary Table 1). All participants had a clearly defined episode of symptomatically mild acute COVID-19 confirmed via direct molecular evidence of infection with SARS-CoV-2. None required hospitalization. Cases were diagnosed according to the National Institute for Health and Care Excellence guideline NG188 (https://www.nice.org.uk/guidance/ng188). Groups were matched as closely as possible for age, BMI, race, sex, time since infection, and vaccination against SARS-CoV-2 (Fig. 1a,b and Table 1). Eligible individuals were men and nonpregnant women over the age of 18 years with no alternative explanatory disease and symptoms that persisted for at least 12 weeks after the initial diagnosis of acute COVID-19. One persistent symptom was sufficient for the diagnosis of long COVID. All individuals underwent a comprehensive medical evaluation, including chest radiography, electrocardiography, lung function tests (spirometry with gas transfer as indicated and measurement of exhaled nitric oxide), and standard blood tests (autoantibody screens; bone, liver and kidney function; coagulation screens; full blood count; markers of nutrition). Symptoms were scored individually using a numeric self-rating scale from 0 (no symptom) to 10 (worst possible symptom). Overall general health was scored similarly on an inverse scale from 0 (worst possible) to 10 (best possible). The secondary cohort included healthy convalescent individuals (controls; n = 30) and individuals with long COVID (cases; n = 95) recruited from the Karolinska University Hospital (Table 2). All participants in the primary cohort were recruited between March and August 2022, and all participants in the secondary cohort were recruited between June and October 2022. PBMCs from donors with untreated chronic HIV-1 infection (n = 14) were obtained from the University of Alabama at Birmingham or the University of California, San Francisco.

Samples

PBMCs were isolated via standard density gradient centrifugation and cryopreserved in fetal bovine serum (Thermo Fisher Scientific) containing 10% dimethyl sulfoxide (DMSO; Sigma-Aldrich). EDTA plasma samples were stored at −80 °C.

Ethics

All participants provided written informed consent in accordance with the principles of the Declaration of Helsinki (2013). The primary study was approved by the Cardiff University School of Medicine Research Ethics Committee (21/55) and the Health Research Authority and Health and Care Research Wales (20/NW/0240), and the secondary study was approved by the Swedish Ethical Review Authority (2022-00100-01).

Cells and viruses

A549 and VeroE6 cells expressing human angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2) were used to support viral entry and propagation⁶⁷. Antibody functionality assays were performed using the England-2 strain of SARS-CoV-2 (ref. ³⁸).

Peptides

SARS-CoV-2 peptides were manufactured as 15-mers overlapping by 11 amino acids spanning the spike protein (Peptides & Elephants) or as 20-mers overlapping by 10 amino acids spanning the nucleocapsid, combined membrane and envelope, ORF1a, ORF1b and ORF3–ORF10 proteins (Sigma-Aldrich). EBV peptides were manufactured as 15-mers overlapping by 11 amino acids spanning the BRLF1, BZLF1, BMLF1 and BARF1 proteins (lytic pool) and the EBNA1, EBNA2, EBNA3A, EBNA3B, EBNA3C and LMP2 proteins (latent pool; JPT Peptide Technologies). CMV peptides were manufactured as 15-mers overlapping by 11 amino acids spanning the combined IE-1, IE-2 and pp65 proteins (JPT Peptide Technologies). Lyophilized peptides were reconstituted at a stock concentration of 10 mg ml^–1 in DMSO and further diluted to 100 μg ml^–1 in phosphate-buffered saline (PBS).

