In situ T-cell transfection by anti-CD3-conjugated lipid nanoparticles leads to T-cell activation, migration, and phenotypic shift

Abstract

Ex vivo programming of T cells can be efficacious but is complex and expensive; therefore, the development of methods to transfect T cells in situ is important. We developed and optimized anti-CD3-targeted lipid nanoparticles (aCD3-LNPs) to deliver tightly packed, reporter gene mRNA specifically to T cells. In vitro, targeted LNPs efficiently delivered mCherry mRNA to Jurkat T cells, and T-cell activation and depletion were associated with aCD3 antibody coating on the surface of LNPs. aCD3-LNPs, but not non-targeted LNPs, accumulated within the spleen following systemic injection, with mCherry and Fluc signals visible within 30 min after injection. At 24 h after aCD3-LNP injection, 2-4% of all splenic T cells and 2-7% of all circulating T cells expressed mCherry, and this was dependent on aCD3 coating density. Targeting and transfection were accompanied by systemic CD25⁺, OX40⁺, and CD69⁺ T-cell activation with temporary CD3e ligand loss and depletion of splenic and circulating subsets. Migration of splenic CD8a⁺ T cells from the white-pulp to red-pulp, and differentiation from naïve to memory and effector phenotypes, followed upon aCD3-LNP delivery. Additionally, aCD3-LNP injection stimulated the secretion of myeloid-derived chemokines and T-helper cytokines into plasma. Lastly, we administered aCD3-LNPs to tumor bearing mice and found that transfected T cells localized within tumors and tumor-draining lymph nodes following immunotherapy treatment. In summary, we show that CD3-targeted transfection is feasible, yet associated with complex immunological consequences that must be further studied for potential therapeutic applications.

Keywords: Lipid nanoparticle; Reporter gene; T-cell activation; T-cell transfection; mRNA.

Figure 1.. aCD3-LNPs containing mCherry mRNA transfect, deplete, and activate Jurkat T cells.

A) Transmission electron microscopy images of (i) non-targeted LNPs and (ii) T-cell targeted-LNPs, (iii) the magnified region of targeted LNP, and (iv) the structure of the lipid-nucleic acid assembly within the LNP. The red rectangle in (A-ii) defines the region magnified. The white rectangles in (A-iii) display the conjugated 16% anti-human CD3 on the surface of LNPs. The magnified region in (A-iv) demonstrates the lipid bilayer coating of the LNP with two leaflets at 6 nm apart from each other. The color scale details the electron scattering power (density) in each leaflet of the bilayer compared to the water in the background. Jurkat cells were treated with either non-targeted LNPs (LNPs) or 16% aCD3-LNPs for 24 h or remained untreated (NTC). B) mCherry fluorescence images of Jurkat cells treated with either non-targeted LNPs or 16% anti-human CD3-targeted LNPs including the corresponding amount of mCherry mRNA in the core of the particles. C, D) Percentage of mCherry⁺ cells (C), and the median mCherry fluorescence intensity (D) 24 h following incubation with various treatments or no treatment. E) Depletion of Jurkat cells 24 h after treatment with ISO-LNPs (with IgG on the surface), aCD3-LNPs, or aCD3-LIPOs (not containing mRNA), all with 2% antibody coating on the particle surface compared to no-treatment control (NTC) cells. Next, the transfection efficiency and activation of Jurkat cells treated with 16% aCD3-LNPs were compared with those of mCherry mRNA (mCh), mCherry mRNA complexed with jetMESSENGER (JET), non-targeted (0%) LNPs, 16% aCD3-liposomes (aCD3-LIPOs, not containing mRNA), and 2% aCD3-LNPs. F–H) Percentage of mCherry⁺ cells (F), the median mCherry fluorescence intensity (G), and percentage of CD69⁺ cells 24 h following incubation with various treatments or no treatment (H). Scale bars are 50 or 100 nm in A and 100 μm in B as displayed. All data are plotted as mean ± SD. Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparisons test. ns: not significant. *p < 0.05, **p < 0.01, ***p < 0.001. ψ represents p < 0.0001 of the below group across all other groups unless otherwise denoted with statistical analysis lines above. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 2.. Pharmacokinetics and transfection resulting from aCD3-LNPs reveal splenic and lymph node transfection.

