Antigen-specific T cell immunotherapy by in vivo mRNA delivery

Abstract

Immunotherapy has shown promise for treating patients with autoimmune diseases or cancer, yet treatment is associated with adverse effects associated with global activation or suppression of T cell immunity. Here, we developed antigen-presenting nanoparticles (APNs) to selectively engineer disease antigen (Ag)-specific T cells by in vivo mRNA delivery. APNs consist of a lipid nanoparticle core functionalized with peptide-major histocompatibility complexes (pMHCs), facilitating antigen-specific T cell transfection through cognate T cell receptor-mediated endocytosis. In mouse models of type 1 diabetes and multiple myeloma, APNs selectively deplete autoreactive T cells leading to durable control of glycemia, and engineer virus-specific T cells with anti-cancer chimeric antigen receptors (CARs), achieving comparable therapeutic outcome as virally transduced ex vivo CAR. Overall, our work supports the use of APNs to engineer disease-relevant T cells in vivo as Ag-specific immunotherapy for autoimmune disorders and cancer.

Fig. 1.. APNs transfect antigen-specific T cells in TCR-dependent manner.

(A) Schematic showing APN delivery of diverse mRNA cargo to cognate T cells. b-g, Activated NOD8.3 CD8 T cells were transfected in vitro with PBS, K^d/Ctrl (non-cognate control) APNs, or K^d/NRP-V7 (cognate) APNs. After 24 hours, transfection readout was measured. (B) T cells transfected by APNs carrying secreted nLuc mRNA were analyzed via IVIS and quantified. (C) Representative flow plots and frequency bar plot of intracellular BFP expression. (D) Representative flow plots and frequency bar plot of surface-bound VHH expression. One-way analysis of variance (ANOVA) and Tukey post-test and correction for multiple comparisons; ****P<0.0001. All data are means ± SD; n=3 independent wells. (E) Schematic showing the internalization mechanism of APN by T cells through T cell receptor (TCR), not low-density lipoprotein receptor (LDLR). (F) Activated NOD8.3 CD8 T cells were coincubated with either a LDLR blocking antibody (αLDLR Ab) or an isotype antibody control (isotype Ab). APNs were encapsulated with nLuc and transfection was measured by IVIS 24 hours post transfection. (G) Quantification of APN transfection in the presence of the TCR signaling inhibitor dasatinib (dasa), measured via flow cytometry. Two-way ANOVA with Sidak post-test and correction for multiple comparisons. NS= not significant; ****P<0.0001.

Fig. 2.. Casp6 APNs deplete autoreactive T cells in ACT model of T1D and maintain total T cell homeostasis.

(A) Timeline describing T1D model development and treatment; APNs selectively target NOD8.3 T cells T cells in vivo and delivery of Casp6 mRNA triggers apoptosis. (B) Representative flow plots showing NOD8.3 T cells present in the pancreatic lymph node (pLN) after respective treatments. (C) Quantification of % NOD8.3 T cells in peripheral blood, pLN, and spleen after treatment. (D) Total CD8 T cells in the peripheral blood, pLN, and spleen after treatment. Organs isolated with less than 1% viable cells after processing were excluded from analysis. One-way ANOVA with Tukey’s post-test and correction for multiple comparisons, n=4–6 biological replicates. ns = not significant; *, **P < 0.01, ***P < 0.001, ****p<0.0001.

Fig. 3. Casp6 APNs durably prevent onset of hyperglycemia and are well tolerated at given dose.

(A) Timeline and schematic showing treatment strategies; low dose aCD3 and APN treatment was administered at 0.1 mg/kg and high dose αCD3 was administered at 2.5 mg/kg at the specified timepoints. APNs prevent hyperglycemia by selectively depleting autoreactive NOD8.3 T cells and sparing β-islet cell function. (B) Blood glucose traces of mice after receiving respective treatments; light-blue shaded region underneath dashed line represents healthy blood glucose levels (<250 mg/dL). (C) survival curve of mice; survival refers to living mice with blood glucose levels <250 mg/dL. Log-rank (Mantel-Cox) test, n=4–6 biological replicates, NS = not significant. (D) Long term blood glucose traces of Kd/NRP-V7 Casp6 APNs treated mice. (E) Survival curve of mice; survival refers to living mice with blood glucose levels <250 mg/dL. Log-rank (Mantel-Cox) test, n=8–16 biological replicates. (F) measure of liver enzyme alanine transaminase (ALT) and aspartate aminotransferase (ALT) levels and total protein levels in serum following administration of Kd/NRP-V7 Casp6 APNs. One-way ANOVA with Tukey’s post-test and correction for multiple comparisons; means ± SD, n=8–16 biological replicates. ns = not significant, **P<0.01.

Fig. 4. APN transfect human virus-specific T cells from a pretreated multiple myeloma patient with CAR mRNA in vitro.

(A)-(C) Enriched human influenza A virus (IAV)-specific T cells from healthy donor were treated with APNs functionalized with human leukocyte antigen A2.1+ (HLA-A2.1) bound to IAV peptide (HLA/IAV APN). APNs were encapsulated with either nLuc or αBCMA CAR mRNA and transfection was analyzed through IVIS readout of nLuc (A) or flow cytometry detection of αBCMA CAR (B) to show transfection was limited to on-target IAV-specific T cells (C). (D)-(E), HLA/IAV APNs transfect human-IAV specific T cells with functional αBCMA CAR and kill luciferized MM.1R cancer cells in in vitro cocultures. (D) and quantified via bioluminescence IVIS readout (E). One-way ANOVA with Tukey’s post-test and correction for multiple comparisons; means ± SD, n=3 independent wells. NS= not significant, ****P<0.0001. (F)-(G), Frozen peripheral blood mononuclear cells (PBMC) were obtained from HLA-A2.1 positive MM patients who were previously treated with daratumumab and peptide pulsed with peptide epitopes (F), leading to IAV and CMV-specific T cell expansion (G). (H) Human MM patient CMV- and IAV-specific T cells were mixed together and transfected in vitro with HLA/IAV and HLA/CMV APNs encapsulated with αBCMA CAR and αGPRC5D CAR mRNA respectively. Two-way ANOVA with Sidak post-test and correction for multiple comparisons; means ± SD, n=3 independent wells. NS = not significant, ***P<0.001, ****P<0.0001.

Fig. 5. In vivo transfected αBCMA CAR T cells induce anti-cancer efficacy in NSG mice bearing systemic human multiple myeloma (MM).

(A)-(C) Enriched human influenza A virus (IAV)-specific T cells from a healthy donor were intravenously injected into NSG mice and dosed with HLA/IAV APNs carrying αBCMA CAR mRNA (A) and IAV-specific T cells in the spleen were quantified (B). Student’s t-test; means ± SD, n=4–5 biological replicates. NS= not significant. (C) In vivo transfection efficiency with HLA/IAV APNs and HLA/CMV APNs carrying αBCMA CAR mRNA was analyzed via flow cytometry. One-way ANOVA with Tukey’s post-test and correction for multiple comparisons; means ± SD, n= 7 biological replicates. ****P<0.0001. (D) NSG mice were intravenously inoculated with luciferized human BCMA+ U266 MM tumor cells, injected with enriched human IAV-specific T cells, and dosed with HLA/IAV APNs encapsulated with αBCMA CAR mRNA. Tumor burden was measured via luminescence by IVIS (E) and quantified up to 30 days after tumor inoculation (F). Two-way ANOVA with Sidak post-test and correction for multiple comparisons; means ± SEM, n=5 biological replicates. NS = not significant, ***P<0.001, ****P<0.0001.