Dottermusch, M. | bioRxiv Science

bioRxiv2024

Microgliosis, astrogliosis and loss of aquaporin-4 polarity in frontal cortex of COVID-19 patients

The severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), causing human coronavirus disease 2019 (COVID-19), not only affects the respiratory tract, but also impacts other organs including the brain. A considerable number of COVID-19 patients develop neuropsychiatric symptoms that may linger for weeks and months and contribute to "long-COVID". While the neurological symptoms of COVID-19 are well described, the cellular mechanisms of neurologic disorders attributed to the infection are still enigmatic. Here, we studied the effect of an infection with SARS-CoV-2 on the structure and expression of marker proteins of astrocytes and microglial cells in the frontal cortex of patients who died from COVID-19 in comparison to non-COVID-19 controls. Most of COVID-19 patients had microglial cells with retracted processes and rounded and enlarged cell bodies in both gray and white matter, as visualized by anti-Iba1 staining and confocal fluorescence microscopy. In addition, gray matter astrocytes in COVID-19 patients were frequently labeled by intense anti-GFAP staining, whereas in non-COVID-19 controls, most gray matter astrocytes expressed little GFAP. The most striking difference between astrocytes in COVID-19 patients and controls was found by anti-aquaporin-4 (AQP4) staining. In COVID-19 patients, a large number of gray matter astrocytes showed an increase in AQP4. In addition, AQP4 polarity was lost and AQP4 covered the entire cell, including the cell body and all cell processes, while in controls, AQP4 immunostaining was mainly detected in endfeet around blood vessels and did not visualize the cell body. In summary, our data suggest neuroinflammation upon SARS-CoV-2 infection including microgliosis and astrogliosis, including loss of AQP4 polarity.

neuroscience↗

bioRxiv2023

Multiomic profiling of medulloblastoma reveals subtype-specific targetable alterations at the proteome and N-glycan level

Medulloblastomas (MBs) are malignant pediatric brain tumors that are molecularly and clinically very heterogenous. To unravel phenotypically relevant MB subtypes, we compiled a harmonized proteome dataset of 167 MBs and integrated findings with DNA methylation and N-glycome data. Six proteome MB subtypes emerged, that could be assigned to two main molecular programs: transcription/translation (pSHHt, pWNT and pGroup3-Myc), and synapses/immunological processes (pSHHs, pGroup3 and pGroup4). Multiomic analysis revealed different conservation levels of proteome features across MB subtypes at the DNA-methylation level. Aggressive pGroup3-Myc MBs and favorable pWNT MBs were most similar in cluster hierarchies concerning overall proteome patterns but showed different protein abundances of the vincristine resistance associated multiprotein complex TriC/CCT and of N-glycan turnover associated factors. The N-glycome reflected proteome subtypes and complex-bisecting N-glycans characterized pGroup3-Myc tumors. Our results shed light on new targetable alterations in MB and set a foundation for potential immunotherapies targeting glycan structures. SignificanceWhereas the application of omics technologies has significantly improved MB tumor classification and treatment stratification, it is still of debate, which features predict best clinical outcome. Moreover, treatment options - especially for high-risk groups - are still unsatisfactory. In contrast to nucleic acids, the proteome and their N-glycans may reflect the phenotype of a tumor in a more direct way and thus hold the potential to discover clinically relevant phenotypes and potentially targetable pathways. We show that these analyses are feasible on formalin fixed and paraffine embedded tissue. Compiling a comprehensive MB dataset, we detected new biomarkers and characteristics for high- and low-risk MB subtypes that were not reflected by other omic data modalities before. Specifically, we identified subtype specific abundance differences in proteins of the vincristine resistance associated multiprotein complex TriC/CCT and in proteins involved in N-glycan turnover. Changes in the N-glycans are considered as potential hallmarks of cancer and we show that N-glycan profiles can distinguish MB subtypes. These tumor-specific N-glycan structures hold a strong potential as new biomarkers, as well as immunotherapy targets. Highlights- Integration of in-house proteome data on formalin fixated paraffine embedded medulloblastoma (MB) and publicly available datasets enables large scale proteome analysis of MB - Six proteome MB subtypes can be assigned to two main molecular programs: replication/ translation versus synapse/immune system - Identification and validation of IHC compatible protein-biomarkers for high and low risk MB subtypes, such as TNC and PALMD. - Subtype specific correlation of the DNA methylome and the proteome reveals different conserved molecular characteristics across MB subtypes. - pGroup3-Myc subtype MBs are associated with high-risk features including high abundances of vincristine resistance associated TriC/CCT member proteins - Proteome MB subtypes show differential N-glycosylation patterns, revealing complex-bisecting glycans as potentially immunotargetable hallmarks of the high risk pGroup3-Myc subtype.

cancer biology↗