ABSTRACT. Chronic Intervillositis of Unknown Etiology (CIUE) is a histopathological lesion in the placenta, characterized by an abundant intervillous cell infiltrate of CD68-positive macrophages of maternal origin, accompanied by adverse pregnancy outcomes. The pathophysiologic mechanism is unclear. Therefore, studied the placental intervillous cell infiltrate in three dizygotic twin couples with one-sided CIUE: one had CIUE, the other represented the non-affected control. Imaging mass cytometry (IMC), using a 42-marker panel, enabled in-depth assessment of phenotype and spatial orientation of the intervillous maternal immune cells. Approximately 80.4% of the intervillous immune cells were macrophages, 19.0% T-cells, 0.5% B-cells and 0.1% NK cells. The absolute number of macrophages and T cells was higher in CIUE compared to controls. However, the relative frequency of immune cells in the intervillous space was not different between the pairs of co-twin fetuses. IMC data showed a difference in the activation status of macrophages between the affected and unaffected placentas: 44.5% of the CD68+ macrophages in the CIUE cases demonstrated bright expression of the activation markers HLA-DR and CD38, whereas this applied to only 5.0% of the CD68+ macrophages in the control twins. We confirmed previous data that the CD68+ cell population, along with T cells, is increased in the intervillous space during CIUE. Moreover, IMC showed an increased amount of HLA-DR+/CD38+/CD68+ macrophages in the intervillous space of the twin with CIUE. Since maternal intervillous blood is part of the maternal circulation and paired twin couples were studied here, it is likely that macrophages are activated locally in CIUE. Further study should point out if HLA-DR+ macrophages are adhering to the fetal trophoblast cells and if they are locally affecting CD4+ T cells to react toward fetal cells and other immune cells.

ABSTRACT. One major complication after solid organ transplantation (SOT) is antibody-mediated rejection (AMR) of the graft by anti-donor HLA antibodies. Modern immunosuppression mainly addresses T cell-mediated rejection, affecting the B-cell alloimmune response only indirectly. B-cell depletion protocols are inefficient in preventing AMR and associated with an increased infection risk, emphasizing the need for a more precise targeting of alloreactive B cells. By the introduction of CARs, T cells can be redirected against various surface antigens, which was shown to impressively mediate elimination of the targeted cells. B cells with anti-donor HLA specificity are uniquely characterized by extracellular expression of respective B-cell receptors (BCRs). Using the anti-HLA BCR as target molecule, we redirected T cells towards alloreactive B cells by introducing a novel chimeric receptor comprising an HLA molecule fused to intracellular 4-1BB/CD3ξ signalling domains in order to generate T cells overcoming rejection by antibodies (CORA-T cells). As a proof of concept, CORA-T cells harbouring a receptor with an extracellular truncated HLA-A*02:01 molecule were designed, in which amino acid variations were introduced to abrogate CD8 binding. Upon co-cultivation with a B-cell line expressing and releasing anti-HLA-A*02:01 antibodies, CORA-T cells showed a specific upregulation of activation marker (CD25, CD69, CD137) and pro-inflammatory cytokine expression (e.g. IL-2, TNF-α). Moreover, they exhibited cytotoxicity towards the target B-cell line, secreted cytotoxic mediators and effectively reduced the amount of released anti-HLA-A*02:01 antibody. Our results demonstrate that CORA-T cells are able to specifically recognize and eliminate alloreactive B cells, having the potential to selectively prevent the formation of anti-HLA antibodies. This suggests application of CORA-T cells as an innovative approach to specifically combat AMR and to improve long-term graft survival in SOT.

