Cholesterol removal improves performance of a model biomimetic system to co-deliver a photothermal agent and a STING agonist for cancer immunotherapy

Biological membranes often play important functional roles in biomimetic drug delivery systems. We discover that the circulation time and targeting capability of biological membrane coated nanovehicles can be significantly improved by reducing cholesterol level in the coating membrane. A proof-of-concept system using cholesterol-reduced and PD-1-overexpressed T cell membrane to deliver a photothermal agent and a STING agonist is thus fabricated. Comparing with normal membrane, this engineered membrane increases tumor accumulation by ~2-fold. In a melanoma model in male mice, tumors are eliminated with no recurrence in >80% mice after intravenous injection and laser irradiation; while in a colon cancer model in male mice, ~40% mice are cured without laser irradiation. Data suggest that the engineered membranes escape immune surveillance to avoid blood clearance while keeping functional surface molecules exposed. In summary, we develop a simple, effective, safe and widely-applicable biological membrane modification strategy. This “subtractive” strategy displays some advantages and is worth further development.


Supplementary
. Viability of CTLL2-PD1 cells treated with β-CD. Cells were washed with PBS once and incubated with different concentration of (2hydroxypropyl)-β-cyclodextrin (5 mM, 10 mM, 20 mM and 30 mM) for 30 min, and then washed with PBS once. Cell viability was detected by CCK-8 kit. Data represent mean ± SEM, n = 3 independent samples. Statistical significance was determined by one-way ANOVA test, and it was two-sided and adjustments were made for multiple comparisons. The experiments were repeated three times independently with similar results. Source data are provided as a Source Data file.

Supplementary Figure 18. Tumor targeting ability of PEG-nCISP. a, b
The uptake of PEG-nCISP by RAW264.7 cells detected by flow-cytometry (a) and the mean fluorescence intensity (MFI) of DiD in RAW264.7 cells (b). Data represent mean ± SEM, n = 5 independent samples. c, d The uptake of PEG-nCISP by B16F10 cells detected by flow-cytometry (c) and the mean fluorescence intensity of DiD in B16F10 cells (d). Data represent mean ± SEM, n = 5 independent samples. e, f Fuorescence images of B16F10 melanoma at 12 h post injection of QFN717, nCISP and PEG-nCISP (e), and the mean fluorescence intensity of DiD in tumors (f). Data represent mean ± SEM, n = 3 mice. Student's two-sided t test was used for the statistical analysis in b, d. Statistical significance was determined by one-way ANOVA test in f, and it was twosided and adjustments were made for multiple comparisons. The experiments for a, b, c, d were repeated three times independently with similar results. The experiments for e, f were repeated twice independently with similar results. Source data are provided as a Source Data file. QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; PEG-nCISP, normal-Cholesterol cell membrane with PEG coated ICB agent and QFN717. Figure 19. Tumor targeting ability of CBM. a, b Fluorescence images of B16F10 melanoma at 12 h post injection of nCBM and CBM (a), and the mean fluorescence intensity (MFI) of DiD in tumors (b). Data represent mean ± SEM, n = 3 mice. Student's two-sided t test was used for the statistical analysis in b. The experiments were repeated three times independently with similar results. Source data are provided as a Source Data file. CBM, low-Cholesterol cell membrane sourced from B16F10 cells coated PLGA nanoparticles; nCBM, normal-Cholesterol cell membrane sourced from B16F10 cells coated PLGA nanoparticles.

Supplementary Figure 20. Targeting ability of CRM in an arthritis model. a
Representative image of ankle joints (Scale bar = 0.3 cm), H&E (Scale bar = 100 µm) and Safranin-O and toluidine blue staining of ankle joints (Scale bar = 100 µm). n = 3 mice. b Representative TEM images of nCRM and CRM. Scale bar = 50 nm. n = 3 independent samples. c, d Fluorescence images of model ankle joints at 6 h post injection of nCRM and CRM (c), and the mean fluorescence intensity (MFI) of DiD in model ankle joints (d). Data represent mean ± SEM, n = 4 mice. e, f Fluorescence images of plasma at 0.5 h post injection of particles (e), and the mean fluorescence intensity of DiD in the plasma (f). Data represent mean ± SEM, n = 3 mice. g, h The uptake of nCRM and CRM by monocytes in the blood detected by flow-cytometry (g), and the relative mean fluorescence intensity of DiD in monocytes (h). Data represent mean ± SEM, n = 3 mice. i, j Representative fluorescence images of major organs at 0.5 h post injection of nCRM and CRM (i), and the mean fluorescence intensity of DiD in the major organs (j). Data represent mean ± SEM, n = 3 mice. Student's two-sided t test was used for the statistical analysis in d, f, h. Statistical significance was determined by one-way ANOVA test in j, and it was two-sided and adjustments were made for multiple comparisons. The experiments for b, g, h were repeated three times independently with similar results. The experiments for a were repeated twice independently with similar results. Source data are provided as a Source Data file. nCRM, normal-Cholesterol cell membrane sourced from RAW264.7 cells coated PLGA nanoparticles. CRM, low-Cholesterol cell membrane sourced from RAW264.7 cells coated PLGA nanoparticles. Data represent mean ± SEM, n = 3 mice. Statistical significance was determined by one-way ANOVA test in b, and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. QFN, Quercetinferrum nanoparticles; QFN717, QFN loaded with SR-717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717.

