(W0930-04-21) Combination Therapy with Carboplatin and miR-5193/149-5p Mimics Against PD-L1 and FOXM1 Axes Loaded in NK-EVs Show Synergistic Effects in Two EGFR-TKI-Resistant (PDX and H1975R) Lung Tumor Models

Purpose: Lung cancer is still the deadliest cancer globally and accounts for 19% of all cancer deaths. MicroRNA (miRNA)-based gene therapy shows great promise as a clinical antitumor therapy, but needs delivery systems for it to be effective. Extracellular vesicles (EVs) secreted by natural killer (NK) cells have been demonstrated to have anticancer potential in our laboratory, were used as potential miRNA delivery systems. Specifically, we investigated the therapeutic potential of IL-15-stimulated NK-EVs loaded with miRNAs (targeting PD-L1 and FOXM1) and in combination with carboplatin in vitro and in vivo using Osimertinib-resistant PDX and H1975R lung cancer cell lines bearing EGFR mutations (L858R and TM00199) to merge gene therapy, immunotherapy, and chemotherapy to enhance anticancer efficacy. The miRNAs were picked from the gene ontology profiling of the NK-EVs and literature survey.

Methods: NK92 cells were grown in PBS-vertical wheel 0.5L bioreactor at the density of 20,000 cells/ml. NK-EVs isolated by ultracentrifugation were characterized by fluorescent-NTA, AFM, and western blotting to determine size, zeta potential, concentration in particles/ml, surface roughness and biomarkers. Proteomics of the EVs and gene ontology studies was done to identify the proteins present in them and miRNAs. Highly manifested differentially expressed genes (DEGs) were identified and picked for further studies. Later, these miRNAs were loaded into NK-EVs using electroporation. For characterization, nanodrop and agarose gel electrophoresis were used. Furthermore, PDX cells and H1975R cells (Osimertinib-resistant) were seeded in 96-well plates for transfection studies in 3D cultures. The plates were treated with various concentrations of miRNA-loaded NK-EVs, Carboplatin and their combination. For in-vivo studies, PDX cells (4×10⁶ cells) were subcutaneously injected into athymic nude mice. The treatment groups were decided based on the results obtained by in-vitro studies and consisted of eight groups i.e., control, NK-EVs (10¹¹ particles/mice), Carboplatin (25 mg/kg), NK-EVs+Carboplatin, miR-5193-NK-EVs (0.5 µg/kg), miR-149-5p-NK-EVs (0.5 µg/kg), miR-5193-NK-EVs+Carboplatin and miR-149-5p-NK-EVs+Carboplatin. For combinations, same dose of treatments was used. Based on the positive results obtained from PDX model study, combination study with miR-5193-NK-EVs and miR-149-5p-NK-EVs in H1975R tumor xenografts was performed with same doses. The tumor volumes were computed using the formula TV = ½ ab², where 'a' and 'b' stand for the tumors' length and width and TV is the tumor volume. The guidelines of the IACUC approved protocol were followed to evaluate the progression of the tumors. The tumor samples at the end of study were investigated for western blotting and RNA sequencing.

