Nanomaterials Trigger Functional Anti-Tumoral Responses in Primary Human Immune Cells

Abstract

Targeting the immune system with nanoparticles (NPs) to deliver immunomodulatory molecules emerged as a solution to address intra-tumoral immunosuppression and enhance therapeutic response. While the potential of nanoimmunotherapies in reactivating immune cells has been evaluated in several preclinical studies, the impact of drug-free nanomaterials on the immune system remains unknown. Here, the molecular and functional response of human NK cells and pan T cells to a selection of five NPs that are commonly used in biomedical applications are characterized. After a pre-screen to evaluate the toxicity of these nanomaterials on immune cells, ultrasmall silica-based gadolinium (Si-Gd) NPs and poly(lactic-co-glycolic acid) (PLGA) NPs are selected for further investigation. Bulk RNA-sequencing and flow cytometry analysis showcase that PLGA NPs trigger a transcriptional priming toward activation in NK and pan T cells. While PLGA NPs improved NK cells anti-tumoral functions in a cytokines-deprived environment, Si-Gd NPs significantly impaired T cells activation as well as functional responses to a polyclonal antigenic stimulation. Altogether, PLGA NPs are identified as an attractive strategy for reactivating the immune system of cancer patients.

Keywords: immunotherapy; nanomaterial; proteogenomics.

Figure 1

Nanomaterials internalization varies across human NK cells and pan T cells. A) Infographics illustrating the pipeline of the assessment of nanomaterial impact on primary human NK‐ and pan T cells. Nanomaterial physicochemical characteristics are displayed on the right. B) Nanomaterials impact on NK cells (upper panel) and pan T cells (lower panel) viability measured using CellTiterGlo assay. Data are representative of 3 independent donors. Data are presented as mean ± s.d. C) Representative confocal micrographs of NK cells and pan T cells showing internalized Si‐Gd NPs and PLGA NPs after 4 h co‐incubation at 37 °C. In green Wheat Germ Agglutinin (WGA), in red nanomaterial (Cyanine5.5‐labelled), in blue nuclei (DAPI). Scale bar = 10 µm. D) Flow‐cytometry assessment of nanomaterials internalization in NK‐ and pan T cells after 48 h co‐incubation at 37 °C. Left: Representative histograms of the relative internalization of Si‐Gd NPs (upper panel) and PLGA NPs (lower panel). Grey histogram: untreated control. Right: Quantification of the mean fluorescent intensity signal of the Cyanine5.5‐labelled nanomaterials. Data are representative of 3 to 6 independent donors and analyzed by a One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg after assessment of their Gaussian distribution by Shapiro‐Wilk test. E) Flow‐cytometry assessment of nanomaterials internalization in CD56^High and CD56^Low NK cells after 48 h co‐incubation at 37 °C. Left: Representative flow cytometry contour plots of the relative internalization of Si‐Gd NPs (upper panel) and PLGA NPs (lower panel). Right: Quantification of the mean fluorescent intensity signal of the Cyanine5.5‐labelled nanomaterials. Data are representative of 4 independent donors and analyzed by a Student's t‐test (Si‐Gd NPs) or Mann‐Whitney test (PLGA NPs) after assessment of their gaussian distribution by Shapiro‐Wilk test. F) Flow‐cytometry assessment of nanomaterials internalization in CD4⁺ and CD8⁺ pan T cells after 48 h co‐incubation at 37 °C. Left: Representative flow cytometry contour plots of the relative internalization of Si‐Gd NPs (upper panel) and PLGA NPs (lower panel). Grey histogram: untreated control. Right: Quantification of the mean fluorescent intensity signal of the Cyanine5.5‐labelled nanomaterials. Data are representative of five to seven independent donors and analyzed by a Student's t‐test with Welch's correction (Si‐Gd NPs) or Mann‐Whitney test (PLGA NPs) after assessment of their Gaussian distribution by the Shapiro‐Wilk test.

