Session 4 – Clinical Applications
Chairs: Hyun-Dong Chang, Berlin, Axel Schulz, Berlin
Speaker: Birgit Sawitzki, Berlin
Cross-cohort immune profiling reveals treatment-relevant endotypes in Long Covid-ME/CFS
Birgit Sawitzki
BIH/Charité – Universitätsmedizin Berlin
Post-acute infection syndromes (PAIS), including Long Covid (LC), can severely impair quality of life, yet no approved therapies exist. A subset of patients develops myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), a heterogeneous disorder characterized by post-exertional malaise, cognitive dysfunction, pain and functional impairment. Variable responses to experimental therapies suggest the existence of biologically distinct disease endotypes and highlight the need for predictive biomarkers.To identify treatment-relevant immune endotypes, we analysed two complementary LC-ME/CFS cohorts: a non-selected LC-ME/CFS cohort representing the broad disease population and an immunoadsorption (IA) cohort enriched for patients with elevated GPCR-directed autoantibodies and linked treatment response data.Using NULISA plasma proteomics, we constructed correlation networks of soluble mediators and identified an ME/CFS-specific molecular network. Unsupervised clustering based on mediators within this network defined three reproducible molecular clusters characterized by low, medium and high abundances of network-associated soluble mediators. Integration with CyTOF-based whole-blood immune profiling revealed distinct immunological phenotypes underlying these molecular states. The high NULISA cluster was characterized by expansion of CXCR5⁻ non-germinal center (non-GCB) IgM⁺ non-switched memory B cells, consistent with extrafollicular B-cell activation. In contrast, the medium cluster was enriched for germinal center-associated B-cell populations, whereas the low cluster exhibited expansion of CD127⁺CD161⁺ CD4⁺ and CD8⁺ T cells, suggesting a fundamentally distinct immunological state.Importantly, these molecular clusters were associated with clinical characteristics. IA non-responders were strongly enriched within the high NULISA cluster, whereas IA responders preferentially localized to the medium cluster. In contrast, the low cluster consisted almost exclusively of patients from the non-selected LC-ME/CFS cohort, suggesting a distinct disease endotype that was largely absent from the GPCR-directed autoantibody-enriched cohort.To further assess autoimmune features, we applied an autoimmunity score based on soluble mediators that distinguishes Sjögren’s disease from acute SARS-CoV-2 infection. The high NULISA cluster exhibited the highest autoimmunity scores and showed the greatest similarity to Sjögren’s disease, whereas the low cluster had the lowest autoimmunity scores and was most distinct from Sjögren’s disease.Together, our findings identify three biologically distinct LC-ME/CFS endotypes that differ in soluble inflammatory networks, immune cell composition and treatment association. These data provide a framework for biomarker-guided patient stratification and support the development of precision therapies for LC-ME/CFS.
Mass Cytometry Analysis Defines Clusters of SLE Patients with Differential Mucosal Phenotypes that Correlate with Immune Activation and Disease Severity
Adrián Barreno Sánchez1,2, Alissa Karina Yudiputri1, Heike Hirseland1, Thuy-Vi Tran1, Sebastian Ferrara1, Anne Beenken1,4, Paulina Rybakowska3, Concepción Marañón Lizana3, Marta E. Alarcon Riquelme3,5, Tobias Alexander1,4, Henrik E. Mei1, Hyun-Dong Chang1,2, Axel R. Schulz1
- German Rheumatology Research Center Berlin – A Leibniz Institute, Berlin, Germany
- Department for Cytometry, Institute of Biotechnology, Technische Universität Berlin, Berlin, Germany
- Pfizer-University of Granada-Junta de Andalucía Centre for Genomics and Oncological Research (GENYO), Granada, Spain
- Department of Rheumatology and Clinical Immunology, Charité Universitätsmedizin Berlin, Berlin, Germany
- Institute for Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
Keywords: Systemic Lupus Erythematosus, mucosal immunity, mass cytometry
Systemic lupus erythematosus (SLE) is a heterogeneous systemic autoimmune disease driven by loss of immune tolerance and autoantibody production. Disease progression, clinical presentation, and treatment response vary widely across patients and remain poorly understood. Since emerging evidence implicates the gut-immune axis in SLE pathogenesis, we aim to examine mucosal immunity as a potential driver of this heterogeneity.