Tetramers

Peptide–HLA class I complexes were generated and tetramerized with fluorescent tags as described previously^68,69. The following specificities were used in this study: CMV pp65 HLA-A*02:01 NLVPMVATV (BV421), CMV pp65 HLA-B*07:02 TPRVTGGGAM (PE), EBV BMLF1 (lytic) HLA-A*02:01 GLCTLVAML (PE), EBV EBNA3A (latent) HLA-B*07:02 RPPIFIRRL (BV421), HIV-1 p2p7p1p6 Gag HLA-A*02:01 FLGKIWPSHK (PE), HIV-1 p17 Gag HLA-A*02:01 SLYNTVATL (BV421), HIV-1 Pol HLA-A*02:01 ILKEPVHGV (PE), HIV-1 p17 Gag HLA-A*24:02 KYKLHIVW (BV421), HIV-1 Nef HLA-A*24:02 RYPLTFGW (PE), HIV-1 p24 Gag HLA-B*07:02 GPGHKARVL (BV421), HIV-1 p24 Gag HLA-B*08:01 EIYKRWII (PE), HIV-1 p24 Gag HLA-B*57:01 KAFSPEVIPMF (PE), HIV-1 p24 Gag HLA-B*57:01 QASQEVKNW (BV421), IAV matrix protein M1 HLA-A*02:01 GILGFVFTL (BV421), IAV nucleoprotein HLA-B*07:02 LPFDKTTVM (BV421), SARS-CoV-2 spike HLA-A*02:01 YLQPRTFLL (BV421), SARS-CoV-2 nucleocapsid HLA-A*02:01 LLLDRLNQL (PE), SARS-CoV-2 ORF3 HLA-A*02:01 ALSKGVHFV (PE), SARS-CoV-2 ORF3 HLA-A*02:01 LLYDANYFL (PE) and SARS-CoV-2 nucleocapsid HLA-B*07:02 SPRWYFYYL (PE).

Antibody quantification

SARS-CoV-2 spike-specific antibody titers were measured using a SARS-CoV-2 Spike (Trimer) Ig Total ELISA Kit (Thermo Fisher Scientific). Samples were assayed in duplicate and calibrated against a standard curve. Data were analyzed using Prism version 9.5.0 (GraphPad).

Neutralization assay

Antibody neutralization activity was quantified as described previously³⁸. Briefly, serial dilutions of plasma were mixed in duplicate with 600 plaque-forming units of England-2, incubated for 1 h at 37 °C, and added to VeroE6 cells expressing ACE2 and TMPRSS2. After 48 h, cell monolayers were fixed in 4% paraformaldehyde (Thermo Fisher Scientific), permeabilized with 0.5% NP-40 (Merck), and blocked with PBS containing 0.1% Tween-20 (PBST) and 3% nonfat milk for 1 h at room temperature (RT). The primary antibody (anti-SARS-CoV-2 nucleocapsid protein, clone 1C7, Stratech Scientific) was diluted 1:500 in PBST containing 1% nonfat milk and added to the cell monolayers for 1 h at RT. Cells were then washed with PBST. The secondary antibody (anti-mouse IgG-HRP, polyclonal, Jackson ImmunoResearch) was diluted 1:3,000 in PBST containing 1% nonfat milk and added to the cell monolayers for 1 h at RT. Cells were then washed again with PBST. Assays were developed using SIGMAFAST OPD (Sigma-Aldrich) and analyzed at an optical density of 450 nm using a CLARIOstar Plus Microplate Reader (BMG Labtech). Control wells contained no sample, a standardized sample with moderate neutralization activity, or no SARS-CoV-2. The neutralization titer for each sample was calculated as the highest plasma dilution that achieved a 50% reduction in plaque formation (NT₅₀).

ADNKA

ADNKA was quantified as described previously^38,70. Briefly, target A549 cells expressing ACE2 and TMPRSS2 were infected overnight with England-2 (multiplicity of infection = 5), collected using TrypLE Express Enzyme (Thermo Fisher Scientific), mixed with healthy donor PBMCs at a ratio of 1:10, and incubated with serial dilutions of plasma in the presence of anti-CD107a–FITC (clone H4A3, BioLegend) and GolgiStop (0.7 μl ml^–1; BD Biosciences) for 5 h at 37 °C. Cells were then washed with cold PBS, stained with anti-CD3–PE-Cy7 (clone UCHT1, BioLegend), anti-CD56–BV605 (clone 5.1H11, BioLegend), anti-CD57–APC (clone HNK-1, BioLegend) and LIVE/DEAD Fixable Aqua (Thermo Fisher Scientific) for 30 min at 4 °C, washed again with cold PBS, and fixed in 4% paraformaldehyde (Thermo Fisher Scientific). Control wells contained a seronegative sample, uninfected target cells, or a standardized sample that elicited moderate ADNKA. Data were acquired using an Attune NxT Flow Cytometer (Thermo Fisher Scientific). Activation was quantified as a function of degranulation (CD107a⁺) among viable NK cells (Aqua⁻CD3⁻CD56⁺) with potent cytotoxic activity (CD57⁺) using FlowJo version 10.9.0 (FlowJo) and normalized to the standardized sample via area under the curve analyses in Prism version 9.5.0 (GraphPad).