Results: NK-EVs exhibited a size distribution of 131.4±4.8 nm and a zeta potential of -36.75±0.04 mV using NTA, meeting exosomal EV criteria. Over 82% of the EVs showed fluorescence indicating presence of tetraspanins CD9/63/81 in the particles confirming them to be exosomes. Morphological analysis with AFM further confirmed their spherical form and consistent size. Western blotting revealed the presence of typical EV markers Alix, TSG101, flotillin and tetraspanins alongside specific NK markers (perforin and granzyme B). Proteomics demonstrated proteins associated with NK cell-mediated cytotoxicity (granzyme B) and T cell activation (perforin and plastin-2). Gene ontology analysis in comparison to control MSC-derived EVs revealed DEGs involved in programmed positive regulation of cell death. Among them, highly expressed miRNAs i.e., miR-5193, miR-3133, miR-149-5p, and miR-193-3p were picked for further studies. 3D Cytotoxicity assays showed that NK-EVs (1*10¹¹ particles/ml) induced cytotoxicity against PDX and H1975R cells, with 62% and 68% cell viability (p < 0.001), respectively. Combination with Carboplatin reduced Carboplatin’s IC₅₀ by ⁓1.9-fold and ⁓1.7-fold in PDX and H1975R cells, respectively. In PDX cells, miR-5193 and miR-149-5p showed 48% and 40% cell death, respectively and in H1975R cells, they showed 44% and 38% cell death, respectively at 50 nM concentration in 3D (p < 0.001) and were superior to other miRNAs. After electroporation of NK-EVs with the above selected miRNAs, agarose gel electrophoresis confirmed their encapsulation (Fig 1). Transfection studies further showed cytotoxic effects against the PDX and H1975R cell lines with miR-5193 and miR-149-5p loaded NK-EVs. In-vivo administration of all treatment groups in PDX model resulted in significant reduction in TV. Although the miR-5193/149-5p-NK-EVs themselves showed reduced TV by ⁓1.5-fold, miR-5193/149-5p-NK-EVs+Carboplatin groups showed the least tumor volume by ⁓3-fold on day 10 (900-1000 mm³ compared to 2750 mm³ of control group, p< 0.001, Fig 2). Western blotting of the tumor samples showed downregulation of PD-1/PD-L1 along with other markers such as Phospho-STAT3, NF-kB-p65, FOXM1, CD1, CDK6 and IL-1ß indicating apoptosis, cell growth inhibition and anti-inflammatory activity (P < 0.001 vs control, Fig 3). In H1975R xenografts, the combination treatments miR-5193-NK-EVs and miR-149-5p-NK-EVs, showed ⁓3-fold reduction in TV as compared to the control which was further increased to ⁓4-fold when combined with Carboplatin (p < 0.001).

Conclusion: Our results demonstrate the feasibility of NK-EVs as cancer immunotherapeutic agents along with being an excellent carrier for miRNAs (miR-5193/miR-149-5p) that proved to be tumor suppressors. The activity was enhanced by their combination with Carboplatin. Further studies are ongoing to understand the molecular mechanisms.

References: Nathani A, Sun L, Khan I, Aare M, Bagde A, Li Y, Singh M. Combined Role of Interleukin-15 Stimulated Natural Killer Cell-Derived Extracellular Vesicles and Carboplatin in Osimertinib-Resistant H1975 Lung Cancer Cells with EGFR Mutations. Pharmaceutics. 2024 Jan 8;16(1):83. doi: 10.3390/pharmaceutics16010083. PMID: 38258094; PMCID: PMC10821370.


Fig 1. Fluorescent Nanoparticle Tracking Analysis (F-NTA) of NK-EVs. A) Size distribution histogram in scatter mode B) Size distribution histogram in fluorescence mode C) Overlay of size distribution histograms in scatter mode and fluorescence mode combined. D) Agarose gel electrophoresis of miRNA samples for qualitative analysis to confirm miRNA encapsulation. E) Western blots of NK cells, IL-15 stimulated NK-EVs and normal NK-EVs showing exosomal markers TSG101, CD81, CD63, Flotillin, Alix and NK specific markers CD56, Perforin, Granzyme B along with negative marker Calnexin and house-keeping β-actin. The values are expressed as mean ± SD (n = 3).


Fig 2. In vivo studies. Line graphs depicting tumor volume of different animal groups in two tumor models A) Line graph showing tumor progression of control, NK-EVs (1011 particles/mice), Carboplatin (25 mg/kg), NK-EVs+Carboplatin, miR-5193-NK-EVs (0.5 µg/kg), miR-149-5p-NK-EVs (0.5 µg/kg), miR-5193-NK-EVs+Carboplatin and miR-149-5p-NK-EVs+Carboplatin groups in PDX lung tumor model from day 0 to day 10 and B) Line graph showing tumor progression of control, Carboplatin (25 mg/kg), miR-5193-NK-EVs (0.5 µg/kg), miR-149-5p-NK-EVs (0.5 µg/kg), miR-5193-NK-EVs+miR-149-5p-NK-EVs+Carboplatin groups in H1975R lung tumor model from day 0 to day 14. The values are expressed as mean ± SD (n = 5).


Fig 3. Effect of NK-EVs, CBP and their combination in PDX tumors in BALB/c nude mice on immune checkpoints, apoptotic, transcription and inflammatory markers: Representative Western blots of different groups and blot quantification of respective PD-L1, PD-1, Phospho-STAT3, NF-kB, FOXM1, CD1, CDK6, IL-1ß and β-actin in the tumor tissue lysate represented by bar graphs. The values are expressed as mean ± SD (n = 3). ns is not significant. * p < 0.05, ** p < 0.01, and *** p < 0.001.