An "eat me" combinatory nano-formulation for systemic immunotherapy of solid tumors

Abstract

Rational: Tumor immunogenic cell death (ICD), induced by certain chemotherapeutic drugs such as doxorubicin (Dox), is a form of apoptosis potentiating a protective immune response. One of the hallmarks of ICD is the translocation of calreticulin to the cell surface acting as an 'eat me' signal. This manuscript describes the development of a stable nucleic acid-lipid particles (SNALPs) formulation for the simultaneous delivery of ICD inducing drug (Dox) with small interfering RNA (siRNA) knocking down CD47 (siCD47), the dominant 'don't eat me' marker, for synergistic enhancement of ICD. Methods: SNALPs loaded with Dox or siCD47 either mono or combinatory platforms were prepared by ethanol injection method. The proposed systems were characterized for particle size, surface charge, entrapment efficiency and in vitro drug release. The ability of the SNALPs to preserve the siRNA integrity in presence of serum and RNAse were assessed over 48 h. The in vitro cellular uptake and gene silencing of the prepared SNALPs was assessed in CT26 cells. The immunological responses of the SNALPs were defined in vitro in terms of surface calreticulin expression and macrophage-mediated phagocytosis induction. In vivo therapeutic studies were performed in CT26 bearing mice where the therapeutic outcomes were expressed as tumor volume, expression of CD4 and CD8 as well as in vivo silencing. Results: The optimized SNALPs had a particle size 122 ±6 nm and an entrapment efficiency > 65% for both siRNA and Dox with improved serum stability. SNALPs were able to improve siRNA and Dox uptake in CT26 cells with enhanced cytotoxicity. siCD47 SNALPs were able to knockdown CD47 by approximately 70% with no interference from the presence of Dox. The siCD47 and Dox combination SNALPs were able to induce surface calreticulin expression leading to a synergistic effect on macrophage-mediated phagocytosis of treated cells. In a tumor challenge model, 50% of mice receiving siCD47 and Dox containing SNALPs were able to clear the tumor, while the remaining animals showed significantly lower tumor burden as compared to either monotreatment. Conclusion: Therefore, the combination of siCD47 and Dox in a particulate system showed potent anti-tumor activity which merits further investigation in future clinical studies.

Keywords: CD47; Immunogenic cell death; SNALPs; calreticulin; doxorubicin.

Scheme 1

SNALPs formulation process. A lipid mixture of CH: DSPC: DOTAP: C16-PEG2000-Ceramide with different molar ratios were dissolved in absolute ethanol at 60 °C. Stock siRNA solutions in H₂O were diluted in 20 mM citrate buffer, pH 4 and heated at 60 °C. The siRNA solution was titrated by the alcoholic lipid solution dropwise under vigorous vortex to ensure the formation of the SNALPs. The obtained SNALPs were incubated at 40 °C for 1 h. For Dox loaded SNALPs, sodium carbonate (1M) was added to adjust the external buffer to pH8 then the SNALPs were incubated with doxorubicin (20% w/w of total lipid) for 2 h at 60 °C. Buffer exchange was then be carried out using ultrafiltration (14,000 g, 45 min) to replace with HEPES buffer at pH 7 for SNALPs with or without doxorubicin.

Figure 1

Physicochemical characterization of SNALPs formulations. SNALPs_siNeg were formulated as described before being incubated with either 0%, 10% or 50% FCS v/v in PBS. Following 24h or 48h incubation SNALPs size (A), surface charge (B) and poly dispersity index (PDI) (C) were measured using dynamic light scattering. To measure stability of RNA within formulations, SNALPs were incubated with 10% and 50% v/v FCS or RNase (100µg/mL) for 24h. RNAse were then inhibited by addition of EDTA before SNALPs were disassembled with heparin (100 IU). Released RNA was qualitatively assessed using gel electrophoresis with free siRNA used as a positive control (D). Drug release from SNALPs was measured by dialysing SNALPs_siNeg in the presence and absence of 50% FCS against PBS (pH 7.4), or in acetate buffer (pH 5.5). Drug concentration in the dialyzate was assessed by measuring the absorbance at 480 nm. As a control, an experiment was set up concurrently in which the same quantity of doxorubicin (Dox) was dialysed against acetate buffer for comparison (E). Statistical analysis was carried out using ANOVA test followed by Tukey HSD test *p<0.05. Data point represent mean and SD (n=3).