Immune cell lineage analysis

PBMCs were thawed quickly, resuspended in RPMI 1640 Complete Medium (Sigma-Aldrich) supplemented with DNase I (10 U ml^–1; Sigma-Aldrich), and seeded at 1 × 10⁶ cells per well in 96-well U-bottom plates (Corning). Cells were incubated first with Human TruStain FcX (BioLegend) for 10 min at RT and then with LIVE/DEAD Fixable Aqua (Thermo Fisher Scientific) for 10 min at RT. Anti-CCR7–APC-Cy7 (clone G043H7, BioLegend) and anti-CX3CR1–PE (clone 2A9-1, BioLegend) were added for 15 min at 37 °C. Cells were then stained with anti-CD3–BV650 (clone OKT3, BioLegend), anti-CD4–PE-Cy5.5 (clone S3.5, Thermo Fisher Scientific), anti-CD8–BUV396 (clone RPA-T8, BD Biosciences), anti-CD11c–BB515 (clone B-ly6, BD Biosciences), anti-CD14–PE-Cy5 (clone 61D3, Thermo Fisher Scientific), anti-CD16–BUV496 (clone 3G8, BD Biosciences), anti-CD19–BUV563 (clone HIB19, BD Biosciences), anti-CD27–BV786 (clone O323, BioLegend), anti-CD34–BB660 (clone 581, BD Biosciences), anti-CD38–APC (clone HB7, BD Biosciences), anti-CD45–BUV805 (clone HI30, BD Biosciences), anti-CD45RA–BV570 (clone HI100, BioLegend), anti-CD56–BUV615 (clone NCAM16.2, BD Biosciences), anti-CD69–BUV737 (clone FN50, BD Biosciences), anti-CD71–BUV661 (clone M-A712, BD Biosciences), anti-CD83–BB790 (clone HB15e, BD Biosciences), anti-CD86–BB630 (clone 2331 (FUN-1), BD Biosciences), anti-CD123–PE-Cy7 (clone 7G3, BD Biosciences), anti-CD127–BV421 (clone A019D5, BioLegend), anti-HLA-DR–BV605 (clone G46-6, BD Biosciences) and anti-PD-1–R718 (clone EH12.1, BD Biosciences) for 30 min at RT (Supplementary Table 8). Stained cells were washed twice with FACS buffer (PBS containing 2% fetal bovine serum and 2 mM EDTA), fixed in Cytofix Fixation Buffer (BD Biosciences), and acquired using a FACSymphony A3 (BD Biosciences). Data were analyzed using FlowJo version 10.9.0 (FlowJo).

