Preparation and characterization of VNP-CB⁷

CB⁷ modified VNP was obtained through our previously reported work (Figure S7) ³⁰. Firstly, CB⁷ was conjugated with a lipid molecule,1,2-distearoyl-sn-glycero-3-phosphoethanolamine-poly(ethyleneglycol) (DSPE-PEG) ^31-34, through a thiolene click reaction (Figure S5). The successful conjugation of CB⁷ with DSPE-PEG (DSPE-PEG-CB⁷) was confirmed by the ¹H NMR spectra (Figure S8). Subsequently, VNP-CB⁷ was obtained through the co-incubation of VNP with DSPE-PEG-CB⁷. Finally, the ferrocene was incubated with VNP-CB⁷ through the strong host-guest interactions between CB⁷ and ferrocene to measure the concentration of CB⁷ on VNP. The concentration of CB⁷ on VNP was about 4 nmol of CB⁷ per 10⁷ CFU of VNP via calculating iron content. ADA modified with fluorescein isothiocyanate (FITC-ADA) was used for labeling CB⁷ on VNP to evaluate the stability of membrane decoration of CB⁷ in VNP. After incubation for 6 h, a bright green fluorescence of FITC was observed and still maintained a high intensity of fluorescence for 24 h post-incubation (Figure ​(Figure2G2G and ​and2H),2H), which indicated the high stability of membrane decoration of CB⁷ on VNP.

Preparation and characterization of CaPCAV

CaPCAV was obtained through strong host-guest (CB⁷-ADA) interaction between CaPCA and VNP-CB⁷. VNP exhibited the smooth surface (Figure ​(Figure2I),2I), while CaPCAV exhibited the rough surface due to decoration of CaCO₃ on VNP (Figure ​(Figure2J).2J). The loading amount of CaCO₃ nanoparticles was calculated to be about 1.5 mg per 10⁵ CFU of VNP through weight difference method. For studying the stability of CaPCA on CaPCAV, PI with red fluorescence was used to mark VNP-CB⁷. After incubation for 24 h, CUR with green fluorescence still maintained a high intensity of fluorescence (Figure ​(Figure3A),3A), indicating the well stability of CaPCAV, which was beneficial for in vivo delivery. CaPCAV exhibited the similar proliferation profiles with VNP after incubation for 12 h (Figure ​(Figure3B),3B), indicating the negligible effect of CaPCA on the proliferation of VNP. The negligible size variation of CaPCAV after 7 days' incubation with PBS further confirmed the good hydrodynamic stability of CaPCAV (Figure S9).

Open in a separate window

Figure 3

The cytotoxicity and mechanism of Ca²⁺ overload induced by CaPCAV in vitro. A) CLSM images of CaPCAV after incubation for 0 and 24 h, the scar bar was 5 μm. B) The proliferation profiles of VNP and CaPCAV for up to 12 h. Release profiles of C) Ca²⁺ and D) CUR from CaPCAV in PBS solutions with different pH. E) The viability of LO2 cells treated with CaPA and CaPAV containing the different concentrations of CaP. F) The viability of 4T1 cells after 24 h incubation with various formulations containing equivalent concentration of CUR (20 μgꞏmL^-1), CaP (100 μgꞏmL^-1) and VNP (10⁶ CFU). G) The content of intracellular Ca²⁺ of 4T1 cells incubated with CaPA, CaPCA, CaPCAV containing the equivalent amout of CaP (100 μgꞏmL^-1) and CUR (20 μgꞏmL^-1) after 24 h, respectively. H) The content of intracellular Ca²⁺ in 4T1 cells treated with CaPCAV+PBS, CaPCAV+Iono, and CaPCAV+ BATM, respectively, after 24 h. I) The viability of 4T1 cells after 24 h incubated with CaPCAV+PBS, CaPCAV+Iono, and CaPCAV+ BATM, respectively. J) CLSM images of 4T1 cells after 12 h cellular internalization of CaPCA, CaPCVA. the scar bar was 5 μm. K) Quantitative analysis of the fluorescence intensities of CUR. Data are shown as mean ± standard deviation (SD) (n = 3; *P < 0.05, **P < 0.01, ***P < 0.001).