A-C) 16% aCD3-LNPs delivering Fluc mRNA with and without labeling the lipid shell with DSPE-PEG2k-Cy7 were injected in the tail vein in C57BL/6 mice (n = 18). For bioluminescence (BL) imaging of Fluc expression at each indicated time point, mice were injected with D-luciferin at 150 mg/kg body weight 10 min prior to collection of organ/tissue. Ex vivo images were acquired by a Lago optical imaging system, and fluorescence intensity of Cy7 and BL radiance of Fluc expression were quantified. A) Time-course of Cy7 fluorescence intensity of the labeled LNPs in blood over 24 h. B) Fold change of BL radiance of Fluc expression and Cy7 fluorescence intensity in blood at 5 h and 24 h following injection of 16% aCD3-Fluc-LNP-Cy7 to the corresponding control. C) Time-course of mean fluorescence intensity of Cy7 and mean BL radiance of Fluc expression in the liver, spleen, and lymph nodes (LNs). D) C57BL/6 mice (n = 24) were treated with systemic administration of either non-targeted LNPs (LNPs) or 16% aCD3-LNPs delivering mCherry mRNA. At 24 h post treatment, mice were perfused with cold PBS (−/−) and spleens and livers were harvested for RT-qPCR assay. Copy number of mCherry mRNA quantified by RT-qPCR and normalized to the housekeeping GAPDH gene in the liver and spleen 24 h after administration of targeted and non-targeted LNPs compared to the NTC. All data are presented as mean ± SD. Statistical analyses were performed using an unpaired T-test with Welch’s correction in B, one-way ANOVA and two-way ANOVA in C and D, respectively with Tukey’s or Dunnett’s multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ψ represents p < 0.0001 of the below group across all other groups unless otherwise denoted with statistical analysis lines above.

Figure 3.. aCD3-LNPs transfect splenic and circulating CD4⁺/CD8a⁺ T cells.

A-D) mCherry⁺ CD4⁺ (A,C) or mCherry⁺ CD8a⁺ (B, D) T cells as a percentage of CD4⁺ or CD8a⁺ T cells, respectively, at 24 and 48h after control (no treatment, isotype conjugated LNPs, or aCD3 F(ab’)₂) or 1, 2, or 16% aCD3-LNP treatment in the spleen (A, B) or blood (C, D) (n=28). E-F) Normalized count of mCherry⁺ CD3e⁺, CD4⁺, and CD8a⁺ T cells at 24 h after 16% aCD3-LNP treatment from spleen (E) or blood (F). Each replicate was normalized in its respective 1, 2, or 16% group using the following: Normalized Count of Analyzed mCherry⁺ T cells = [(largest T-cell count among replicates in the group) / (T-cell count of replicate)] · mCherry⁺ count of replicate. All data are plotted as mean ± SD. For plots with more than 2 groups, statistical analyses were performed using ordinary one-way ANOVAs with Tukey’s multiple comparison test. Plots with 2 groups were statistically analyzed using an unpaired T test with Welch’s correction, ns: not significant. * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. ψ represents p<0.0001 of the below group across all other groups unless otherwise denoted with statistical analysis lines above.

Figure 4.. aCD3-LNPs activate splenic and circulating CD4⁺/CD8a⁺ T cells with distinct CD69, CD25, and OX40 kinetics (n=45 C57BL/6 mice).

A-D) CD69⁺ CD4⁺ (A, C) or CD69⁺ CD8a⁺ (B, D) T cells as a percentage of CD4⁺ or CD8a⁺ T cells, respectively, at 5 and 24 h after control (no treatment, isotype conjugated LNPs, or aCD3 (Fab’)₂) or 1, 2, or 16% aCD3-LNP treatment in the spleen (A, B) or blood (C, D). E-F) CD25⁺ CD4⁺or CD8a⁺ T cells as a percentage of respective total CD4⁺ or CD8a⁺ splenic (E) or circulating (F) T cells at 5, 24, and 48 h after 16% aCD3-LNP treatment. G-H) OX40⁺ CD4⁺ or CD8a⁺ T cells as a respective percentage of total CD4⁺ or CD8a⁺ splenic (G) or circulating (H) T cells at 5, 24, and 48 h after 16% aCD3-LNP treatment. All data are plotted as mean ± SD. For plots with more than 2 groups, statistical analyses were performed using ordinary one-way ANOVAs with Tukey’s multiple comparison test. Plots with 2 groups were statistically analyzed using an unpaired T-test with Welch’s correction, ns: not significant. *p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. ψ represents p<0.0001 of the below group across all other groups unless otherwise denoted with statistical analysis lines above.