Figure 2

PLGA NPs transcriptionally prime immune cells activation. A) Transcriptomic impact of Si‐Gd NPs and PLGA NPs treatment on immune cells. Upper panel: Schematic representation of the pipeline of the assessment of nanomaterial impact on primary human NK‐ and pan T cells transcriptome. Lower panel: Principal component analysis of the entire transcriptomic profile of NK cells and pan T cells, untreated and treated with Si‐Gd NPs or PLGA NPs. Each point represents one independent replicate. Axis labels represent the percent of variance as in the respective principal components (dimension 1 and dimension 2). B,C) Violin plots comparing aggregate expression distribution of genes related to TNFα signaling via the NFκB pathway (B) or related to IFNγ signaling (C) according to the MSigDB Hallmark 2024 database. The solid line within each violin represents the median, and dotted lines represent quartiles. Data are analyzed by a Kruskal‐Wallis test with original FDR method of Benjamini‐Hochberg after assessment of their gaussian distribution by Shapiro‐Wilk test. D) NK cells expression of NKFB1 and NFKB2 mRNA expression as fold change relative to untreated control calculated using the 2^−∆∆Ct method (housekeeping: GAPDH). Data are representative of three independent donors and analyzed by One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg after assessment of their Gaussian distribution by Shapiro‐Wilk test. E. Pan T cells expression of NKFB1, NFKB2 and IRF1 mRNA expression as fold change relative to untreated control calculated using the 2^−∆∆Ct method (housekeeping: GAPDH). Data are representative of three independent donors and analyzed by One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg after assessment of their Gaussian distribution by the Shapiro‐Wilk test. F. Heat‐map showing Z‐score values for RNA expression of NK cells activation and function‐associated genes. Gene names are represented on the top, while associations are represented on the bottom. Stars represent significant statistical difference between untreated control NK cells and PLGA NPs‐treated NK cells. G) Flow‐cytometry analysis of NK cells activation. Left: Representative histograms of CD69 expression at the NK cells surface in the different treatment conditions. Right: Quantification of CD69 expression as fold change relative to untreated control. Data are representative of seven to eight independent donors and analyzed by a Mann‐Whitney test after assessment of their Gaussian distribution by the Shapiro‐Wilk test. H) Heat‐map showing Z‐score values for RNA expression of pan T cells activation and function‐associated genes. Gene names are represented on the top, while associations are represented on the bottom. Stars represent a significant statistical difference between untreated control pan T cells and PLGA NPs‐treated pan T cells. G) Flow‐cytometry analysis of pan T cells activation. Left: Representative histograms of CD69 expression at the pan T cells surface in the different treatment conditions. Right: Quantification of CD69 expression as fold change relative to untreated control. Data are representative of eight independent donors and analyzed by a Mann‐Whitney test after assessment of their Gaussian distribution by the Shapiro‐Wilk test.

Figure 3

PLGA NPs enhance anti‐tumor functions of NK cells by promoting perforin polarization at the immune synapse. A) Schematic representation of the pipeline of the assessment of Si‐Gd NPs and PLGA NPs functional impact on primary human NK cells. B) Flow‐cytometry assessment of K562 lysis induced by increasing nanomaterials‐treated NK cells ratio (0.625:1; 1.25:1; 2.5:1, and 5:1) after 4 h of co‐incubation with 10 ng/mL of IL‐15 (left) or without (right). Data are representative of six to eight independent donors and analyzed by a Two‐way ANOVA corrected with the original FDR method of Benjamini‐Hochberg. C) Percentage of CD69‐positive NK cells assessed after co‐incubation with K562 cells for 4 h. Data are representative of 3 independent donors and analyzed by a Two‐way ANOVA corrected with the original FDR method of Benjamini‐Hochberg. D) Mean fluorescence intensity of NKG2D expressed at the NK cells surface after co‐incubation with K562 cells for 4 h. Data are representative of three independent donors and analyzed by a two‐way ANOVA corrected with the original FDR method of Benjamini‐Hochberg. E) Representative conjugate formation assay. Left: Representative contour plots conjugates formed between CTV‐stained NK cells and CFSE‐stained K562 cells after co‐incubation at a 1:1 ratio for 20 min. Right: Quantification of the percentage of conjugates in each condition. Data are representative of three independent donors and analyzed by a two‐way ANOVA corrected with the original FDR method of Benjamini‐Hochberg. F) Nanomaterials‐treated NK cells immune synapse (IS) formation with K562. Left: Schematic representation of IS polarization and perforin distance evaluation. Center: Representative confocal micrographs of NK‐K562 IS. In green Phalloidin‐iFluor488, in white perforin dG9 (Alexa Fluor 647), in red K562 palmitoylated‐tdTomato and in blue nuclei (DAPI). Scale bar = 10 µm. Right: Percentage of polarized IS (upper panel) and mean perforin distance to the IS (lower panel). Data are representative of between 43 and 73 synapses coming from three independent donors displayed as mean per experiment. Data are analyzed by a One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg after assessment of their Gaussian distribution by the Shapiro‐Wilk test. G. Concentration of TNFα (upper panel) and IFNγ (lower panel) released in the supernatant after 4 h of NK cells co‐incubation with K562 cells at a 2:1 ratio. Data are representative of four independent donors and analyzed by a One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg (TNFα) or by a Kruskal‐Wallis test (IFNγ) after assessment of their Gaussian distribution by the Shapiro‐Wilk test. LOD: limit of detection.

