Acute myeloid leukemia cell membrane-coated nanoparticles for cancer vaccination immunotherapy

Abstract

Cancer vaccines are promising treatments to prevent relapse after chemotherapy in acute myeloid leukemia (AML) patients, particularly for those who cannot tolerate intensive consolidation therapies. Here, we report the development of an AML cell membrane-coated nanoparticle (AMCNP) vaccine platform, in which immune-stimulatory adjuvant-loaded nanoparticles are coated with leukemic cell membrane material. This AMCNP vaccination strategy stimulates leukemia-specific immune responses by co-delivering membrane-associated antigens along with adjuvants to antigen-presenting cells. To demonstrate that this AMCNP vaccine enhances leukemia-specific antigen presentation and T cell responses, we modified a murine AML cell line to express membrane-bound chicken ovalbumin as a model antigen. AMCNPs were efficiently acquired by antigen-presenting cells in vitro and in vivo and stimulated antigen cross-presentation. Vaccination with AMCNPs significantly enhanced antigen-specific T cell expansion and effector function compared with control vaccines. Prophylactic vaccination with AMCNPs enhanced cellular immunity and protected against AML challenge. Moreover, in an AML post-remission vaccination model, AMCNP vaccination significantly enhanced survival in comparison to vaccination with whole leukemia cell lysates. Collectively, AMCNPs retained AML-specific antigens, elicited enhanced antigen-specific immune responses, and provided therapeutic benefit against AML challenge.

Fig. 1. Schematic of AML membrane-coated nanoparticles (AMCNPs) production and anti-leukemic vaccination.

A The immunostimulatory adjuvant, CpG oligodeoxynucleotide 1826, was encapsulated into biodegradable poly (lactic-co-glycolic acid) (PLGA) polymer nanoparticle cores (small gray spheres) via a double emulsion process. Through sonication, CpG-loaded nanoparticle cores were coated with isolated AML cell membrane (red circle), including membrane-associated MHC-I-presented antigens (blue, green, purple, and yellow dots), to form AMCNPs. B Delivery of AMCNPs to immature APCs (blue cell) stimulates maturation and AML-associated antigen presentation. The mature APCs (blue cell) present AML antigens and co-stimulatory molecules to naive T cells (green cells), resulting in activation and proliferation of T cells specific for different AML antigens (blue, green). Activated T cells (green cells) can initiate AML cell death, after detecting the MHC-I-presented antigens on AML cells (red cells).

Fig. 2. Characterization of C1498-OVA cell line.

A MIP vector and MIP-OVA retrovirus constructs used in generation of C1498-MIP and C1498-OVA cell lines. The murine Cadm1 signal peptide (SP) and transmembrane domain (TMD) were cloned 5’ to full-length chicken ovalbumin into the MIP vector. B Representative flow cytometry plots and histogram of C1498-MIP and C1498-OVA cell lines stained with antibodies against MHC class I-presented OVA peptide 257–264 (H-2K^b:SIINFEKL), which demonstrate OVA antigen presentation in C1498-OVA cells compared to C1498-MIP or fluorescence minus one (FMO) negative control staining. C OVA-specific CD8⁺ T cell (B3Z) lacZ activation assay. C1498-MIP or C1498-OVA cells were incubated with B3Z CD8⁺ T cell hybridoma reporter cells, in which OVA-specific T cell receptor activation drives lacZ expression. Representative image demonstrating OVA-specific T cell activation (red color) in B3Z lysates, as assayed with the β-gal substrate chlorophenol red-β-galactoside (CPRG).

Fig. 3. Characterization of AMCNPs.

Uncoated nanoparticles (CpG NP), AMCNPs (C1498, C1498-MIP, and C1498-OVA), and isolated membrane material (C1498, C1498-MIP, and C1498-OVA) were analyzed for size (A) and zeta potential (B) through dynamic light scattering analysis. C Representative transmission electron microscopy images of C1498, C1498-MIP, and C1498-OVA AMCNPs. D Coomassie blue staining of whole cell lysates, isolated membrane material, and AMCNPs from C1498, C1498-MIP, and C1498-OVA cells.

Fig. 4. AMCNPs are taken up by APCs efficiently, stimulate maturation, and promote antigen presentation.

A BMDCs were pulsed for 30 min with free dye-labeled CpG or equivalent C1498-OVA AMCNPs with encapsulated dye-labeled CpG. Representative images show cellular DNA staining by DAPI (blue), labeled CpG (green), and merged. B BMDCs were pulsed for 2 h with C1498-OVA AMCNPs, equivalent C1498-OVA whole cell lysate (WCL) vaccine, or OVA SIINFEKL peptide with CpG. 48 h post-pulsing, CD11c⁺ BMDCs were gated for high expression of the activation markers CD40, CD80, CD86, and MHC-II. Activated CD11c⁺CD40^hi BMDCs were further gated for MHC-I presentation of OVA (H-2K^b:SIINFEKL). Data is presented as the mean percentage of total live BMDCs. C Labeled C1498 AMCNPs or mock controls were injected into C57BL/6 mice via the hock. 24 h post-injection, CD11c⁺ cells in the draining lymph node (dLN) and spleen were examined for presence of labeled C1498 AMCNPs. Representative flow cytometry plots are shown. D Mice received mock, C1498-OVA WCL, or equivalent C1498-OVA AMCNP vaccination. 24 h post-vaccination, CD11c⁺ cells in the dLN were gated for high expression of CD80, CD83, CD86, and MHC-II. Data is presented as mean percentage of total live cells. Significance was determined using one-way ANOVA with a post-hoc test using the Holm-Šídák method.

