An ER–Horse Detonating Stress Cascade for Hepatocellular Carcinoma Nanotherapy

doi:10.21203/rs.3.rs-2168182/v1

Download PDF

Article

An ER–Horse Detonating Stress Cascade for Hepatocellular Carcinoma Nanotherapy

https://doi.org/10.21203/rs.3.rs-2168182/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Persisting and excessive endoplasmic reticulum stress (ERS) can evoke rapid cell apoptosis. Therapeutic interference of ERS signaling holds enormous potentials for cancer nanotherapy. Herein, a hepatocellular carcinoma (HCC) cell–derived ER vesicle (ERV) encapsulating siGRP94, denoted as ER–horse, has been developed for precise HCC nanotherapy. Briefly, ER–horse, like Trojan horse, was recognized via homotypic camouflage, imitated physiological function of ER and exogenously opened the Ca²⁺ channel. Consequently, the mandatory pouring–in of extracellular Ca²⁺ triggered the aggravated stress cascade (ERS and oxidative stress) and apoptosis pathway with the inhibition of unfolded protein response by siGRP94. Collectively, our findings provide a paradigm for potent HCC nanotherapy via ERS signaling interference and exploring therapeutic interference of physiological signal transduction pathways for precision cancer therapy.

Biological sciences/Cell biology

Biological sciences/Biotechnology

Endoplasmic reticulum vesicles

homologous recognition

stress cascade

Ca2+ homeostasis disruption

endoplasmic reticulum stress

oxidative stress

Endoplasmic reticulum (ER) is an essential cellular compartment for protein synthesis and maturation.^1–4 The ER stress (ERS) signaling caused by accumulation of unfolded or misfolded proteins inside the ER, is involved in the pathogenesis of various human diseases.^5,6 In human tumors, ERS has been shown to be hyperactivated and beneficial for cancer cell survival in harsh environments.⁷ Persistent ERS occurred in tumor cells is manifested as ER homeostasis imbalance, which ultimately impacts cell function, fate and survival.^8–10 Typically, ER homeostasis is restored by unfolded protein response (UPR), eventually adapting to stress and survival. However, persistent or severe ERS can directly lead to cancer cell death.¹¹ Therefore, therapeutic interference of ERS signaling in the cancer cell will be a potential therapeutic modality. To this end, a variety of anticancer agents towards ERS have been explored, which mainly target UPR components or proteasomes, and have shown promise in clinical trials.^{10, 12} Despite these advances, these agents are more suitable for some polysecretory cell types due to the over–dependence on the progress of misfolded protein accumulation.^13–15 It is worth hypothesizing the combination of the early disruption of ER homeostasis and the interference of protein–folding process will realize more effective cancer therapy. Besides, the undesired side effect of the current agents with poor tumor specific distribution also needs to be resolved.¹⁶ Therefore, exogenous nanomedicine intervention for the purpose of activating the body’s cell stress system against cancer has gained significant attention and been sought to join traditional treatment modalities as an adjuvant treatment.

One of the approaches disrupting the ER homeostasis is adjusting the Ca²⁺ homeostasis. Under physiological conditions, the ER regulates intracellular Ca²⁺ homeostasis through an endogenous route.¹⁷ When intracellular Ca²⁺ stores become depleted, STIM1, which senses the ER luminal Ca²⁺ concentration, translocates to vesicular structures (punctae) underneath the plasma membrane (PM). Then, STIM1 interacts with ORAI1 and activates calcium release–activated calcium (CRAC) channels residing in the PM, which enables Ca²⁺ entry into the cytoplasm.^18–20 Intracellular Ca²⁺ homeostasis is closely related to the maintenance of normal cell functions,^{21, 22} while disruptions in this homeostasis decrease protein folding capacity, leading to the accumulation of unfolded proteins and eventually ERS, which usually be relieved by UPR.^23–25 However, if UPR does not work effectively, a specific apoptotic cascade pathway of caspase–12 is activated in damaged cells, accompanied by the trigger of oxidative stress (OS).^{23, 26–27} Inspired by this, an exogenous biovesicle nanosystem featuring particular physiological changes induced by simulation of ER function is desirable to realize precise cancer therapy and mitigate cytotoxicity.