Figure 4

Silica‐based gadolinium NPs (Si‐Gd NPs) treatment impairs pan T cells spreading and degranulation. A) Bubble plot showing Z‐score values for protein expression associated with pan T cells functions. Protein names are represented on the top, while associations are represented on the bottom. Bubble size represents ‐Log₁₀(q‐value). Dashed lines and bubbles correspond to non‐significantly deregulated proteins. B) Schematic representation of the pipeline of the assessment of Si‐Gd NPs and PLGA NPs functional impact on primary human pan T cells. C) Pan T cells sensitivity to polyclonal activation. Left: Percentage of CD69‐positive pan T cells assessed after culture on a culture plate coated with increasing concentration of anti‐CD3 antibodies for 24 h. Right: Half maximal effective concentration (EC₅₀) calculated from the dose‐response curve in the left graph. Data are representative of four independent donors and analyzed by a Mann‐Whitney test after assessment of their Gaussian distribution by the Shapiro‐Wilk test. D) Flow‐cytometry assessment of OX‐40 expression at the pan T cells surface after activation with 5 µg mL⁻¹ of anti‐CD3 and 5 µg mL⁻¹ of anti‐CD28 antibodies for 24 h. Left: Representative flow cytometry contour plots of the relative OX‐40 expression in activated pan T cells previously treated for 48 h with Si‐Gd NPs (upper panel) or PLGA NPs (lower panel). Right: Quantification of OX‐40 expression as fold change relative to untreated control. Data are representative of three independent donors and analyzed by a Student's t‐test after assessment of their Gaussian distribution by the Shapiro‐Wilk test. E) Nanomaterials‐treated pan T cells spreading on an antigen‐presenting cells mimicking surface. Left: Representative confocal micrographs of pan T cells spreading. In orange Phalloidin‐iFluor488, in cyan nuclei (DAPI). Scale bar = 10 µm. Right: Pan T cells median spreading area. Left: Relative filamentous actin content in pan T cells as fold change relative to untreated control. Data are representative of 3 independent donors and analyzed by a One‐way ANOVA test with the original FDR method of Benjamini‐Hochberg (Spreading area) or a Student's t‐test (F‐actin content) after assessment of their gaussian distribution by Shapiro‐Wilk test. F) Flow‐cytometry analysis of pan T cells degranulation. Left: Representative histograms of CD107a (LAMP‐1) expression at the pan T cells surface after activation with 5 µg/mL of anti‐CD3 and 5 µg/mL of anti‐CD28 antibodies for 4 h. Right: Quantification of CD107a expression as fold change relative to untreated control. Data are representative of four independent donors and analyzed by a Student's t‐test after assessment of their Gaussian distribution by the Shapiro‐Wilk test. G) TNFα (left panel) and IFNγ (right panel) secretion in the supernatant after activation with 5 µg mL⁻¹ of anti‐CD3 and 5 µg mL⁻¹ of anti‐CD28 antibodies for 4 h. Data are represented as fold change relative to untreated control. Data are representative of four independent donors and analyzed by a Student's t‐test after assessment of their Gaussian distribution by the Shapiro‐Wilk test. H. Flow‐cytometry analysis of CD4⁺ and CD8⁺ T cells degranulation. Left: Representative histograms of CD107a (LAMP‐1) expression at the CD4⁺ T cells (upper panel) and CD8⁺ T cells (lower panel) surface after activation with 5 µg/mL of anti‐CD3 and 5 µg mL⁻¹ of anti‐CD28 antibodies for 4 h. Right: Quantification of CD107a expression as fold change relative to untreated control. Data are representative of four independent donors and analyzed by a Student's t‐test after assessment of their Gaussian distribution by the Shapiro‐Wilk test.