Determination of CAR T cell metabolism in an optimized protocol

Abstract

Adoptive transfer of T cells modified to express chimeric antigenic receptors (CAR) has emerged as a solution to cure refractory malignancies. However, although CAR T cell treatment of haematological cancers has now shown impressive improvement in outcome, solid tumours have been more challenging to control. The latter type is protected by a strong tumour microenvironment (TME) which might impact cellular therapeutic treatments. Indeed, the milieu around the tumour can become particularly inhibitory to T cells by directly affecting their metabolism. Consequently, the therapeutic cells become physically impeded before being able to attack the tumour. It is therefore extremely important to understand the mechanism behind this metabolic break in order to develop TME-resistant CAR T cells. Historically, the measurement of cellular metabolism has been performed at a low throughput which only permitted a limited number of measurements. However, this has been changed by the introduction of real-time technologies which have lately become more popular to study CAR T cell quality. Unfortunately, the published protocols lack uniformity and their interpretation become confusing. We herein tested the essential parameters to perform a metabolic study on CAR T cells and propose a check list of factors that should be set in order to draw sound conclusion.

Keywords: CAR T cell; Seahorse XF96; cancer therapy; immunotherapy; metabolim.

FIGURE 1

Mitochondrial metabolic analysis by Seahorse XF Cell Mito Stress Kit. (A) Oxygen consumption rate (OCR) profile and (B) extracellular acidification rate (ECAR) profiles during the same metabolic monitoring. Basal OCR measurement is performed over the three first time points. This is followed by the measurement of OCR after sequential addition of Oligomycin (inhibits ATP synthesis, which stops the electron transport and blocks OXPHOS), Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP, proton uncoupler that mimics a sudden high need in energy, hence will show the maximum respiration rate) and Rotenone/Antimycin A (strong blockers of electron transport, lead to complete inhibition of OXPHOS). Three measures/drug are performed.

FIGURE 2

Optimization of cell concentration on mock T cells. (A) Quantification of CD19CAR expression in primary T cells at 5 and 10 days following transduction. (B) Cell distribution microscopy (left, scale bar: × 400 pm, × 10 objective) and respiration profile (right, dotted lines are the basal OCR range, 20 and 160 pmol/min) for the indicated cell concentration, these data were obtained from a single donor. (C) Basal OCR obtained for each cell concentration of mock and CD19CAR T cells. These results are presented as mean ± SD; (n = 4 donors). (D) SRC was calculated from the average maximal OCR—the average basal OCR for each cell concentration of mock and CD19CAR T cells. The results are mean ± SD; (n = 4 donors) and comparision between groups was performed with Two-way ANOVA test, ****p < 0.0001 (Only statistically different conditions are indicated on the graph).

FIGURE 3

Optimization of drug concentration. Indicated T cells were seeded at 200,000 cells/well and their respiration was tested with varying (A) Oligomycin, (B) FCCP, and (C) Rotenone/Antimycin concentrations. The OCR values plotted on the graphs (left) correspond to the OCR counted after the addition of the indicated drug, results are mean ± SD, N = 4 donors in sextuplicate (n = 6) and comparison between groups was performed with Two-way ANOVA test, ****p < 0.0001, ns; no significance.

FIGURE 4

Reproducibility and variability. (A) OCR calculation plots of T cells isolated from four healthy donors (N = 4), mock and CD19CAR, run in sextuplicate (n = 6) in ideal conditions (200,000 cells, 1.5 µM of Oligomycin, 1 µM of FCCP, and 0.5 µM of Rotenone/Antimycin). The results are mean ± SD. Two-way ANOVA test was performed to compare groups. ****p < 0.0001.

FIGURE 5

Stimulation of CD19CAR T cell. (A) Increased SRC of stimulated CD19CAR T cells. SRC was calculated from the average maximal OCR—the average basal OCR for each cell concentration of mock and CD19CAR T cells. (B) Increased ECAR for stimulated T cells (mock and CD19CAR). The experiments were run in optimal conditions as defined in the present report. The results are represented as mean ± SD; N = 4 donors in sextuplicate (n = 6); One-way ANOVA with multiple comparisons was performed to compare groups. *p < 0.1, **p < 0.01.