Mannose–rich ER vesicles (ERVs), which retain the specific properties of the parental tumor cells, can specifically target homologous tumors and derivatives for selective binding.^28–33 In particular, ERVs are considered to manipulate the fate of cells under ERS by activating CRAC channels. Inspired by these natural properties, herein, we report a hepatocellular carcinoma (HCC) cell–targeted ERV encapsulating siGRP94, denoted as ER–horse for the precise nanotheraoy. The ER–horse imitates the physiological regulation function of ER to exert induce cell pathological changes via physiological signal transduction pathways (Fig. 1a). For a proof–of–concept demonstration, HCC cells were selected as the homologous target cells due to high expression of mannose receptor (MR). Usually, HCC cells survive from ERS that caused by Ca²⁺ influx or other adverse microenvironments through the activation of UPR, which is an adaptive mechanism capable of reinstating ER homeostasis (Fig. 1b, top). For the precision treatment, ER–horse sourced from HepG2 cells are developed into a smart platform with the homotypic camouflage recognition ability for permitting entry of the vesicle into source cells, like Trojan horse (Fig. 1b, below). After mannose–MR recognition, STIM1 on ERVs couples with ORAI1 on HepG2 cells to activate CRAC channels. Then, the designed ER–horse triggers stress cascade, starting by the pouring–in of the Ca²⁺: 1) Ca²⁺ homeostasis disruption, attributing to Ca²⁺ influx from CRAC channels and causing unfolded protein accumulation; 2) ERS, following unfolded protein accumulation and strengthened by siGRP94 release in the cytoplasm; 3) OS, induced by the excessive ERS and aggravating ERS in turn accompanied by the production of reactive oxygen species (ROS);³⁴ 4) sustained Ca²⁺ homeostasis disruption, deriving from the depletion of ER Ca²⁺ stores and STIM1 protein oligomerization induced by OS, and restarting the circle. This could form a viciously stress cascade, where Ca²⁺ homeostasis disruption, ERS and OS amplify mutually, forcing cells to apoptosis and achieving nanotherapy for HCC (Fig. 1b, below). Therefore, determination of ER–horse with the homotypic recognition–driven targetability and the mechanism triggering the stress cascade is expected to facilitate further tumor therapy development and diverse biomedical applications.

Synthesis and characterization. The ER–horse was prepared with the following three steps (Fig. 1a): 1) ER membrane was isolated from HCC cells (HepG2 cells was used here), 2) ERVs were constructed by extraction and extrusion, and 3) siGRP94 selected as intracellular GRP94 downregulator was encapsulated by extrusion. For control ERVs, only steps 1 and 2 were performed. The mannose expression on ER membrane was verified, which was necessary for the targetability of ER–horse to MR–rich hepatoma cells, by detecting the co–localization of Con A (binding to mannose) and calnexin (CNX) (an ER chaperone) in the HepG2 cells (Supplementary Fig. 1). The overlapped signals suggested that the ER membranes of HepG2 cells had potential for endowing ER–horse with a mannose–rich character.

The composition and physical properties of the ERVs and ER–horse were investigated. First, the primary protein components of the ER membrane, including CNX, calreticulin (CRT) and the putative calcium sensor STIM1, were detected in ERVs by Western Blotting (WB) (Fig. 2a). Moreover, the ERVs were found to retain surface mannose by fluorescence microscopy, which was important for tumor targeting (Fig. 2a, box in upper right). To ascertain the stability of the ERVs in a physiological environment, we performed dynamic light scattering (DLS) and measured the zeta potential of ERVs incubated in PBS at different pH values (5.5, 6.0 and 7.4). The size (Fig. 2b) and zeta potential (Fig. 2c) of the ERVs changed slightly but not significantly in response to pH, indicating that these vesicles are stable in acidic environments such as the tumor microenvironment.

Then, we explored the specific characteristics of ER–horse and ERVs in PBS at physiological pH (7.4). Transmission electron microscopy (TEM) images showed that the ER–horse exhibited a typical globular morphology (Fig. 2d). Nanoparticle tracking analysis (NTA) showed that the average hydrodynamic diameter of the ER–horse was 187 ± 2.8 nm at a concentration of 2.22e + 09 ± 1.56e + 08 particles/mL (Fig. 2e). Moreover, DLS analyses of ER–horse and ERVs performed at pH 7.4 showed that the two vesicle types maintained nearly the same particle size range centered at approximately 200 nm in a physiological environment (Fig. 2f), consistent with the TEM results, indicating that siGRP94 encapsulation had no impact on vesicle stability. Compared with larger nanoparticles, nanoparticles smaller than 200 nm were more inclined to penetrate tumor tissue due to the enhanced permeability and retention (EPR) effect.³⁵ To verify the successful encapsulation of siGRP94 in ER–horse, we next compared the zeta potentials of ER–horse, empty ERVs, and siGRP94. The zeta potential of ER–horse was − 7.39 ± 1.28 mV, which was significantly lower than that of siGRP94 (–3.19 ± 0.88 mV) and ERVs (–3.95 ± 1.22 mV) (Fig. 2g). The lower zeta potential of ER–horse reflected the successful encapsulation of siGRP94 by negatively charged ERVs. Prior to biological evaluation of the ER–horse, it was necessary to confirm the integrated structure of it. This desired structure of ER–horse was confirmed by confocal laser scanning microscopy (CLSM) imaging of the ER membrane (red) and siGRP94 (green) (Fig. 2h). The high fluorescence co–localization rate indicated the successful preparation and integrity of ER–horse.