AIM assay

PBMCs were thawed quickly, resuspended in RPMI 1640 Complete Medium (Sigma-Aldrich) supplemented with DNase I (10 U ml^–1; Sigma-Aldrich), and rested at 1 × 10⁶ cells per well in 96-well U-bottom plates (Corning) for 3 h at 37 °C. The medium was then supplemented with unconjugated anti-CD40 (clone HB14, Miltenyi Biotec) and anti-CXCR5–BB515 (clone RF8B2, BD Biosciences), followed 15 min later by the relevant peptides (each at 0.5 μg ml^–1), and the cultures were incubated for 12 h at 37 °C. Negative-control wells contained equivalent DMSO. After incubation, cells were washed with PBS, labeled with LIVE/DEAD Fixable Aqua (Thermo Fisher Scientific) for 10 min at RT, washed with FACS buffer, and stained with anti-CCR4–BB700 (clone 1G1, BD Biosciences), anti-CCR6–BUV737 (clone 11A9, BD Biosciences), anti-CCR7–APC-Cy7 (clone G043H7, BioLegend), anti-CX3CR1–PE (clone 2A9-1, BioLegend) and anti-CXCR3–AF647 (clone G025H7, BioLegend) for 10 min at 37 °C. Cells were then stained further with anti-CD3–BUV805 (clone UCHT1, BD Biosciences), anti-CD4–BUV496 (clone SK3, BD Biosciences), anti-CD8–BUV395 (clone RPA-T8, BD Biosciences), anti-CD14–BV510 (clone M5E2, BioLegend), anti-CD19–BV510 (clone HIB19, BioLegend), anti-CD28–BUV563 (clone CD28.2, BD Biosciences), anti-CD38–APC-R700 (clone HIT2, BD Biosciences), anti-CD39–BV711 (clone A1, BioLegend), anti-CD45RA–BV570 (clone HI100, BioLegend), anti-CD69–BV650 (clone FN50, BioLegend), anti-CD71–BUV661 (clone M-A712, BD Biosciences), anti-CD95–PE-Dazzle594 (clone DX2, BioLegend), anti-CD127–PE-Cy5 (clone A019D5, BioLegend), anti-CD137–PE-Cy7 (clone 4B4-1, BioLegend), anti-CD154–BV421 (clone 24-31, BioLegend), anti-HLA-DR–BV605 (clone G46-6, BD Biosciences), anti-PD-1–BUV615 (clone EH12.1, BD Biosciences) and anti-TIGIT–BV786 (clone 741182, BD Biosciences) for 30 min at RT in the presence of Brilliant Stain Buffer Plus (BD Biosciences; Supplementary Table 9). Stained cells were washed twice with FACS buffer, fixed in Cytofix Fixation Buffer (BD Biosciences), and acquired using a FACSymphony A5 (BD Biosciences). Data were analyzed using FlowJo version 10.9.0 (FlowJo).

Tetramer staining and phenotypic analysis

PBMCs were thawed quickly, resuspended in RPMI 1640 Complete Medium (Sigma-Aldrich) supplemented with DNase I (10 U ml^–1; Sigma-Aldrich), and seeded at 2 × 10⁶ cells per well in 96-well U-bottom plates (Corning). Cells were incubated first with dasatinib (50 µM; STEMCELL Technologies) for 10 min at RT and then with the relevant peptide–HLA class I tetramers (each at 1 µg per stain) for 20 min at RT (Supplementary Table 10). After incubation, cells were washed with PBS, labeled with LIVE/DEAD Fixable Aqua (Thermo Fisher Scientific) for 10 min at RT, washed with FACS buffer, and stained with anti-CCR7–APC-Cy7 (clone G043H7, BioLegend), anti-CX3CR1–BUV615 (clone 2A9-1, BD Biosciences) and anti-CXCR3–PE-Cy5 (clone G025H7, BioLegend) for 10 min at 37 °C. Cells were then stained further with anti-CD3–BUV805 (clone UCHT1, BD Biosciences), anti-CD4–PE-Cy5.5 (clone RM4-5, Thermo Fisher Scientific), anti-CD8–BUV395 (clone RPA-T8, BioLegend), anti-CD14–BV510 (clone M5E2, BioLegend), anti-CD19–BV510 (clone HIB19, BioLegend), anti-CD27–BV786 (clone O323, BioLegend), anti-CD38–BUV496 (clone HIT2, BD Biosciences), anti-CD39–BV711 (clone A1, BioLegend), anti-CD45RA–BV570 (clone HI100, BioLegend), anti-CD95–BB700 (clone DX2, BD Biosciences), anti-CD127–BB630 (clone HIL-7R-M21, BD Biosciences), anti-HLA-DR–BV650 (clone G46-6, BD Biosciences), anti-LAG-3–BUV661 (clone 3DS223H, Thermo Fisher Scientific), anti-PD-1–BUV737 (clone EH12.1, BD Biosciences), anti-TIGIT–PE-Dazzle594 (clone A15153G, BioLegend) and anti-TIM-3–BV605 (clone F38-2E2, BioLegend) for 20 min at RT, washed twice with FACS buffer, fixed/permeabilized using a FoxP3 Transcription Factor Staining Buffer Set (Thermo Fisher Scientific), and stained intracellularly with anti-EOMES–EF660 (clone WD1928, eBioscience), anti-granzyme B–BB790 (clone GB11, BD Biosciences), anti-Ki67–AF700 (clone B56, BD Biosciences), anti-T-BET–PE-Cy7 (clone 4B10, eBioscience) and anti-TCF-1–AF488 (clone C63D9, Cell Signaling Technology) for 30 min at RT (Supplementary Table 11). Stained cells were washed twice with FACS buffer and acquired using a FACSymphony A3 (BD Biosciences). Data were analyzed using FlowJo software version 10.9.0 (FlowJo).