ABSTRACT. Introduction: The pandemic caused by SARS-CoV-2 remains a global concern. Continued efforts are still needed to tackle with the emerging variants. While the impacts of mutations on B-cell epitopes can be predicted and confirmed soon after the emergence of variants, the influence of mutations on the immunogenicity of T-cell epitopes have been less well understood. In the present study, we used a newly developed HLA II-peptide interaction assay to identify HLA II-binding regions in SARS-CoV-2 Spike and the mutations that might affect the interaction with HLA II. Method: We developed an assay that estimated the binding strength of HLA II-peptide interaction by measuring the cell-surface expression of HLA II-peptide fusion construct. In the present study, we analyzed the interaction of HLA-DPA1*02-DPB1*05:01 (DP5) and peptides (15-mer) that covered the entire Spike protein. The cell-surface expression of HLA II-peptide complex was measured by flow cytometry and normalized to the expression level of HLA II-negative ctrl peptide fusion. The ratio of the test peptide/negative ctrl peptide was used as an indicator of HLA II-peptide interaction. Results: Among the 211 peptides analyzed, we found > 8 peptides that showed strong binding to DP5, indicating that these regions might act as T-cell epitopes. Additional 20 regions showed intermediate binding. Part of the binding regions were overlapped with T-cell epitopes restricted by various HLA II alleles (Knierman et al. 2020; Tarke et al. 2021). Some of the DP5-binding regions were also overlapped with mutation sites of contagious variants. While most of the mutations (non-synonymous substitution) did not alter the interaction with DP5, some of the deletions in Delta and Omicron variants decreased the interaction with DP5, indicating that limited number of viral mutations might alter CD4+ T-cell responses. Further immunological analysis will be needed to confirm these findings.

ABSTRACT. Amyotrophic Lateral Sclerosis (ALS) is a rare neurologic disease with incidence ranging from 1 to 6 cases per 100,000 individuals in the United States and Europe. As the disease develops, muscle weakness and atrophy progress until most voluntary muscles are paralyzed, causing death between 3 to 5 years after the first symptoms. Despite its high heritability (40-50%), only a few genetic factors have been described for ALS. HLA (human leukocyte antigen) associations, for example, are restricted to a few studies from the 1980s that only analyzed 1-field resolution in limited sample sizes. Here, we present the first modern study searching for HLA associations in a large sample of patients with ALS (n = 1,849) and controls (2,026). Genotyping was performed by next-generation sequencing using previously validated methods, and association analysis was adjusted for sex. We initially found that the overall frequencies of HLA-DPB1 alleles were different among patients and controls (p = 0.01). Next, we found several amino acids in HLA-DPB1 significantly associated with ALS: positions 56, 69, and 85-87. Because positions 85-87 are in linkage disequilibrium, and our previous work identified the combination of positions 56 with 85-87 as defining the main serological epitopes of the HLA-DPB1 molecule and also being the primary units of natural selection for HLA-DPB1, we examined this combined amino acid motif in patients and controls. We found that the combination of position 56A with 85-87GPM is protective in ALS (OR=0.77, CI: 0.64-0.93, p = 0.005). Further, we found that glutamic acid at HLA-DPB1 position 69 predisposes to ALS (OR=1.23, CI:1.08-1.39, p = 0.002). This result is especially intriguing, as this residue has long been established as the primary mediator of risk for Chronic Beryllium Disease (CBD). Because some epidemiology of ALS overlaps with CBD, specifically with respect to electronics and heavy metal exposure, these results warrant further investigation.

ABSTRACT. HLA sensitized patients harbour antibodies and memory B cells directed against allogeneic HLA molecules, thereby significantly decreasing the chance to receive a compatible donor organ. Desensitization can be used to non-specifically target circulating antibodies and B-cells, however, this comes with several side effects and success is limited. Therefore, more precise and potent therapies are needed to specifically target HLA-specific humoral immune memory. Whereas Chimeric Antigen Receptor (CAR) T-cells are genetically modified cells that have been proven to specifically eliminate target cells, we decided to use this concept for targeting HLA-specific B cells. We designed Chimeric HLA Antigen Receptor (CHAR) T-cells, which harbour a specific HLA molecule as the extracellular domain coupled to a CD28 costimulatory domain and a CD3ζ intracellular signalling domain. In K562 cell lines, CHAR molecules could be detected on the cell surface with the correct conformation, requiring endogenously expressed β2m. Additionally, the presentation of household peptides in the HLA-A2 CHAR molecule was confirmed. HLA-A2 CHARs were also expressed in Jurkat Triple Parameter Reporter (TPR) cells, which express reporter constructs encoding T-cell activation markers NF-κB CFP, NFAT eGFP and AP-1 mCherry. Stimulation with microbeads coated with HLA-A2 specific monoclonal antibodies (mAbs) resulted in induction of NFkB, NFAT and AP1 in these cells, whereas HLA-A3 mAb coated beads did not. Furthermore, incubation with different HLA-A2-surface antibody bearing cell lines led to IFN-γ secretion by HLA-A2 CHAR primary CD8 T cells, whereas incubation with HLA-A3 surface antibody-positive cells did not. These data are promising for further research into CHAR T-cells. We hypothesise that CHAR T-cells can serve as personalized precision immunotherapy to desensitize (highly) sensitized kidney transplant candidates and will be a modality to treat humoral rejection after solid organ transplantation.

