Immune Cell Infiltration

Immune Cell Infiltration2019-09-12T10:46:00-05:00

Background

  • Presence of Tumor-infiltrating immune cells (e.g. CD3/CD4/CD8/CD45+ T cells) is associated with better prognosis for cancer patients
  • Typical analysis involves sectioning of tumor tissue samples and processing for immunohistochemistry (IHC) to label T cells. This method works well but is very low throughput and costly, and as such not amenable to use in screening.
  • In vitro assay of immune cell infiltration: monitoring interaction of therapeutic with immune cells in 2D assay format (e.g. Boyden chambers, scratch assay). These models do not replicate the tissue microenvironment and thus do not adequately mimic in vivo conditions for study of immune cell infiltration.
  • 3D cell culture models are more biologically relevant and allow for proper assessment of immunotherapies designed to promote immune cell response to tumor cells
  • This assay can be utilized to assess effect of compounds and therapeutic antibodies on immune cell infiltration, or to screen immune cell populations for use in immunotherapy.

Protocol

InstrumentThermoFisher CX7 LZR
Analysis MethodHigh content screening
MarkersDAPI (total cell count)
CD3, CD8, CD45RO
Other markers available on request
Cell Model Type3D cell models (e.g. tumor spheroids, organoids, etc.)
Cell Types AvailableATCC cancer cell lines (e.g. A549, BT549, MCF-7)
Primary cells, cancer stem cells
Client provided cells
Custom models available on request
Test Article ConcentrationSingle point assay
8 point assay (20 ug/mL, 10 ug/mL, 2 ug/mL, 1 ug/mL, 400 ng/mL, 200 ng/mL, 100 ng/mL, 20 ng/mL)
(custom concentrations available)
Number of Replicates3 replicates per concentration
Quality ControlsVehicle control
Data DeliveryDose response curves illustrating number of infiltrating T cells vs dose of test compound,
Total cell counts for each treatment group,
Distribution of distance of infiltration for immune cells
Evaluation of statistical significance of results with respect to vehicle and treatment groups

General Procedure

  1. 3D cell models, grown to approximately 200 μm in diameter.
  2. Treatment with test compounds (if applicable)
  3. Seeding PBMCs at 4000 cells/well
  4. Fixation after 24 hours
  5. Nuclei are labeled with DAPI. Immunolabeling for selected CD markers
  6. Tissue clearing is applied to render 3D cell models transparent
  7. High content imaging is conducted on well plates
  8. Images are analyzed to quantify total number of cells and total number of infiltrating immune cells by CD expression

Data

Figure 1. Maximum Z-projection of Z-stack obtained from High Content Imaging of tumor spheroids in the presence of activated and naive immune cells. Data was quantified by automated image analysis to count total number of cells, and number of infiltrating immune cells.

This video depicts a representation of how immune cell infiltration is measured in 3D. Each immune cell is mapped to its position within the 3D cell model and the distance to the boundary is measured. Dark lines indicate cells that are close to the surface, orange/yellow lines depict cells that are far into the interior.

Figure 2. Representative data showing the observed number of CD3+ cells detected within the boundaries of the tumor spheroid, and a breakdown of CD3+ cell counts by location in the tumor spheroid.

Figure 3. Heatmap overlay depicting relative number of detected CD3+ cells in concentric regions of tumor spheroid.

Figure 4. Montage of z slices from confocal stack obtained from cleared tumor spheroid, depicting CD3+ cells in green, and nuclei in blue.

This website uses cookies to enhance the user experience. Ok