To further examine the biocompatibility of the ER–horse, serum stability assays were performed in which ER–horse or siGRP94 was incubated with 10% fetal bovine serum (FBS). CLSM imaging showed that ERVs protected siGRP94 from nucleases, as evidenced by the continued detection of fluorescence from ER–horse (green) but not from siGRP94 in tumor cells (Supplementary Fig. 2). These results suggested that ER–horse was successfully prepared, and used as a safe and stable delivery system with high practicality for in vivo application.

Homologous cells–targeting ability of ER–horse in vitro. We hypothesized that the ER–horse constructed from the mannose–rich ER membrane would homologously bind to MR–rich HepG2 cells and then activate CRAC channels (Fig. 3a). As the first step in determining the potential mechanisms responsible for ER–horse–mediated tumor targeting, it was necessary to determine the capacity of ER–horse for targeted recognition of human cell lines, including two liver cancer cell lines (HepG2 and SMMC–7721 cells, hereinafter referred to as 7721 cells), a normal human liver cell line (L02), a normal human breast epithelial cell line (MCF 10A) and a breast cancer cell line (SK–BR–3). Thus, surface MR expression (red) on these cell lines was evaluated by CLSM analysis (Fig. 3b, d). Liver cancer cells (HepG2 and 7721 cells) showed a robust MR fluorescence signal at the cell membrane, with a particularly stronger signal in HepG2 cells than in 7721 cells. The other cell lines showed negligible MR expression, supporting the feasibility of using mannose as the identification element of liver cancer.

To confirm the specific interaction of mannose on ER–horse and MR on HepG2 cells, the co–localization of mannose (green) and MR (red) was investigated after incubating HepG2 cells with ER–horse. The ER–horse treatment group had a higher fluorescence co–localization rate (89.57 ± 3.51%) than the control (untreated) group (18.48 ± 1.61%), indicating the successful targeting of HepG2 cells by ER–horse (Fig. 3c). Subsequently, to verify the requirement for mannose on ER–horse, we blocked the mannose–binding sites on HepG2 cells with free mannose (red) for 1 h. CLSM analysis revealed a lack of cancer cell targeting by ER–horse in the group pretreated with free mannose, demonstrating that the homologous targetability depended on the interaction of mannose on ER–horse and MR on HepG2 cells (Fig. 3e, f).

Next, to verify the ability of ER membrane encapsulation to improve the cellular uptake of ER–horse (green), specific cell lines that differ in their expression of MR (HepG2, 7721, and L02 cells) were investigated at diverse duration (0, 2, 5, 10, and 30 min) (Fig. 3g). CLSM images showed that cells with high MR expression had a strong ability to capture ER–horse. HepG2 cells exhibited strong fluorescence of ER–horse after 10 min of incubation, indicating the time necessary for considerable fusion to occur (Fig. 3g, Supplementary Fig. 3a). At 30 min, the intracellular ER–horse fluorescence intensity was higher in HepG2 cells than in 7721 and L02 cells. The results indicated that the ER–horse was able to specifically identify HepG2 cells rich in MR, which provided the appropriate conditions for subsequent activation of CRAC channels and precise delivery of siGRP94.

To verify that the ability to specifically recognize MR–rich cells was derived from the encapsulation of the ER membrane, we detected the cellular uptake efficiency of vesicles composed of different biomembranes, including ERVs, RBCVs and CCVs (green). Compared to siGRP94–RBCVs and siGRP94–CCVs with a low cellular uptake efficiency, the ER–horse showed markedly greater intracellular fluorescence intensity in a very short time (10 min) (Fig. 3h, Supplementary Fig. 3b). These results indicated that encapsulation by mannose–rich ER membranes could enhance the cellular uptake of ER–horse by HepG2 cells with greater selectivity and specificity.

