Impact of 3D cell culture hydrogels derived from basement membrane extracts or nanofibrillar cellulose on CAR-T cell activation

Abstract

Hydrogel-based 3D culture systems are increasingly used for preclinical evaluation of cell-based immunotherapies, including chimeric antigen receptor T (CAR-T) cells. However, hydrogel properties can influence T cell behavior, potentially affecting interpretation of immunotherapy studies. We assessed CD4⁺ T and CAR-T cell responses in two chemically undefined matrices-Matrigel and basement membrane extract (BME)- and in a synthetic nanofibrillar cellulose (NFC) hydrogel. Although NFC was mechanically stiffer, T cell activation and proliferation were higher in NFC than in Matrigel or BME. Murine CD4⁺ T cells acquired a regulatory phenotype in Matrigel and BME but not in NFC. Similarly, CAR-T cell function was reduced in Matrigel and BME but maintained in NFC. These findings underscore how matrix composition can shape T cell responses in 3D culture. NFC provides a chemically defined alternative that preserves T cell activity, supporting its use in more accurate preclinical testing of immunotherapies.

Keywords: Biological sciences; Biomaterials; Cell biology; Immune response; Materials science.

Graphical abstract

Figure 1

Rheological properties of NFC hydrogel, Matrigel, and BME (A) Rheological time and temperature sweep measurements displaying the crosslinking kinetics and behavior of NFC, Matrigel (MG), and BME, with a temperature ramp from 4°C to 37°C at 5°C per minute before stabilization. (B) Rheological frequency sweep measurements at 37°C, showing the frequency dependent viscoelastic behavior of NFC, Matrigel, and BME. (C) Rheological amplitude sweep at 37°C, showing the amplitude dependent viscoelastic behavior of NFC, Matrigel, and BME. (D) Storage modulus values derived from frequency sweep experiments at a 1 Hz frequency of NFC, MG, and BME. Data representative of three independent measurements are presented as means ± SD. Data were analyzed using unpaired one-way ANOVA with Tukey’s multiple comparison test (∗p < 0.05). NFC, nanofibrillar cellulose hydrogel; MG, Matrigel; BME, basement membrane extract.

Figure 2

ECM-derived hydrogels (Matrigel and BME) influence murine CD4⁺ T cell activation, function and differentiation to regulatory T cells (A) Representative bright field images (EVOS microscope, scale bar 2000 μm) of CD4⁺ T cells purified from Foxp3eGFP mice stimulated ex vivo with anti-CD3 (1 μg/mL) and anti-CD28 (1 μg/mL) in standard 2D suspension (control) or embedded in different hydrogels (NFC, Matrigel, and BME). CD4⁺ T cells were analyzed by flow cytometry after 5 days in culture. (B) Number of viable CD4⁺ T cells recovered after culture. (C) Median fluorescent intensity (MFI) of CD25 in CD4⁺ T cells. (D) Percentage of IFNγ-expressing CD4⁺ T cells. (E and F) Representative histograms showing CellTrace Violet expression on CD4⁺ T cells (E) and proportion of proliferative CD4⁺ T cells (f; CTV-low). (G and H) Percentage of Foxp3 eGFP⁺ CD25^hi Treg cells (G) and representative histograms showing Foxp3 expression on Foxp3-eGFP⁺ Treg cells (H). Each point represents an individual mouse. Data from three independent experiments are presented as means ± SD. Data were analyzed using unpaired one-way ANOVA with Tukey’s multiple comparison test (∗p < 0.05, ∗∗p < 0.005, ∗∗∗∗p < 0.0001). NFC, nanofibrillar cellulose; MG, Matrigel; BME, basement membrane extract.