ABSTRACT. Unrelated donors (UD) with single mismatches at HLA-A, B, C (9/10) are a valid alternative when a fully 10/10 matched UD is not available for hematopoietic cell transplantation (UD-HCT). In the latter, we have shown that permissive HLA-DPB1 disparity is mechanistically based on the degree of immunopeptidome divergence between mismatched alleles, which in turn reflects differences in their peptide binding motifs (PBM). We hypothesized that this mechanism might also apply to HLA class I mismatches in the 9/10 setting. To test this, we used the hierarchical PBM clustering of 122 HLA class I allotypes (44 HLA-A, 63 HLA-B, 18 HLA-C) to identify 21 distinct PBM groups. Taking into account also the PBM group of the shared allele, single HLA class I mismatches were classified as PBM-matched or PBM-mismatched in the graft-versus-host (GvH), host-versus-graft (HvG) or both directions (bidirectional). In 2,562 9/10 UD-HCT from the Center of International Blood and Marrow Transplant Research, 1,629 (63.6%) pairs could be classified with the available PBM data. Of these, 386 (23.7%) were PBM-matched and 1,243 (76.3%) were PBM-mismatched, the latter including 254 (15.6%), 238 (14.6%) and 751 (46.1%) unidirectional HvG, unidirectional GvH or bidirectional mismatches, respectively. No significant difference in survival or other clinical endpoints was found between PBM-matched and PBM-mismatched pairs. In contrast, the absence of a PBM-mismatch in GvH direction in the unidirectional HvG or PBM-matched group was significantly associated with improved overall survival in multivariate analysis compared to the unidirectional GvH or bidirectional reference group (HR 0.76, 95% CI 0.63-0.92, P = 0.0036), reflecting lower graft-versus-host disease and transplant-related mortality. These observations support the notion that immunopeptidome divergence defines permissive HLA class I mismatches in 9/10 UD-HCT, a finding with both immunological and clinical implications.

ABSTRACT. HLA haplotype frequencies are estimated from ambiguous unphased HLA genotyping data using Expectation-Maximization (EM) algorithms. Current population genetics methods require independent EM frequency estimates for each population, and assume that each population is in Hardy-Weinberg Equilibrium (HWE). The HWE assumption of EM has thus far resulted in the exclusion of individuals from mixed or unknown ethnic backgrounds from reference datasets. Multi-region populations are currently poorly served by stem cell donor registry HLA imputation and matching implementations due to the inability of such algorithms to incorporate admixture into their population genetics models. To address this unmet need, we have expanded the imputation component of our GRaph IMputation and Matching (GRIMM) framework, where imputation becomes the expectation step in an iterative EM algorithm. Our novel multi-region EM implementation considers region as a Bayesian prior, enabling integration of HLA information from multiple single-region population groups, and for the first time including individuals with ambiguous or mixed ethnic backgrounds. We show that our multi-region EM produces much higher likelihood values and better haplotype recovery as measured by Kullback-Leibler divergence than all evaluated EM implementations when tested on real datasets of US donor registry HLA typings as well as simulated multi-region datasets of ambiguous HLA typings.