Furthermore, we observed the coupling of exogenous STIM1 on ER–horse with ORAI1 on the HepG2 cells at the moment of ERV fusion with HepG2 cells (Fig. 3i). The higher fluorescence co–localization rate in the ER–horse–treated group (84.48 ± 3.16%) than the control group (21.8 ± 2.41%) indicated an increase in the binding of STIM1 to ORAI1; moreover, the images showed an increase in STIM1 at the PM upon exposure of HepG2 cells to ER–horse. Taken together, these results indicated that the ER–horse showed precise homologous targetability to HepG2 cells and assembled CRAC channels (STIM1–ORAI1), which might improve the treatment effect.

Intracellular trafficking mechanisms. After binding to the cell membrane, cellular uptake of ER–horse is the next important step in biovesicle delivery. Therefore, we explored the processes of ER–horse fusion with the PM, endosomal escape and siGRP94 release by monitoring the endocytosis and intracellular distribution of ER–horse (ERVs, red; and siGRP94, green) in HepG2 cells (Fig. 4a). To investigate the fusion of ER–horse with the cell membrane, we labeled the HepG2 cell membrane (green) and monitored the fusion process by CLSM. After 10 min of incubation, the fluorescence signal of ER–horse appeared on the HepG2 cell surface (Fig. 4b), indicating fusion between ER–horse and the cell membrane. At 15–20 min, the ER–horse signal gradually increased before gradually moving away from the cell membrane. As the next step necessary for siGRP94 delivery, the co–localization of ER–horse (green) and endosomes (purple) was assessed to verify the transport of ER–horse into the cytoplasm and the fusion of ER–horse with endosomes. CLSM images showed an increase in fusion between ER–horse and endosomes after incubation for 60 min (Fig. 4c). These results demonstrated that ER–horse was taken up into cells through endocytosis as the form of endosome (Endo)–ERVs.

The endosomal release of siGRP94 near the ER is crucial for overall siRNA–mediated silencing efficiency. Successful delivery of siGRP94 from Endo–ERVs to the ER (red) was evidenced by the separation of the siGRP94 signal (green) from the Endo–ERV signal (purple) (Fig. 4d). During the 30–60 min incubation period, the co–localization (white) of siGRP94 (green), Endo–ERVs (purple) and ER (red) increased, indicating an increase in the delivery of siGRP94 to the ER membrane area by Endo–ERVs. After 90 min of incubation, no co–localization was observed, suggesting that all siGRP94 had been released from the endosomes and delivered to the ER. The targeted release of siGRP94 encapsulated by Endo–ERVs implies that ERVs in Endo–ERVs probably plays an important role in promoting the ER uptake of ER–horse due to the similar features; thus, the homology between Endo–ERVs and the ER membrane immediates membrane fusion. Therefore, the ER–horse is a potential strategy for effectively improving cellular uptake and drug release.

To explore the exact course of endosomal reorganization and escape of siGRP94 molecules in the cytoplasm, the specific dynamic process was characterized in further detail by fluorescence imaging of HepG2 cells. In this experiment, fluorescent–labeled dextran (purple), a water–soluble and membrane–impermeable molecule, was added to the culture medium to label endosomes. As dextran enters the cytosol only if the endosomal membrane is damaged, dextran can be used to assess the integrity of the endosomal membrane during intracellular tracing. CLSM images showed the time–dependent appearance of endosomes (purple), siGRP94 (green) and ERVs (red) in both merged and individual channels (Fig. 4e). After 60 min of incubation, Endo–ERVs containing siGRP94 were visible. Over the next 30 min of monitoring, the fluorescent signal of siGRP94 (green) quickly disappeared, and the fluorescence signals for dextran (purple) and ERVs (red) remained visible and co–localized. These results indicated the rapid escape and complete release of siGRP94 in the constructed system without endosomal membrane rupture, reasonably indicating a highly efficient membrane fusion mechanism by taking advantage of the transporting process.

