Using Visualization of t-Distributed Stochastic Neighbor Embedding To Identify Immune Cell Subsets in Mouse Tumors

Area of Science:

• Immunology
• Computational Biology
• Bioinformatics

Background:

• High-dimensional flow cytometry is crucial for studying tumor-associated immune cells.
• Traditional gating methods may miss subtle immune cell populations or minor marker changes.
• Complex flow panels necessitate advanced analytical tools.

Purpose of the Study:

• To evaluate visualization of t-distributed stochastic neighbor embedding (viSNE) as an unsupervised tool for high-dimensional flow cytometry data analysis.
• To identify patterns of anti-tumor immune responses in mouse MB49 bladder tumors using viSNE.
• To compare immune cell populations in tumors, spleens, and tumor-draining lymph nodes.

Main Methods:

• Immune cells from syngeneic mouse MB49 bladder tumors, spleens, and tumor-draining lymph nodes were analyzed.
• A 15-color flow cytometry panel with a cell viability probe was employed.
• viSNE was used to visualize and analyze the high-dimensional data.

Main Results:

• viSNE mapped known immune cell populations, including T cells, B cells, eosinophils, neutrophils, dendritic cells, and NK cells.
• Distinct CD8+ T cell populations were identified based on CD86 and programmed cell death protein 1 expression.
• CD8+ T cells and CD8+ dendritic cells were detected within the tumor microenvironment.
• Differences in polymorphonuclear cells/granulocytic myeloid-derived suppressor cells between spleen and tumor were observed, attributed to CD44 loss during digestion.

Conclusions:

• viSNE is a valuable tool for high-dimensional immune cell analysis in tumor-bearing mice.
• viSNE effectively eliminates gating biases inherent in traditional methods.
• The tool successfully identifies immune cell subsets that might be overlooked by conventional gating strategies.