To do so, we developed a robust CyTOF-based workflow for ex-vivo phosphorylation profiling and mucosal immune phenotyping using fixed whole blood, designed to be scalable for large clinical cohorts. The pipeline uses a cell-sorting strategy to separate blood into lymphoid and myeloid fractions, each analyzed with optimized panels of more than 50 antibodies. These panels capture mucosal-associated populations, alongside phosphoprotein markers reflecting intracellular signaling activity.
Applying this workflow to 326 SLE samples reproduced known lymphoid and myeloid differences versus healthy controls, and additionally revealed significant reductions in mucosa-associated populations, particularly MAIT cells and αEβ7+ CD8+ T cells. Clustering patients by abundance of these mucosal-related lymphoid populations identified three subgroups with distinct phenotypic and clinical profiles. One cluster, marked by elevated MAIT cells and naive-like γδ T cells but reduced circulating IgA+ B cells, showed higher rates of lupus nephritis and disease activity, along with altered B cell, NK cell, and monocyte subsets and increased pZAP70/Syk and pCREB signaling in several immune populations.
The findings demonstrate this workflow’s value for deep immune phenotyping focused on gut-related populations and signaling activity, and its potential to reveal clinically meaningful subtypes of SLE patients.
An optimized flow cytometry-based method for the isolation of foreign rare cell populations in human stem cells using monoclonal HLA class I specific antibodies.
Authors
Bernadette L. Bramreiter1, Rachel C. Quilang2, Carin van der Keur2, Anne Wagenmakers2, Katja Sallinger1, Emiel Slaats1, Julia Schönberger1, Katharina Schuch1, Hyun-Dong Chang3,4, Dave Roelen2, Marie-Louise van der Hoorn5 , Thomas van den Akker5, Michael Eikmans2, Thomas Kroneis1
1: Division of Cell Biology, Histology and Embryology, Gottfried Schatz Research Center, Medical University of Graz, Graz, Austria
2: Department of Immunology, Leiden University Medical Center, Leiden, Netherlands
3: German Rheumatism Research Center Berlin (DRFZ), Berlin, Germany
4: Institute for Biotechnology, Technische Universität, Berlin, Germany
5: Department of Obstetrics and Gynaecology, Leiden University Medical Center, Leiden, Netherlands
Objectives
Microchimerism, (Mc) defined as the presence of genetically distinct cells from another individual, occurs during pregnancy through a bi-directional cell transfer yielding maternal and fetal Mc (mMc, fMc). Current hypothesis assign microchimeric cells stem and progenitor cell-like properties contributing to a lifelong establishment of Mc. This study aimed to develop and optimize a sex-unbiased flow cytometry-based method to isolate viable potential mMc cells from human fetal stem cell populations using monoclonal HLA class I-specific antibodies (HuMoAbs).
Methods
Hybridoma derived HuMoAbs of IgG and IgM isotypes were generated at LUMC. Eleven HuMoAbs targeting HLA-A and HLA-B were selected and validated for the separation of maternal and fetal cells. To improve population discrimination and detection sensitivity, maternal specific HuMoAbs were separately conjugated to PE and AF488, respectively, and used simultaneously for staining. Fetal specific HuMoAbs were labeled with AF647. PE/AF488 double-positive and AF647 negative cells were considered of maternal origin. This approach was optimized in spike-in experiments and subsequently applied to fetal stem cell populations derived from bone marrow (BM), liver (L), amniotic membrane (AM) and amniotic fluid (AF).
Results
The optimized method enabled reliable separation of maternal and fetal cells, with detection sensitivity down to 0.004% maternal cells. Dual fluorochrome labelling significantly improved population discrimination. Across multiple fetal tissue-derived stem cell samples, viable maternal cells were isolated at average frequencies of 0.04% in AM, 0.7% in BM, 0.1% in L and 0.03% in AF.
Conclusion
We established a robust, sex-unbiased flow cytometry-based method for the pre-enrichment and isolation of viable potential maternal Mc cells from human fetal stem cell populations. This approach provides a valuable tool for downstream functional and molecular analyses of mMC cells.