Furthermore, we monitored the trafficking of labeled ER–horse (green) from the cell membrane to other intracellular organelles (red) by CLSM for 0–8 h (Supplementary Fig. 4). The fluorescent signal of ER–horse was exclusively observed at the cell membrane (green) at the initial time point (1 h), with a high Pearson’s correlation coefficient (~ 0.84). However, after 1 h, the fluorescence intensity of ER–horse at the cell membrane declined, and then the signal appeared at the ER membrane (CNX antibody), indicating that the ER–horse effectively transported siGRP94 to the ER area. Given that the main pathway of membrane transport from the cell membrane to the ER membrane is dependent on endosomal vesicles, these observations provide further evidence for the transport of ER–horse through the endosomal vesicular transport system. After siGRP94 delivered to the ER, fluorescence was observed at the mitochondria (MitoTracker) and lysosomes (LysoTracker) at 4–8 h, indicating the subsequent movement of the ERVs. These results showed the fate of siGRP94 and ERVs in the constructed ER–horse after cellular uptake by HepG2 cells.

Stress cascade induced by ER–horse. Inspired by the effective cellular uptake and delivery of ER–horse, we next verified the specific ability of ER–horse inducing the stress cascade (Fig. 5a). As the indicators of OS, we compared the concentration of ROS in the treatment groups of ERVs and ER–horse (Fig. 5b–d). It is clear that the fluorescence of ROS in ER–horse group is higher than that of ERVs after 12 h treatment. Then, we detected the specific mechanism of cell stresses. Briefly, after treatment with ER–horse, Ca²⁺ influx disrupted Ca²⁺ homeostasis and triggered unfolded protein accumulation, followed by ERS and OS. With the downregulation of ERS protective protein GRP94, UPR was inhibited, while ERS and OS were amplificated and further boosted Ca²⁺ influx to from a vicious cycle, eventually inducing cell apoptosis. While in the ERV group without siGRP94, UPR alleviated the ERS and OS by reducing the accumulation of unfolded protein, and made cells survival. Considering Ca²⁺ influx and ROS production played vital roles in the cyclical process; therefore, we performed CLSM and flow cytometry to determine whether ER–horse enhanced cytoplasmic Ca²⁺ and ROS contents in HepG2 cells (Fig. 5b–d). CLSM images of Ca²⁺ and ROS after treatment with ERVs (Fig. 5b, 5d, below) and ER–horse (Fig. 5c, 5d, below) for different durations (0, 0.5, 1, 4, 8, and 12 h) showed similar increasing trends at 0–1 h (Supplementary Fig. 5a). Increases in the fluorescence signals of Ca²⁺ and ROS over the next 11 h were observed in the ER–horse group, whereas these signals decreased in the ERV group after 4 h. We hypothesized that the interaction of STIM1 and ORAI1 caused by the fusion of ER–horse with the HepG2 cell membrane activated the CRAC channels in the early treatment period, inducing an increase in intracellular Ca²⁺; and the changes in ROS fluorescence might be the result of ERS sourced by Ca²⁺ homeostasis imbalance in 0–4 h. Then, the fluorescence intensities of Ca²⁺ and ROS continued to increase in the ER–horse group in 4–12 h, indicating that gene silencing by siGRP94 further aggravated ERS. In contrast, the fluorescence signals of Ca²⁺ and ROS in the ERV group tended to recover due to cell self–recovery. Moreover, the fluorescent changes in the different treatment groups (PBS, siGRP94, ERVs, negative control (NC–) ERVs, and ER–horse) at 1 h verified our hypothesis, that is, the fusion of ERVs to cell membrane could trigger Ca²⁺ influx and increase ROS levels and become parts of stress cascade (Fig. 5d–e, Supplementary Fig. 5b).

To clarify the underlying mechanism by which Ca²⁺ influx is induced by ER–horse in the stress cascade, we monitored intracellular Ca²⁺ mobilization and ROS production in the ER–horse group after cotreatment with different drugs (2–APB, BAPTA and DIDS). The STIM1–ORAI1 calcium channel inhibitor 2–APB and the calcium chelator BAPTA reduced the Ca²⁺ influx induced by ER–horse. A similar effect was not observed following treatment with DIDS, which inhibited the voltage–dependent anion channel type 1 calcium channel on the outer mitochondrial membrane (Fig. 5f, Supplementary Fig. 5c). These data indicated that the increase in cytoplasmic Ca²⁺ originated from both extracellular Ca²⁺ influx and intracellular Ca²⁺ release from the ER in response to the activation of CRAC channels rather than other calcium channels. As for ROS production, cytosolic ROS levels decreased significantly after 2–APB and BAPTA treatment but showed no obvious change after DIDS treatment compared with the control. The trend changes of ROS were roughly consistent with that of Ca²⁺. Moreover, both BAPTA and 2–APB significantly decreased mitochondrial ROS production induced by ER–horse treatment, indicating that elevated cytoplasmic Ca²⁺ contributed to the increase in ROS, which was consistent with the flow cytometry results (Supplementary Fig. 6).