The acid-responsive release of Ca²⁺ and CUR

The loading ratio of CUR in CaPC was as high as 25.5%. After conjugation of ADA, the loading ratio of CUR in CaPCA was 24%, indicating the high stability of CUR in CaPCA during the process of esterification. CaPCA were immersed in different PBS solution to evaluate their behaviors of acid-responsive decomposition. CaPCA was stable in PBS at pH 7.4 (Figure S7A), slightly decomposed at pH 6.5 (Figure S7B), while almost completely decomposed at pH 5.5 (Figure S7C). CaPCAV also exhibited acid responsive release of Ca²⁺ (Figure ​(Figure3C)3C) and CUR (Figure ​(Figure3D).3D). The minor release of Ca²⁺ and CUR from CaPCAV was observed at neutral environment even after 24 h, while about 89% of Ca²⁺ and 80% of CUR were released from CaPCAV after 4 h under pH 5.5, respectively. The burst release of Ca²⁺ and CUR at acidic environment from CaPCAV was beneficial to Ca²⁺ overload inside tumor cells.

In vitro cytotoxicity induced by Ca²⁺ overload and VNP

The carrier of CaPA and CaPAV had negligible cytotoxicity against LO2 cells even at the high concentrations of 100 μgꞏmL^-1 (Figure ​(Figure3E),3E), indicating the well biocompatibility of CaPA and CaPAV. VNP with anti-cancer activity could prohibit the cell viability of 4T1 cells (Figure ​(Figure3F)3F) ¹⁴. CaPC and CaPCA exhibited stronger cytotoxicity against 4T1 cells than that of free CUR and CaP (Figure ​(Figure3F),3F), due to that the CaP and loaded CUR could synergistically enhance Ca²⁺ overload mediated apoptosis of tumor cells. CaPCA exhibited similar cytotoxicity against 4T1 cells with that of CaPC, indicating the coating of ADA has no effect on the cytotoxicity of CaPC against 4T1 cells. As expected, combing the cytotoxicity of VNP and Ca²⁺ overload mediated apoptosis induced by CaPCA, CaPCAV exhibited the highest cytotoxicity against 4T1 cells among all the studied groups.

The intracellular Ca²⁺ level was investigated via a Ca²⁺ quantitative assay kit to assess intracellular Ca²⁺ overload.²⁹ Compared with CaPA, both CaPCA and CaPCAV exhibited a higher intracellular Ca²⁺ content (Figure ​(Figure3G),3G), due to the Ca²⁺ efflux inhibiting agent of CUR ³⁵.The higher Ca²⁺ content in 4T1 cells treated with CaPCAV than that of CaPCA indicated that VNP could reduce the reflux of Ca²⁺, due to the reduced cell viability by VNP with anti-cancer activity. As a cell-permeant Ca²⁺ chelator, BAPTA-AM (BATM) could bind the intracellular Ca²⁺ and hence decrease the level of intracellular free Ca²⁺. Ionomycin (Iono) is an ionophore, which could specifically delivery extracellular Ca²⁺ into cells, resulting in the increase of intracellular Ca^{2+ 11}. For investigating the effect of Ca²⁺ overload on cells viability, 4T1 cells were incubated with the CaPCVA containing PBS, BATM, Iono and CaCl₂, respectively. The intracellular Ca²⁺ content treated with BAPTA-AM exhibited a dramatic decrease than that of PBS group (Figure ​(Figure3H),3H), which induced the corresponding increase of cell viability (Figure ​(Figure3I).3I). The intracellular Ca²⁺ content treated with Ca²⁺ and Iono exhibited an obvious increase than that of PBS group (Figure ​(Figure3H),3H), which induced the corresponding decrease of cell viability (Figure ​(Figure3I).3I). Compared with high intensity of green fluorescence of CUR in 4T1 cells treated with CaPCA (Figure ​(Figure3J),3J), 4T1 cells treated with CaPCAV exhibited the slightly higher fluorescence intensity of CUR (Figure ​(Figure3K),3K), which might be the reduced cell viability by VNP.

