Stratifying neutrophil responses in Glioblastoma

Background

Glioblastoma (GBM) is an aggressive solid cancer with a 5-year survival rate of 3.4%¹.  Limited advances beyond surgical and radio/chemotherapy have been made in the last 20 years, despite the evolution of immune therapies.  More recently, expansion of the neutrophil compartment has been correlated with disease progression questioning the therapeutic potential of targeting the neutrophil response².  Whilst training of granulopoiesis has been reported to promote anti-tumour neutrophil activities in murine models of lung cancer³, pro-tumoral neutrophil responses have also been described most notably through the ability of tissue neutrophils to facilitate tumour metastasis⁴.  New data from our group has revealed that there is expansion of circulating mature normal density (NDN) and immature low density (LDN) neutrophil subsets in patients with GBM.  Importantly, whilst LDN demonstrate an enhanced ability to kill primary human-derived GBM tumour cells, preliminary transcriptomics and immunohistochemical data would suggest that LDN signatures are absent from the primary tumour site. It remains unclear as to whether this is the result of active exclusion of LDN populations from the primary tumour site, or whether neutrophils demonstrate plasticity and can switch between different maturation and functional states. Given this heterogeneity, we now aim to stratify blood and tissue neutrophils by phenotype and function to allow precision targeting of anti-tumoural neutrophil responses.

Aims

1. Define neutrophil subtypes in blood, tumour and peri-tumour tissue of patients with Glioblastoma using single-cell proteomic technology.
2. Characterise pro and anti-tumour responses of neutrophil subsets using patient-derived glioma cell lines.
3. Determine location and dynamics of neutrophil subsets within tumour and peri-tumour tissues using 3D multiplexed immunofluorescence microscopy and brain tumour slice imaging.  

Training Outcomes

Training will be provided in core neutrophil biology techniques including isolation, flow cytometry phenotyping, tumour co-cultures, functional assays (NETosis, respiratory burst, phagocytosis and killing), single cell sorting and generation of neutrophil libraries for subsequent proteomic analysis (SW).  Training in mass spectrometry sample preparation and data analysis will be undertaken in Edinburgh (AvK). Cells will be analysed using a timsTOF single-cell-proteomic (SCP) mass spectrometer.  Initial data will be searched using Data Independent Analysis DIA-Neural-Network (DIA-NN) followed by batch correction and quality control to identify and quantify the neutrophil proteome. The student will first train in classical bioinformatic/machine learning (ML) applications to perform an in-depth analysis of the generated proteomic/clinical data. Specific training in the use of 0-value imputation using K-Nearest-Neighbours, differential expression using support vector machines, dimensionality reduction (PCA, tSNE etc.) and clustering analysis will be provided for analysis of SC-proteomic data.  Training in culture and imaging of human (PB, JB) and murine (JB) brain tumour slices will occur across partners in Edinburgh and Glasgow, with 3D multiplexed immunofluorescence microscopy undertaken at the Beatson (LC). Nanostring gene expression data sets have already been generated from GBM and healthy control neutrophil subsets. Scientific expertise in neutrophil and GBM biology, mass spectrometry, leucocyte dynamics and quantitative histopathology will be provided together with day to day support from postdoctoral and clinical fellows in the Walmsley group which has core funding from a Wellcome Discovery Award.

References

¹Brodbelt, A. et al., 2015. Eur J Cancer. 51;533-542.  ²Bambury, R.M. et al., 2013.  J Neuro-Oncol. 114;149-154^{.  3}Kalafati, L. et al., 2020. Cell 183;771-785.  ⁴Wculek, S. K., and Malanchi, I. 2015. Nature 528;413-417.