Matrix-assisted laser desorption ionization (MALDI) imaging mass spectrometry (MALDI-MSI) allows us to investigate the distribution of lipid molecules within tissues. We used MALDI-MSI to identify prognostic gangliosides in tissue sections of rat intracranial allografts of rat glioma and mouse intracranial xenografts of human medulloblastoma. In the healthy adult rodent brain, GM1 and GD1 were the main types of glycolipids. Both gangliosides were absent in both intracranial transplants. The ganglioside GM3 was not present in the healthy adult brain but was highly expressed in rat glioma allografts. In combination with tandem mass spectrometry GM3 (d18:1/C24:0) was identified as the most abundant ganglioside species in the glioma allotransplant. By contrast, mouse xenografts of human medulloblastoma were characterized by prominent expression of the ganglioside GM2 (d18:0/C18:0). Together, these data demonstrate that tissue-based MALDI-MSI of gangliosides is able to discriminate between different brain tumors and may be a useful clinical tool for their classification and grading.

Medulloblastoma comprises four main subgroups (WNT, SHH, Group 3 and Group 4) originally defined by transcriptional profiling. In primary medulloblastoma tissues, these groups are thought to be distinguishable using the immunohistochemical detection of β-catenin, filamin A, GAB1 and YAP1 protein markers. To investigate the utility of these markers for in vitro studies using medulloblastoma cell lines, immunoblotting and indirect immunofluorescence were employed for the detection of β-catenin, filamin A, GAB1 and YAP1 in both DAOY and D283 Med reference cell lines and the panel of six medulloblastoma cell lines derived in our laboratory from the primary tumor tissues of known molecular subgroups. Immunohistochemical detection of these markers was performed on formalin-fixed paraffin-embedded tissue of the matching primary tumors. The results revealed substantial divergences between the primary tumor tissues and matching cell lines in the immunoreactivity pattern of medulloblastoma-subgroup-specific protein markers. Regardless of the molecular subgroup of the primary tumor, all six patient-derived medulloblastoma cell lines exhibited a uniform phenotype: immunofluorescence showed the nuclear localization of YAP1, accompanied by strong cytoplasmic positivity for β-catenin and filamin A, as well as weak positivity for GAB1. The same immunoreactivity pattern was also found in both DAOY and D283 Med reference medulloblastoma cell lines. Therefore, we can conclude that various medulloblastoma cell lines tend to exhibit the same characteristics of protein marker expression under standard in vitro conditions. Such a finding emphasizes the importance of the analyses of primary tumors in clinically oriented medulloblastoma research and the urgent need to develop in vitro models of improved clinical relevance, such as 3D cultures and organotypic slice cultures.

Canonical Wnt signaling is known to promote proliferation of olfactory stem cells. In order to investigate the effects of a constitutive activation of Wnt signaling in Sox2-positive precursor cells of the olfactory epithelium, we used transgenic mice that allowed an inducible deletion of exon 3 of the Ctnnb1 gene, which is responsible for the phosphorylation and degradation of Ctnnb1 protein. After induction of aberrant Wnt activation by Ctnnb1 deletion at embryonic day 14, such mice developed tumor-like lesions in upper parts of the nasal cavity. We still observed areas of epithelial hyperplasia within the olfactory epithelium following early postnatal Wnt activation, but the olfactory epithelial architecture remained unaffected in most parts when Wnt was activated at postnatal day 21 or later. In summary, our results suggest an age-dependent tumorigenic potential of aberrant Wnt signaling in the olfactory epithelium of mice.

Young children with brain tumours are at high risk of developing treatment-related sequelae. We aimed to assess neuropsychological outcomes 5 years after treatment. This cross-sectional study included children under 4 years of age with medulloblastoma (MB) or ependymoma (EP) enrolled in the German brain tumour trials HIT2000 and HIT-REZ2005. Testing was performed using the validated Wuerzburg Intelligence Diagnostics (WUEP-D), which includes Kaufman-Assessment-Battery, Coloured Progressive Matrices, Visual-Motor Integration, finger tapping "Speed", and the Continuous Performance Test. Of 104 patients in 47 centres, 72 were eligible for analyses. We assessed whether IQ was impacted by disease extent, disease location, patient age, gender, age at surgery, and treatment (chemotherapy with our without craniospinal irradiation [CSI] or local radiotherapy [LRT]). Median age at surgery was 2.3 years. Testing was performed at a median of 4.9 years after surgery. Patients with infratentorial EPs (treated with LRT) scored highest in fluid intelligence (CPM 100.9±16.9, mean±SD); second best scores were achieved by patients with MB without metastasis treated with chemotherapy alone (CPM 93.9±13.2), followed by patients with supratentorial EPs treated with LRT. In contrast, lowest scores were achieved by patients that received chemotherapy and CSI, which included children with metastasised MB and those with relapsed MB M0 (CPM 71.7±8.0 and 73.2±21.8, respectively). Fine motor skills were reduced in all groups. Multivariable analysis revealed that type of treatment had an impact on IQ, but essentially not age at surgery, time since surgery or gender. Our results confirm previous reports on the detrimental effects of CSI in a larger cohort of children. Comparable IQ scores in children with MB treated only with chemotherapy and in children with EP suggest that this treatment strategy represents an attractive option for children who have a high chance to avoid application of CSI. Longitudinal follow-up examinations are warranted to assess long-term neuropsychological outcomes.