The mitochondria damage induced by CaPCAV

The cell morphology was observed via bio-TEM to investigate the effect of Ca²⁺ overload on cells.³⁶ Compared with the normal 4T1 cells, the CaPA, CaPCA and CaPCAV could induce distinct changes of cell morphology (Figure ​(Figure4A).4A). 4T1 cells treated with CaPCA exhibited more pronouced mitochondria damage than that of CaPA due to the enhanced Ca²⁺ overload induced by CUR, including the chromatin marginalization and densification into dense masses (marked by yellow arrow), the vesicular-like expansion (marked by blue arrows) and swollen mitochondria (marked by red arrows). 4T1 cells treated with CaPCAV showed the most serious mitochondria damage among all the groups, including more dense chromatin and vesicular-like mitochondria, which indicated that the conjugation of VNP could enhance mitochondria damage though intracellular Ca²⁺ overload.

Open in a separate window

Figure 4

The effect of CaPCAV on mitochondria. A) Bio-TEM images, B) MMP, C) mitochondrial distributions and D) cytochrome C of 4T1 cells after incubation with PBS, CaPA, CaPCA and CaPCAV, respectively, for 12 h. The scale bar was 5 μm. E) The expression of cytochrome C and F) relative ATP levels of 4T1 cells after incubation with PBS, CaPA, CaPCA and CaPCAV, respectively, for 12 h. Data are presented as mean ± standard deviation (SD) (n = 3; *P < 0.05, **P < 0.01, ***P < 0.001).

Mitochondrial membrane potentials (MMP) and mitochondrial distributions were also investigated to further evaluate the mitochondria damage ³⁷. MMP were assessed through staining with JC-1. JC-1 is a monomer with green fluorescence at low MMP but forming J-aggregates with red fluorescence at high MMP ^{29, 38}. 4T1 cells treated with CaPA, CaPCA and CaPCAV exhibited lower MMP than that of PBS group (Figure ​(Figure4B).4B). Notably, 4T cells treated with CaPCAV exhibited the most obvious green fluorescence among all the groups, indicating the lowest MMP and most serious mitochondrial damage.

The mitochondrial distribution was assessed through staining with MitoTracker Deep Red FM ³⁹. As shown in Figure ​Figure4C,4C, 4T1 cells treated with different types of CaP exhibited weaker red fluorescence intensity than that of PBS group. The lowest red fluorescence was observed in 4T1 cells treated with CaPCAV, indicating the lowest number of mitochondria and the most serious mitochondrial damage. Cytochrome C could cleave pro-caspase-3 to form activated caspase-3, resulting in apoptosis of cells ⁴⁰. Hence, the expression of cytochrome C were investigated through immunofluorescence staining to evaluate the cell apoptosis induced by mitochondrial damage (Figure ​(Figure4D).4D). 4T1 cells treated with CaPCAV exhibited the largest upregulation of cytochrome C among all the groups (Figure ​(Figure4E),4E), suggesting the cell apoptosis induced by mitochondrial damage.

The mitochondria damage could directly induce the decrease of intracellular adenosine triphosphate (ATP) levels ⁴¹. Therefore, intracellular ATP level in 4T1 cells was assessed through an ATP probe to further evaluate the damage of mitochondria. 4T1 cells treated with CaPCAV had the lowest ATP level among all the groups (Figure ​(Figure4F),4F), indicating the most severe mitochondrial damage.