Plasma proteomics

A data-driven approach was used to select healthy convalescent individuals (n = 51) and individuals with long COVID (n = 51) for plasma proteome characterization via a Proximity Extension Assay (Olink Proteomics). Immune cell subset proportions were summarized using a PCA. Outlier samples were excluded based on the greatest deviation from the origin along PC1 to PC4. Plasma samples were analyzed in two batches using Explore 3072 (Olink Proteomics). Sixteen bridge samples were included for quality control purposes in each batch.

General statistics

Differences between groups were assessed using a two-tailed Mann–Whitney U-test. Raw P values are shown. Correlations were evaluated using the two-tailed Pearson coefficient or a two-tailed Spearman rank test. Significance was assigned at P < 0.05. Basic statistical analyses were performed using Prism version 9.5.0 (GraphPad).

Flow cytometry data analysis

Samples acquired for immune cell lineage analysis were gated to the single-cell/viable/CD45⁺ population and subsequently exported to contain only 3,000 events using the FlowJo Plugin DownSample version 3. Exported fcs files were loaded into R using flowCore version 2.6.0. All data were concatenated into a single matrix with compensated markers (excluding viability, CD34 and CD45). Data for each marker were scaled and centered for analysis using umap version 0.2.10.0. Clustering was performed using a Gaussian mixture model (maxNumComponents = 10) implemented in mclust version 6.0.0. Data were visualized using ggplot2 version 3.4.2. Antigen-specific CD4⁺ and CD8⁺ T cell frequencies assessed via the AIM assay were calculated after background subtraction. Samples acquired for detailed phenotypic characterization were excluded below a threshold of five tetramer⁺ CD8⁺ T cells per specificity. The expression of each marker was then normalized to the average geometric mean fluorescence intensity across all samples and specificities and used to calculate the co-inhibitory score, representing the summed data for PD-1, TIM-3, LAG-3 and TIGIT. Statistical analyses were performed using R version 4.2.1.

Plasma proteome data analysis

Bridge sample data were normalized using the olink_normalization function implemented in OlinkAnalyze version 3.4.1. Differential expression analyses were performed using a Wilcoxon rank-sum/Mann–Whitney U-test with Benjamini–Hochberg correction implemented via the olink_wilcox function in OlinkAnalyze version 3.4.1. GSEA was performed using fgsea version 1.20.0 incorporating lists of all analyzed proteins ordered by correlation coefficient or fold change. Gene sets were downloaded from the MSigDB using msigdb version 7.5.1. Overrepresentation analysis was performed using the fora function implemented in fgsea version 1.20.0 incorporating all measured proteins as the ‘universe’. At least five proteins were required in each gene set for consideration. Significance was evaluated using a hypergeometric test. Correlations were calculated using the cor.test function implemented in stats version 4.1.3. PCAs were performed using the prcomp function implemented in stats version 4.1.3. Data were visualized using ggplot2 version 3.4.2 and pheatmap version 1.0.12. All analyses were performed using R version 4.2.1. Network analyses of plasma proteins that were differentially expressed as a function of symptom severity were performed using the stringApp in Cytoscape version 3.10.3 (ref. ⁷¹).