Fig. 5. AMCNPs enhance antigen-specific T cell activation.

A BMDCs were pulsed with C1498-MIP AMCNPs or C1498-OVA AMCNPs before co-culture with B3Z CD8⁺ T cells. OVA-specific B3Z CD8⁺ T cell activation was measured by a CPRG assay. Data shown as mean optical density at 570–650 nm from three experiments. Significance was determined by unpaired t-test. B–D Mice were vaccinated 3 times as indicated, with C1498-MIP AMCNPs, equivalent C1498-OVA WCL vaccine, or C1498-OVA AMCNPs; splenocytes were collected and re-stimulated ex vivo with OVA SIINFEKL peptide for 7 days (n = 3). C OVA-specific T cell expansion was measured by H-2K^b:SIINFEKL tetramer staining of CD3⁺CD8⁺ T cells. Data is presented as the mean frequency of CD3⁺CD8⁺OVA tetramer⁺ cells among CD3⁺ cells. D The concentration of secreted IFN-γ was measured by ELISA. C, D Significance was determined using one-way ANOVA with a post-hoc test using the Holm-Šídák method. E, F Mice were vaccinated 3 times, as indicated, with C1498-OVA AMCNPs (n = 3), equivalent C1498-OVA WCL vaccine (n = 3), or mock vaccination control (n = 3). Total number of CD69⁺ or CD25⁺ CD8⁺ T cells among peripheral blood (PB) mononuclear cells was determined by flow cytometry on day 17 and normalized to 1 ml of PB. Significance was determined using one-way ANOVA with a post-hoc test using the Holm-Šídák method. G–J Mice were vaccinated 3 times, as indicated, with C1498-OVA AMCNPs (n = 9) or equivalent C1498-OVA WCL vaccines (n = 8). H, I OVA-specific T cell expansion was determined through staining with H-2K^b:SIINFEKL dextramer (OVA-dextramer) of PB mononuclear cells on day 21. H Total CD3⁺CD8⁺OVA dextramer⁺ events observed were normalized to 1 ml of PB and adjusted for background staining by subtracting the average number of events in unvaccinated controls (n = 5). Significance was determined using one-way ANOVA with a post-hoc test using the Holm-Šídák method. I Representative flow cytometry plots of CD3⁺ gated live cells used to quantify CD3⁺CD8⁺OVA dextramer⁺ events are shown. J OVA-specific central memory (CM, CD62L^hiCD44^hiCD8⁺OVA dextramer⁺) and effector memory (EM, CD62L^lowCD44^hiCD8⁺OVA dextramer⁺) expansion was determined through flow cytometry of live splenocytes on day 57. Total events observed were normalized to the total number of live splenocytes collected and adjusted for background staining by subtracting the average number of events in unvaccinated controls (n = 5). Significance was determined using one-way ANOVA with a post-hoc test using the Holm-Šídák method.

Fig. 6. Post-remission AMCNP vaccination promotes long-lasting anti-leukemic immunity and survival benefit.

A Mice were challenged with 1 × 10⁵ C1498 cells, followed by either cytarabine and doxorubicin chemotherapy or mock chemotherapy (n = 7). B Mice were then vaccinated at 26-, 33-, and 40-days post-challenge with C1498 AMCNPs (n = 13) or equivalent C1498 whole cell lysate vaccine (WCL) (n = 15). Unvaccinated “chemotherapy only” mice were used as controls (n = 5). C Surviving mice were re-challenged at day 163 with 2 × 10⁶ C1498 cells. Kaplan–Meier survival plots are shown with significance determined by the Mantel–Cox test. D–G Mice were challenged with 1 × 10⁵ C1498 cells, followed by cytarabine and doxorubicin chemotherapy. Mice were then vaccinated at 26-, 33-, and 40-days post-challenge with C1498 AMCNPs (n = 4) or equivalent C1498 WCL vaccine (n = 4). Mice were re-challenged at day 72 with 2 × 10⁶ C1498-eGFP cells and analyzed at day 93. E Frequency of eGFP⁺ cells among mononuclear cells isolated from the bone marrow (BM) or liver of vaccinated mice. Representative flow plots are shown. F MFI of PD-1 expression among BM and liver CD3⁺ T cells from AMCNP-vaccinated and WCL-vaccinated mice. G Splenic CD3⁺CD8⁺ T cells from AMCNP-vaccinated and WCL-vaccinated mice were analyzed for the frequency of naive T cells (CD62L^hiCD44^low). Significance was determined using unpaired t-tests.