Figure 3

ECM-derived hydrogels (Matrigel and BME) hamper human CD4⁺ T cell activation and function (A–H) CD4⁺ T cells isolated from cord blood mononuclear cells were stimulated ex vivo with anti-CD3 (1 μg/mL) and anti-CD28 (1 μg/mL) for 5 days in standard 2D suspension (control) or embedded in NFC, Matrigel and BME. (A) Number of viable CD4⁺ T cells recovered after culture. (B and C) Median fluorescent intensity (MFI) of CD25 in CD4⁺ T cells (B) and representative histograms showing CD25 expression on CD4⁺ T cells (C). (D and E) Percentage of IFNγ-expressing CD4⁺ T cells (D) and representative histograms showing IFNγ expression on CD4⁺ T cells (E). (F and G) Representative histograms showing CellTrace Violet expression on CD4⁺ T cells (F) and proportion of proliferative CD4⁺ T cells (CTV-low; G). (H) Percentage of FOXP3⁺ CD25^hi Treg cells. Each point represents an individual donor. Data from three independent experiments are presented as means ± SD. Data were analyzed using unpaired one-way ANOVA with Tukey’s multiple comparison test (∗p < 0.05, ∗∗p < 0.005, ∗∗∗∗p < 0.0001). NFC, nanofibrillar cellulose; MG, Matrigel; BME, basement membrane extract.

Figure 4

CAR-T cell cytotoxicity in short term cultures is comparable between NFC, Matrigel, and BME CD20 CAR-T cells were co-cultured with Daudi (Burkitt lymphoma cell line) cells labeled with CellTrace Violet (CTV) for 24 h at the different effector to target (E:T) ratios specified, in standard 2D suspension (control) or embedded in the specified hydrogels. (A) Representative gating strategy for analyzing the viability of Daudi and CD20 CAR-T cells by flow cytometry. (B) Number of viable CAR-T cells recovered after culture in the 1:1 E:T condition (50,000 Daudi +50,000 CAR-T cells/well) as measured by flow cytometry using counting beads (Flow-Count fluorospheres). Data were analyzed using a one-way ANOVA with Tukey’s multiple comparison test. (C) Specific lysis (%) of Daudi cells induced by CD20 CAR-T cells at the specified E:T ratios. Specific apoptosis was calculated by applying the following formula: [(%viable untreated − %viable treated)/%viable untreated] × 100. (D and E) Bar plot showing median fluorescence intensity (MFI) of (D) IFNγ and (E) CD69 in CAR-T cells cultured for 24 h in the described conditions. Data from two or three independent experiments are presented as means ± SD. Data were analyzed using a two-way ANOVA with Tukey’s multiple comparison test (∗p < 0.05, ∗∗p < 0.005). NFC, nanofibrillar cellulose; MG, Matrigel; BME, basement membrane extract.

Figure 5

ECM hydrogels (Matrigel and BME) reduce CAR-T cell activation and proliferation (A) Representative bright field images (scale bar, 1,000 μm) of CD20 CAR-T cells stimulated in vitro for 4 days with coated anti-CD3 (1.6 μg/mL) and soluble anti-CD28 (1 μg/mL) in either standard 2D suspension (control) or the indicated gels. (B and C) Number of viable CAR-T cells recovered after 4 (B) or 10 days (C) of culture as measured by flow cytometry using counting beads (Flow-Count fluorospheres). (D) Representative histograms showing CD25 expression in CAR-T cells. (E and F) median fluorescence intensity (MFI) of CD25 in CAR-T cells after 4 (E) or 10 days (F) of culture. (G–I) For flow cytometry-based analysis of proliferation, CD20 CAR-T cells were labeled with CellTrace Violet (CTV) dye on day 0 and stimulated as indicated in (A). (G) Representative histograms showing CTV expression on CAR-T cells and (H and I) proportion of proliferative CAR-T cells (low CTV expression) after (H) 4 or (I) 10 days of culture. (J and K) Median fluorescence intensity (MFI) of (E) TIM3 and (F) HLADR in CAR-T cells after 10 days culture. Data from two or three independent experiments are presented as means ± SD. Data were analyzed using unpaired one-way ANOVA with Tukey’s multiple comparison test (∗p < 0.05, ∗∗∗p < 0.0005, ∗∗∗∗p < 0.0001). NFC, nanofibrillar cellulose; MG, Matrigel; BME, basement membrane extract.