Supplementary Figure 23. Semi-quantitative analysis of nCISP and CISP in serum.
Mice bearing B16F10 melanoma were treated with nCISP and CISP (marked with DiD), and sacrificed at different time points to isolate serum. The semi-quantitative analysis of nanoparticles in serum is based on the fluorescence images. Data represent mean ± SEM, n = 3 mice. Source data are provided as a Source Data file. QFN, Quercetinferrum nanoparticles; QFN717, QFN loaded with SR-717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717. Uptake of nCISP and CISP by monocytes in human blood detected by flow-cytometry (b) and the relative mean fluorescence intensity (MFI) of DiD in monocytes (c). Data represent mean ± SEM, n = 6 independent samples. d, e Representative fluorescence images of plasma (d) and MFI of DiD in plasma (e). Data represent mean ± SEM, n = 6 independent samples. f, g Representative fluorescence images of A375 cells treated with nCAM and CAM (Red) (f), and the relative MFI of DiD in cells (g). Scale bar = 20 μm. Data represent mean ± SEM, n = 5 independent samples. h, i Fluorescence images of plasma at 0.5 h post incubation (h), and MFI of DiD in plasma (i). Data represent mean ± SEM, n = 3 independent samples. Student's two-sided t test was used for the statistical analysis in c, e, g, i. The experiments for b, c, d, e, f, g were repeated twice independently with similar results. Source data are provided as a Source Data file. QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; nCAM, normal-Cholesterol cell membrane sourced from A375 cells coated PLGA nanoparticles. CAM, low-Cholesterol cell membrane sourced from A375 cells coated PLGA nanoparticles. Figure 26. Concentration of C5a in plasma. Mice with B16F10 melanoma were treated with nCISP and CISP (10 mg/kg), and the plasma was collected to detect C5a at 0.5 h post injection. Data represent mean ± SEM, n = 3 mice. Statistical significance was determined by one-way ANOVA test, and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717. Figure 27. Characterization of liposomes. a Images of coomassie brilliant blue staining of SDS-PAGE gel. Liposomes (0.5 mg/mL) with different content of cholesterol were incubated with serum of mice, and the protein absorbed on them were assessed then. b Western blot analysis of complement C3 absorbed on liposomes. c Water contact angle of Liposome 2/10 and Liposome 6/10 at 0.08s. Data represent mean ± SEM, n = 3 independent samples. d TEM images of Liposome 2/10, Liposome 4/10 and Liposome 6/10. Scale bar = 100 nm. The experiments for a, b, c, d were repeated twice independently with similar results. Lip 2/10, liposomes made from cholesterol (2 mg) and phospholipids (10 mg); Lip 4/10, liposomes made from cholesterol (4 mg) and phospholipids (10 mg); Lip 6/10, liposomes made from cholesterol (6 mg) and phospholipids (10 mg). CD11c + CD80 + cells (c) in B16F10 melanoma at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. Statistical significance was determined by one-way ANOVA test in a, b, c, and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. L, laser; QFN, Quercetinferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN. Figure 31. Macrophages in B16F10 melanoma. a Flow-cytometric analysis of CD80 + CD86 + cells in CD11b + F4/80 + cells (M1) in B16F10 melanoma at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. b Flow-cytometric analysis of CD206 + F4/80 + cells in CD11b + cells (M2) in B16F10 melanoma at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. Statistical significance was determined by one-way ANOVA test in a, b, and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. L, laser; QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN.

Supplementary Figure 32. Memory T cells in B16F10 melanoma. a, b, c, d Memory
T cells including CD8 + Tcm (a), CD8 + Tem (b), CD4 + Tcm (c) and CD4 + Tem (d) in tumors were examined by flow cytometer. Tumors were harvested at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. Statistical significance was determined by one-way ANOVA test in a, b, c, d and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. L, laser; QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN.

Supplementary Figure 33. Activated T cells in TdLNs in mice bearing B16F10
melanoma. a Flow-cytometric analysis of CD69 + cells in CD3 + CD8 + cells in TdLNs at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. b Flowcytometric analysis of CD69 + cells in CD3 + CD4 + cells in TdLNs at 6 days post laser irradiation. Data represent mean ± SEM, n = 4 mice. Statistical significance was determined by one-way ANOVA test in a, b and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. L, laser, QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN; TdLNs, Tumor draining lymph nodes. Figure 34. Images of melanoma in mice at 16 days post tumor inoculation. Red cross line represents the dead mouse. Scale bar = 0.5 cm. n = 8 mice. L, laser; QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN. Figure 35. Tumor growth curve of mice bearing B16F10 melanoma. n = 8 mice. Source data are provided as a Source Data file. L, laser; QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717; CIP, low-Cholesterol membrane coated ICB agent and QFN.

Supplementary Figure 36. CISP targeted to MC38 tumors in mice. a, b
Fluorescence images of MC38 tumors at 6 h post treated with of nCSP, CSP, nCISP and CISP (a) and the mean fluorescence intensity (MFI) of DiD in tumors (b). Data represent mean ± SEM, n = 3 mice. The experiments were repeated twice independently with similar results. Statistical significance was determined by one-way ANOVA test in b and it was two-sided and adjustments were made for multiple comparisons. Source data are provided as a Source Data file. QFN, Quercetin-ferrum nanoparticles; QFN717, QFN loaded with SR-717; nCSP, normal-Cholesterol cell membrane coated QFN717; CSP, low-Cholesterol membrane coated QFN717; nCISP, normal-Cholesterol cell membrane coated ICB agent and QFN717; CISP, low-Cholesterol membrane coated ICB agent and QFN717.