Figure 2

Intracellular uptake, cytotoxicity and silencing efficiency of SNALPs in CT26 murine colon carcinoma cells in vitro. CT26 cells were incubated with either the soluble Dox or SNALPs_siNeg-Dox at a range of concentrations for 4, 6, 8 and 24 h before being analyzed by flow cytometry. Cellular uptake across the entire study was assessed by measuring the mean fluorescence intensity (MFI) using flow cytometry (n=3) The uptake of Dox either free (A) of loaded into SNALPs (B) were assessed in CT26 cells over 4, 6, 8 and 24 h. SNALPs demonstrated increased Dox uptake compared to soluble Dox at all concentrations and time points tested. To measure cytotoxicity of the formulation, CT26 cells were incubated with either Dox or SNALPs_siNeg-Dox for 48 h at increasing drug concentrations (0.01- 100 μM). Cell viability was determined by MTT assay and data is presented as viable cells as a percentage of non-treated cells (n= 5) (C). Intracellular delivery of SNALPs_siAtto655 at concentrations 10, 30, 90 nM after 4 and 24 h was assessed using flow cytometry. Representative flow cytometry histograms obtained at the 24 h time point are shown in (D). Quantitative uptake of siRNA expressed as MFI is shown in (E). siRNA uptake was higher at increasing concentrations and incubation times (* p<0.05). The effect of Dox on the uptake of SNALPs_{siAtto655-Dox} is shown in (F). Cells were incubated with SNALPs_{siAtto655-Dox} at increasing concentrations of soluble Dox (10, 30 and 60 nM) and fixed Atto 655-siRNA (30 nM). Uptake was assessed by flow cytometry as described. Co-incubation with soluble Dox increased uptake of siRNA in cells only at an earlier timepoint of 4h (*p<0.05). To evaluate gene silencing, CT26 cells were incubated with SNALPs_siCD47 at three different siCD47 concentrations (0, 10, 30 and 90 nM) at 48 and 72 h. Representative flow cytometry contour plot for 30 nM siRNA at 48 h is shown in (G). Gates were drawn based on isotype controls. Histograms for each concentration of siRNA at 48 h is shown in (H). The knock-down efficiency of CD47 is presented as MFI as percentage untreated control which has been normalized to 100% (I). Data points represent mean and SD (n=3). Statistical analysis was performed using One-way ANOVA followed by Tukey's post-test *p<0.05.

Figure 3

Effect of Dox on calreticulin expression on CT26 murine colon cancer cells in vitro. CT26 cells were pulsed with either cisplatin (5, 10 and 20 µM), Dox or SNALPs_siNeg-Dox (0, 5, 10 and 20 µM Dox) for 4 h before being stained with anti-calreticulin monoclonal antibody and acquired by FACs Calibur flow cytometer. Representative expression of calreticulin is shown in overlaid flow cytometry histograms (A). Cellular expression of calreticulin for each of the groups was calculated from the relative mean fluorescence intensity (MFI) (B).Statistical analysis was performed using ANOVA test followed by Tukey HSD test. Incubating CT26 cells with either Dox or SNALPs_siNeg-Dox significantly increased calreticulin expression when compared to either untreated cells or that incubated with the same concentrations of cisplatin (p< 0.05). To measure acellular ATP, CT26 cells were either left untreated as a control of pulsed with SNALPs_siNeg-Dox. The cell supernatants were removed, and cellular debris cleared by centrifugation. ATP was quantified by luciferase-based ATP detection reagent (C). Statistical analysis was performed using Student T-test followed by Mann Whitney post-test *p<0.05, **p<0.005, ns non-significant. Bars represent mean and SD (n=3-6).