In vitro immune activation induced by CaPCAV

The antigens could promote the maturation of immature DCs (iDCs) ⁴², hence, the immune activation was evaluated through assessing the level of maturation of DCs ⁴³. As shown in Figure ​Figure5A,5A, firstly, 4T1 cells were cultured in the donor wells (upper wells) with different formulations. Subsequently, iDCs were seeded in the receptor wells (bottom wells) and co-cultured with the upper 4T1 cells for another 24 h. Finally, the DCs were collected to analyze costimulatory molecules CD86 and CD11c, the markers of DCs maturation. In contrast with the free CUR and CaPA group, the ratio of CD86⁺ in CD11c⁺ in CaPCA and VNP group significantly increases (Figure ​(Figure5B),5B), suggesting both VNP and Ca²⁺ overload could trigger immune activation and induce the maturation of DCs. The group of CaPCAV exhibited the largest ratio of CD86⁺ in CD11c⁺ (Figure ​(Figure5B)5B) due to the synergistically enhanced immune activation induced by Ca²⁺ overload and VNP.

Open in a separate window

Figure 5

In vitro immune activation. A) The schematic co-culture system of 4T1 cells and bone marrow-derived iDCs. B) The quantitative analysis of mature DCs (CD86⁺/CD11c⁺, gated on CD11c⁺ cells) after incubation with different formulations for 24 h. C) TNF-α and D) IL-6 expression level in Raw264.7 cells after incubation with different formulations for 24 h. Data are shown as mean ± standard deviation (SD) (n = 3; *P < 0.05, **P < 0.01, ***P < 0.001).

TNF-α and interleukin 6 (IL-6) are the DC activation-related cytokines, hence the level of TNF-α and IL-6 were also investigated to further assess immune activation ⁴⁴. Raw264.7 cells treated with CaPCAV generated a larger amount of IL-6 (Figure ​(Figure5C)5C) and TNF-α (Figure ​(Figure5D)5D) than other groups, indicating the largest level of DC maturation. All the above results confirmed that Ca²⁺ overload and VNP could significantly improve the cellular immune response.

In vivo bio-distribution

The bio-distribution of various formulations was evaluated in 4T1 tumor-bearing mice model (Figure ​(Figure6A).6A). Cy5.5 exhibit higher intensity of fluorescence than that of CUR, hence, Cy5.5 was selected to replace the CUR. For labeling the VNP, the ADA modified liposome loaded with Cy5.5 was conjugated with VNP-CB⁷ through the CB⁷-ADA host-guest interaction. Among all the organs of CaPA group, liver exhibited the largest fluorescence intensity of Cy5.5 (Figure ​(Figure6B6B and ​and6C),6C), suggesting CaP was mainly metabolized through the liver. In VNP group, the larger intensity of Cy5.5 fluorescence in spleen was observed than other organs, indicating VNP was mainly metabolized through the spleen (Figure ​(Figure6B).6B). A large fraction of accumulation of VNP in tumors (Figure ​(Figure6D6D and ​and6E)6E) indicated the excellent targeting ability of VNP. Benefiting from the unique property of hypoxia targeting, CaPAV exhibted a higher fluorescence intensity of Cy5.5 than that of CaPA in tumors, confirming the well tumor-targeting efficiency of CaPAV (Figure ​(Figure6D6D and ​and66E).

Open in a separate window

Figure 6

In vivo bio-distribution and anti-tumor efficacy. A) Fluorescence imaging of the tumor-bearing mice (the scale bar: 10 mm) and B) ex-vivo imaging and C) the quantitative analysis of harvested organs (the scale bar: 3 mm). D) The amount of VNP in harvested organs. E) The corresponding photographs of VNP in tumor. F) The animal study protocol. G) The tumor images, H) tumor growth profiles and I) body weight changes of 4T1 tumor-bearing mice during the treatment with PBS, CaPA, VNP, CaPCA and CaPCAV, respectively. J) Histological observation of the harvested tumor tissues stained with TUNEL, Scale bar: 50 μm. Data is expressed as the mean ± SD; *P <0.05, **P < 0.01, ***P < 0.001.

