Researchers explore immunomodulating targets for improving glioblastoma therapies

In a recent article published in EMBO Molecular Medicine, researchers investigated cellular, molecular, and spatiotemporal heterogeneity of glioblastoma (GBM), a malignant and highly aggressive primary brain tumor.

They performed single-cell ribonucleic acid sequencing (scRNA-seq) of seven sets of patient-derived GBM stem cell cultures (GSCCs) combined with GBM-associated macrophage (GAM)-GBM co-cultures. Importantly, these GSCCs covered a broad spectrum of genetic aberrations frequently present in GBM.

Additionally, they performed real-time in vivo monitoring of GAM-GBM interactions in orthotopic zebrafish xenograft models.

Background

GBM treatment has not changed in the past 15 years and patients rarely survive over two years. GBM tumors' heterogeneity and an immunosuppressive tumor microenvironment (TME) contribute to hindering response to treatment. 

In contrast to other macrophages, GAMs are highly plastic immune cells exhibiting immunosuppressive properties that promote tumor progression instead of preventing tumor formation.

When co-cultured with GSCCs, they polarize toward a more immunosuppressive phenotype, even though this shift does not follow the M1/M2 markers-based classification for characterizing macrophage subtypes. Since they constitute 30–50% of the tumor, they are a valid target for new therapeutic approaches.

There is a need to evaluate multiple therapeutic approaches targeting GAMs, such as blocking GAM recruitment, reprogramming, and GAM-mediated phagocytosis, in parallel.

About the study

In the present study, researchers developed an in vitro model to analyze how GSCCs influenced the phenotypic features of GBMs in a patient-specific manner using scRNA-seq. They co-cultured GAMs and GSCCs for four days in a 1:5 ratio using the hanging-drop method.

Then, they collected cells in various conditions, labeled them using the MULTI-seq methodology, and pooled them for scRNA-seq.

Next, they optimized a zebrafish avatar model (an in vivo model) in which they engrafted green fluorescent protein (GFP)+ GSCCs into a macrophage reporter line.

They grew Zebrafish avatars at an elevated temperature of 34°C to allow normal development of the embryos while maintaining an environment that favored tumor cell proliferation. 

High-resolution live imaging of the zebrafish embryos in time-lapse movies helped the researchers capture and visualize dynamic interactions between GAMs and transplanted tumor cells in real time.

In addition, an image analysis pipeline processed in vivo recordings, which helped identify distinct spatiotemporal behavioral patterns of GSCCs and GAMs. 

Results

Sc-RNA seq and demultiplexing fetched as many as 5,320 cells from eight co-cultures and one monoculture of GAMs that met this study's quality control thresholds. The average number of genes detected per cell was 3,334.

The study analyses uncovered molecular-level GAM heterogeneity and a clear phenotype switch in vitro and in vivo. It also captured patient-specific cell–cell interaction patterns. 

Further, the results demonstrated that the extent of macrophage polarization correlated with GAM recruitment and activity in the corresponding zebrafish avatar models.

The advanced image processing pipeline captured the complex spatiotemporal interactions between GSCCs and GAMs. They uncovered differential tumor cell invasion and infiltration of reactive GAMs across different in vivo models. In addition, they revealed how tumor cell invasion and infiltration of reactive GAMs varied across patients.

It is noteworthy that the zebrafish xenograft models, owing to their unique features like ease of genetic manipulation, small size, and economic feasibility, have proven highly suited for studying GBM initiation, progression, migration, vasculature, and invasion. 

More importantly, they appear most apt for xenotransplantation of GBM tumor cells, given their high resemblance to the human brain structure. In the context of GBM, they could help researchers evaluate drugs that cross the blood-brain barrier (BBB).

Furthermore, gene set enrichment analysis (GSEA) found that non-invasive GSCCs in the zebrafish avatar model showed enrichment for extracellular matrix (ECM) production and deposition. 

Differential gene expression analysis and immunohistochemistry on GBM tumor samples and knock-out (KO) experiments in zebrafish identified the LGALS1 gene as a primary regulator of immunosuppression. Studies have implicated the LGALS1 gene encoding galectin-1 protein (GAL-1) in modulating cell-cell and extracellular cell-matrix (ECM) interactions.

Given that LGALS1 is involved in immunosuppression, observed changes in its expression might be correlated to reduced survival in GBM patients. Thus, LGALS1 KO in GBM tumor cells could transform the immune landscape. 

Studies have also shown that LGALS1 drives resistance to chemo- and immunotherapy; thus, incorporating LGALS1-targeted drugs into the current GBM treatment schedules could effectively improve outcomes for GBM patients.

Furthermore, the analyses highlighted the potential role of triggering receptor expressed on myeloid cell-2 (TREM2) signaling in GBM progression. Studies have implicated TREM2 signaling in immunosuppression witnessed during many different pathologies, such as neurodegenerative diseases, obesity, and several cancers.

One study found that TREM2 expression was associated with poor prognosis in GBM. Thus, more studies are warranted to elucidate mechanisms governing the immunosuppressive roles of GAL1 and TREM2 signaling.

Conclusions

The scRNA-seq profiling and the zebrafish avatar model revealed two features correlated to GBM patients' survival. First, they showed "substantial heterogeneity across GBM patients in GBM-induced GAM polarization", and second, they showed "the ability to attract and activate GAMs-features that correlated with patient survival".

In preclinical research, these models could work as a functional screening platform to help improve GBM treatments and identify novel and promising immune-modulating targets.

More importantly, these models hold the potential to maximize the efficacy of GBM therapy, predict a patient's response to specific drugs, and establish inclusion criteria for clinical trials evaluating GBM therapeutics.

Given the consistent failure of current GBM therapies, developing customized GBM therapeutic strategies is crucial.

In this regard, preclinical models based on patient-derived GBM tumor samples could allow a more precise reproduction of the patient's tumor complexity, which, in turn, would facilitate identifying exceptional responders who could benefit from a specific therapy.

Written by

Neha Mathur

Neha is a digital marketing professional based in Gurugram, India. She has a Master’s degree from the University of Rajasthan with a specialization in Biotechnology in 2008. She has experience in pre-clinical research as part of her research project in The Department of Toxicology at the prestigious Central Drug Research Institute (CDRI), Lucknow, India. She also holds a certification in C++ programming.