biostudies-arrayexpressUnknownTranscriptomicsGenomicsProteomicsfrederic lepretreIllumina NovaSeq 6000RNA-seq of coding RNAHomo sapiensHomo sapienshttps://www.ebi.ac.uk/biostudies/studies/E-MTAB-15068COVID-19 induces profound B-cell dysregulation, notably a marked expansion of plasmablasts (PB), whose functional role remains unclear. This study aimed to characterize PB dynamics and functions in COVID-19 and their association with disease severity. We performed longitudinal immune profiling in a prospective cohort of 50 patients with COVID-19 (cohort 1), including flow cytometry-based B-cell immunophenotyping and multiplex cytokine analysis at days 1, 7, 14, and 30. A second retrospective cohort of 282 corticosteroid-naïve patients (cohort 2) was used to validate PB dynamics, model PB trajectories, and perform transcriptomic profiling of sorted PB. PB expansion occurred early in COVID-19 and was positively correlated with maximal disease severity (r=0.53, p<0.0001). Two distinct PB expansion trajectories were identified: one rapidly resolving, and one persistent and amplified, the latter being associated with higher severity scores and 30-day mortality (31% vs. 5%, p<0.001). In cohort 1, BAFF levels at day 7 correlated positively with both PB proportion (r=0.59, p=0.002) and maximal disease severity (r=0.74, p<0.001). Transcriptomic profiling of PB in cohort 2 revealed severity-specific signatures: in severe cases, early PB upregulated genes related to purine metabolism and CD39 expression, suggesting a pro-inflammatory role. In non-severe cases, PB expressed interferon-related and CIITA-mediated MHC-II programs, indicative of antiviral function. PB display dual functional profiles in COVID-19, acting either as regulators of antiviral immunity or as amplifiers of inflammation in severe disease. These findings support exploring therapeutic strategies targeting the BAFF-PB axis in severe COVID-19.biostudies-arrayexpressSequencing - The libraries were sequenced on a NovaSeq 6000 system using a 100-cycle chemistry. All samples, except one, achieved at least 20 million reads. To eliminate low-quality regions and poly(A) tails from the reads, we used the fastp program with a quality score threshold of 20, removing reads shorter than 25 bp. Read alignments were performed using the STAR program with the human reference genome GRCh38 and reference gene annotations from Ensembl. UMIs were utilized to reduce errors and PCR bias through fastp and umi-tools. Based on deduplicated read alignments (UMI), gene-level molecule counts were determined using FeatureCounts. Additional programs such as qualimap, FastQC, and MultiQC were used for quality control of the reads.Nucleic Acid Extraction - Total RNAs were extracted using the NucleoSpin® RNA kit (MACHEREY-NAGEL) as per the manufacturer’s instructions.Library Construction - The samples exhibited heterogeneity in total RNA quantity. To address this, we concentrated the samples with the lowest RNA concentrations using a SpeedVac and limited the input to a maximum of 50 ng for the most concentrated ones (n=2) to avoid an excessive concentration range. After concentration, the estimated total RNA amounts ranged from 0.1 ng to 50 ng in a volume of 4 µL. From 4 µL of total RNA, we added 1 µL of ERCC spike-in control. Library preparation was then initiated using oligo(dT) priming, following the QuantSeq FFPE 3' mRNA FWD protocol with UDI (Unique Dual Indexes) from Lexogen, incorporating the low-input specific workflow. Unique Molecular Identifiers (UMIs) were introduced to facilitate the removal of PCR duplicates during analysis. After generating the double-stranded cDNA library, we purified it using magnetic beads and amplified it. Library amplification was performed with 22 cycles. Libraries with the lowest yields (less than 600 pM in the 200 bp–1000 bp range) were re-amplified using the reamplification add-on kit for Illumina (LEXOGEN) with six to eight additional cycles. Final library concentrations ranged from 222 pM to 11,187 pM (200 bp–1000 bp interval). After equimolar pooling of the libraries, the pool was repurified to eliminate primer dimers.Sample Collection - All patients with a laboratory-confirmed diagnosis of COVID-19 who were hospitalized at the CHU Lille and had available lymphocyte immunophenotyping performed between April 8th and May 1st, 2020, were retrospectively enrolled. Patients had to have symptomatic SARS-Cov2 infection confirmed by positive nasopharyngeal swab polymerase chain reaction. Samples used for lymphocyte immunophenotyping were stored for B-cell isolation and DNA extraction. Data on demographics, clinical manifestations, routine laboratory features, and therapeutic management, and outcomes were collected. The alpha (B.1.1.7) SARS-CoV-2 variant was dominant in France during this period. Among the patients in cohort 2, we retrospectively constituted a cohort named 2b, by collecting 41 samples from 23 patients with blood samples available according to maximum severity and time since onset of disease to perform RNA sequencing on PB cells and other B cells.OrganizationMINSEQE ScoreAssays and DataMAGE-TAB Filesfrederic lepretrefalsePlasmablasts in COVID-19: Dual Role as Immune Effectors or Amplifiers of InflammationCOVID-19 induces profound B-cell dysregulation, notably a marked expansion of plasmablasts (PB), whose functional role remains unclear. This study aimed to characterize PB dynamics and functions in COVID-19 and their association with disease severity. We performed longitudinal immune profiling in a prospective cohort of 50 patients with COVID-19 (cohort 1), including flow cytometry-based B-cell immunophenotyping and multiplex cytokine analysis at days 1, 7, 14, and 30. A second retrospective cohort of 282 corticosteroid-naïve patients (cohort 2) was used to validate PB dynamics, model PB trajectories, and perform transcriptomic profiling of sorted PB. PB expansion occurred early in COVID-19 and was positively correlated with maximal disease severity (r=0.53, p<0.0001). Two distinct PB expansion trajectories were identified: one rapidly resolving, and one persistent and amplified, the latter being associated with higher severity scores and 30-day mortality (31% vs. 5%, p<0.001). In cohort 1, BAFF levels at day 7 correlated positively with both PB proportion (r=0.59, p=0.002) and maximal disease severity (r=0.74, p<0.001). Transcriptomic profiling of PB in cohort 2 revealed severity-specific signatures: in severe cases, early PB upregulated genes related to purine metabolism and CD39 expression, suggesting a pro-inflammatory role. In non-severe cases, PB expressed interferon-related and CIITA-mediated MHC-II programs, indicative of antiviral function. PB display dual functional profiles in COVID-19, acting either as regulators of antiviral immunity or as amplifiers of inflammation in severe disease. These findings support exploring therapeutic strategies targeting the BAFF-PB axis in severe COVID-19.2025-07-01T00:00:00Z2025-04-23T13:14:25Z2025-04-23T13:14:25ZE-MTAB-15068ERP171892EFO_0002944EFO_0004170EFO_0005518EFO_0003738EFO_0004184