Figure 4

Effect of incubation of CT26 murine colon cancer cells by different SNALPs on macrophage uptake. The ability of both siCD47 and Dox containing SNALPs to alter expression of calretculin on cells was tested using flow cytometry. CT26 cells were pulsed with either PBS (control), cisplatin, SNALPs_siNeg-Dox, SNALPs_siCD47 or SNALPs_siCD47-Dox for 48 h (30 nM siRNA and 5 µM Dox/Cisplatin). Cells were harvested and stained with anti-calreticulin monoclonal antibody prior to acquisition on FACS Calibur flow cytometer^TM. Data is expressed as mean fluorescence intensity and SD (n=3-6) (A). Statistical analysis was performed using a Student T-test followed by Mann Whitney post-test * p<0.05 **p<0.005. CT26 cells were labelled with CellTrace^TM before being incubated with SNALPs_siNeg, SNALPs_siNeg-Dox, SNALPs_siCD47 and SNALPs_siCD47-Dox at concentration 30 nM and 60 nM for siRNA and Dox, respectively, for 48 h. The cells were collected and co-cultured with J774 macrophage cells for 6 h. Cells were harvested and stained with anti-mouse CD45 monoclonal before being acquired on a FACs Calibur flow cytometer. For analysis, cells were first gated on CD45 expression before CellTrace^TM fluorescence was quantified in CD45+ population, a representative histogram for each group is shown in (B). Relative MFI of CD45+ J774 cells is shown in (C). Bars represent the average of mean fluorescence intensity and SD (n=3). One-way ANOVA followed by Tukey's post-test ***p<0.001.

Figure 5

In vivo whole body IVIS imaging and biodistribution of DiR-labelled SNALPs_siNeg in CT26 tumour-bearing BALB/c mice after systemic administration. Mice were inoculated subcutaneously with 1 x 10⁶ CT26 cells in right flank. When the tumour reached ~ 70-80 mm³, mice were i.v. injected with 200 µl of DiR labelled (1 mole%) of total lipid) SNALPs_siNeg in HEPES buffer pH 7 containing approximately 1 nmole siNeg. Animals were imaged using IVIS imaging system, (A) shows representative whole body images using 750/780 nm for DiR labelled SNALPs_siNeg tracking, immediately after injection and at 1, 4 and 24 h post-injection. The tumour site is highlighted. Animals were culled at 24 h post-injection and their organs were excised for analysis (B). Ex vivo organ biodistribution profile of DiR labelled SNALPs_siNeg in CT26 tumour-bearing BALB/c mice normalised to organ weight is shown in (C). Bars represent mean and SD (n=3).

Figure 6

In vivo assessment of SNALPs in CT26 tumour model. Mice (BALB/c n=10 per group) were implanted subcutaneously with 1 x 10⁶ CT26 cells. On day 7 and 17 (dashed lines), mice were i.v. injected with either PBS, SNALPs_siNeg-Dox, SNALPs_siCD47 or SNALPs_siCD47-Dox. Dox was used at (5 mg/kg) while siRNA was used at (0.1 mg/kg). Tumour size was monitored for each mouse until the sacrifice of animals (A). Statistical analysis was performed using 2-way ANOVA followed by Tukey post-test, ***p<0.001, data points represents the mean and SEM. The percentage of mice resolving the tumour was recorded (B). Statistical analysis was carried out using a Mantel-Cox test *p<0.05. Changes in mouse weight relative to starting weight throughout the time course is reported in (C). At the terminal time point, mice were culled and individual organ weight was recorded, spleen weight has been magnified (inset) (D). Statistical analysis was performed using Student's T-test *p<0.05. Bars represent Mean and SEM.

Figure 7

Co-delivery of siCD47 and Dox in SNALPs significantly alters the immunological outcomes of tumour challenge. Following termination of previously described experiment, mice were analysed for several immunological parameters. Cells were extracted from tissues by physical dissociation: tissues analysed included spleens (A, B) and tumours where present and (C, D). Splenocytes were cultured in the presence of brefeldin A for 6 h. Cell surface was stained with anti CD8 before being stained intracellularly with Anti IFNγ-APC. Cells were acquired on a FACs Calibur^TM flow cytometer and gated based on FSC/SSC. Representative flow plots CD8+ fraction population is shown in (A). The corresponding data is expressed graphically in (B). Cells extracted from tumour were stained with Anti CD8α-PE and CD4-FITC and acquired as described. Cells were first gated based on FSC/SSC profile before the relevant marker was assessed. Absolute cell numbers gathered from tumours were first normalised for tumour volume and presented as relative cells/mm3 for both CD8 (C) and CD4 (D). Data analysis was performed using Graphpad Prism *p<0.05 Student's T-test.