Figure 5.. aCD3-LNPs temporarily deplete splenic and circulating CD4⁺/CD8a⁺ T cells while reducing CD3 surface expression (n=31 C57BL/6 mice).

A-B) CD4⁺ (A) or CD8a⁺ (B) T cells as percentage of total CD45⁺ splenic leukocytes at 5, 24, and 48 h after control or 1, 2, or 16% aCD3-LNP treatment. C) Representative flow cytometry dot plots (down sampled to 15,000 events each) of splenic and circulating lymphocytes at 5 h after 16% aCD3-LNP or NTC treatment showing CD3e ligand internalization as denoted by red arrows. D) CD4⁺ and CD8a⁺ as a percentage of CD3e⁺ splenic T cells at 24 h after control or 1, 2, or 16% aCD3-LNP treatment. E-F) CD4⁺ (E) or CD8a⁺ (F) T cells as a percentage of total CD45⁺ circulating leukocytes at 5, 24, and 48 h after control or 1, 2 or 16% aCD3-LNP treatment. G) CD4⁺ and CD8a⁺ as a percentage of CD3e⁺ circulating T cells at 24 h after control or 1, 2, or 16% aCD3-LNP treatment. All data are plotted as mean ± SD. For plots with more than 2 groups, statistical analyses were performed using ordinary one-way ANOVAs with Tukey’s multiple comparison test. Plots with 2 groups were statistically analyzed using an unpaired T-test with Welch’s correction. CD4⁺ and CD8a⁺ were compared in D and E with an unpaired T-test with Welch’s correction. * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. ψ represents p<0.0001 across all groups unless otherwise denoted with statistical analysis lines above.

Figure 6.. aCD3-LNPs mobilize and activate splenic T cells (n=21 C57BL/6 mice).

Mice were treated with systemic administration of either 16% aCD3-LNPs or non-targeted LNPs, each carrying mCherry, and compared to no-treatment control (NTC). A, B) Spleen weight (A) and size (B) one day after treatment. C, D) Spleen weight (C) and size (D) one-week after treatment. E, F) Histological sections of the spleens harvested one day (E) and one week (F) following each treatment and stained with H&E, aCD8, aCD4, and aOX40. Black and white dashed circles represent the white pulp and the periarterial lymphatic sheaths (PALS) in the spleen sections, respectively. Scale bars are 2.5 mm for the whole section and 200 μm for the magnified section. In A and C, all data are presented as mean ± SD. Statistical analyses were performed using an ordinary one-way ANOVA with Tukey’s multiple comparisons test in A and unpaired T-tests with Welch’s correction in C. ns: not significant. **p<0.01

Figure 7.. 16% aCD3-LNPs promote secretion of T-helper cytokines into murine plasma (n=9 C57BL/6 mice).

A-C) Fold change of Th1 (A), Th2 (B), or T-regulatory, Th17, Th22, and Th25 (C) associated cytokines secreted in murine plasma at 5 h and 24 h after 16% aCD3-LNP administration. Median Fluorescence Intensity (MFI) Fold Change = average MFI of sample / average MFI of PBS buffer associated with that sample. Raw data passed internal quality control such that each replicate’s MFI was plotted only if at least 20 Luminex beads were recovered during acquisition. All data are plotted as mean ± SD. For Luminex plots, statistical analyses were performed with NTC against 5 or 24 h using an unpaired T-test with Welch’s correction. * p<0.05, ** p<0.01, *** p<0.001.

Figure 8.. 16% aCD3-LNPs shift the phenotype of splenic and circulating CD4⁺/CD8a⁺ T cells (n=12 C57BL/6 mice).

A-B) Splenic (A) and circulating (B) CD4⁺ and CD8a⁺ CD44^+/−CD62L^+/− T-cell frequencies at 5, 24, and 48 h after no-treatment control or 16% aCD3-LNP administration. All data are plotted as mean ± SD. Statistical analyses were performed using ordinary one-way ANOVAs with Tukey’s multiple comparison test. * p<0.05, ** p<0.01, *** p<0.001. ψ represents p<0.0001 across all groups unless otherwise denoted with statistical analysis lines above.