In vivo anti-tumor therapy

The anti-tumor efficiency was evaluated in 4T1-tumor-bearing mice. The various formulations were injected into the 4T1-tumor-bearing mice on Day 1, 3, 5, 7, 9 via the tail veins, respectively (Figure ​(Figure6F).6F). The tumors of control group grew more rapidly than other groups (Figure ​(Figure6G6G and ​and6H).6H). VNP exhibited a modest anti-tumor effect, due to bacteria-induced innate and adaptive immune responses ²¹. Beneficial from the enhanced Ca²⁺ overload induced by CUR, the CaPCA group exhibited better anti-tumor effects than that of CaPA group. CaPCAV group showed the best anti-tumor effects among all the groups, due to the enhanced Ca²⁺ overload and immune response. Moreover, the moderate change of body weight in CaPCAV group suggested the low toxicity of CaPCAV (Figure ​(Figure5D).5D). The immunofluorescence TUNEL analysis of tumors also confirmed the highest antitumor effects of CaPCAV among all the tested groups (Figure ​(Figure6J).6J). The above results of anti-tumor efficiency confirmed that CaPCAV could synergistically magnify Ca²⁺ overload and immunotherapy.

In vivo immune system activation

The expression level of representative immune cytokines TNF-α and IL-6 was investigated to evaluate the in vivo immune response. PBS group exhibited nearly negligible overexpression of TNF-α (Figure ​(Figure7A)7A) and IL-6 (Figure ​(Figure7B),7B), while CaPA exhibited high expression of TNF-α and IL-6, due to Ca²⁺ overload induced apoptosis. Attribute to the enhanced Ca²⁺ overload induced by CUR, CaPCA group had higher expression of TNF-α and IL-6 than that of CaPA group. VNP group also upregulated the expression of TNF-α and IL-6 due to the induced innate and adaptive immune responses. The TNF-α (Figure ​(Figure7C)7C) and IL-6 (Figure ​(Figure7D)7D) in serum of CaPCAV were also most overexpressed among all the tested groups, which further confirmed that CaPCAV-mediated apoptosis exhibited the highest immune response induced by both Ca²⁺ overload and VNP.

Open in a separate window

Figure 7

In vivo immune response and safety evaluation. A) TNF-α and B) IL-6 quantitative analysis of tumor tissues. The level of C) TNF-α and D) IL-6 in blood samples obtained from the treated mice. E) The ratio of M1/M2 of tumor tissues determined by analysis the ratio of CD11c (an M1 marker) CD206 (an M2 marker). F) WBC, G) AST, LDH and CK, and H) GGT, UREA, CREA and ALT level of blood samples collected from the normal and treated mice. Scale bar: 50 μm. Data is expressed as the mean ± SD (n =3); *P <0.05, **P < 0.01, ***P < 0.001.

The immune response could induce the polarization of anti-inflammatory macrophages (M2) to pro-inflammatory macrophages (M1), hence, ratio of CD11c (an M1 marker) and CD206 (an M2 marker) was used to further analyze the immune response (Figure ​(Figure7E).7E). The largest ratio of M1/M2 was exhibited in CaPCAV group, suggesting that tumor cell apoptosis induced by CaPCAV could significantly improve the immune response. DCs could process antigen materials and present them to T cells, resulting in the activation of immune response. Hence, the T cell level in the tumor was also assessed. The immunostaining of CD4⁺ and CD8⁺ were conducted on the tumor sections (Figure S11), an obviously increased intratumor infiltration of T cells was observed in mice treated with CaPCAV, which confirmed the CaPCAV could enhance the infiltration of immune cell into the tumor tissues. The mice treated with formulations containing VNP exhibited an increased number of white blood cells (WBC) than that of normal mice (Figure ​(Figure7F),7F), due to the mild infections induced by the colonization of VNP.

This study was supported by the National Natural Science Foundation of China (Grant No. 82173042 and 82103248), Shanghai Pujiang Program (2021PJD040) and The Science and Technology Development Fund (FDCT) of Macau SAR (0070/2023/RIA2, 0056/2023/ITP2).