To further assess the specific targetability of ER–horse to HepG2 cells, we detected the fluorescent signals of cytoplasmic Ca²⁺ and ROS in HepG2 cells treated with different siGRP94–vesicles for 1 h and different cells treated with ER–horse for 0, 1 and 12 h. HepG2 cells treated with ER–horse exhibited higher fluorescence intensities of Ca²⁺ and ROS than those treated with other vesicles or other treated cell types (Supplementary Fig. 7–9). The above results showed that ER–horse could lead to the Ca²⁺ influx and the ROS production via activation of CRAC channels (STIM1–ORAI1).

Since the ERS can be induced by the influx of Ca²⁺ and strongthen by siGRP94, we assessed whether different treatments induced ERS. Ultrastructural changes in the ER in HepG2 cells were observed after different treatments (ER–horse, siGRP94, ERVs). Compared with the siGRP94 and ERV control groups, the ER–horse group showed obvious ER swelling and vacuolization after 36 h of treatment, indicating the occurrence of ERS (Fig. 5g).

Next, the expression levels of ERS–related proteins were assessed by WB. First, the concentration of ER–horse between 0 and 10 µM was optimized for 36 h of incubation (Fig. 5h, top left), and 10 µM ER–horse was chosen for the following experiments. The levels of the classic ERS markers GRP78 were higher in the ERV and ER–horse groups than in the PBS and siGRP94 groups, with a greater increase in the ER–horse group than in the ERV group (Fig. 5h, bottom left). Among the four groups, the ER–horse group had the lowest GRP94 levels due to siRNA–mediated silencing; notably, the GRP94 levels did not decrease in the ERV group. Given the induction of ERS, we next determined the levels of the apoptosis–related protein Bcl–2 and the apoptosis regulator BAX to confirm the involvement of apoptosis. The expression of Bcl–2 levels were lower while no significant change in Bax in HepG2 cells treated with ER–horse group compared with those of PBS, siGRP94, or ERVs group. To verify the effect of siGRP94 in inhibiting the UPR, we determined GRP94 and GRP78 protein expressions in the two groups (the ERV and ER–horse groups) of HepG2 cells for different times (0, 2, 4, 8, and 12 h) (Fig. 5h, right). Treatment with ER–horse evoked a continued increase in GRP94 protein from 0–4 h due to ERS caused by subsequent Ca²⁺ influx and ROS production. At 8–12 h, siGRP94–mediated silencing induced decreases in GRP94 levels, which further aggravated ERS. Simultaneously, the corresponding levels of the ERS–related protein GRP78 increased in a time–dependent manner throughout the time course. Conversely, in the ERV group, GRP94 levels peaked at 8 h and then declined, which was consistent with the change in GRP78 (Fig. 5h, bottom right). This difference was hypothesized to stem from the absence of siRNA–mediated silencing, which might allow the activation of cell self–regulation and recovery.

Therapeutic effect in vitro. To further explore the apoptotic pathway induced by ER–horse, a series of apoptosis–related proteins (caspase–12, CHOP, TRAF2, Cl–caspase–3) expressions were investigated. WB analysis demonstrated the high expression of cascade pathway of caspase–12 related proteins, indicating cells apoptosis through a specific apoptotic cascade pathway of caspase–12. (Fig. 6a). ER–horse can augment nanotherapy efficiency by triggering ERS aggravation and GRP94 downregulation to disrupt the self–repair function of cancer cells. To corroborate the antitumor potential, the induction of apoptosis in response to ER–horse treatment was assessed by CLSM. HepG2, 7721 and L02 cells were treated for 36 h with ER–horse, and propidium iodide (PI) staining was used to identify apoptotic cells. Compared with 7721, L02 and HepG2 control cells (no ER–horse treatment), HepG2 cells treated with ER–horse showed higher fluorescence signals, indicating that ER–horse induced the apoptosis of HepG2 cells (Fig. 6b). Additionally, Cell Counting Kit–8 (CCK–8) experiments were performed from 0 to 36 h to compare the survival rates of HepG2 cells subjected to different treatments (PBS, siGRP94, ERVs, and ER–horse) (Fig. 6c), HepG2 cells treated with different vesicles (ER–horse, siGRP94–RBCVs, and siGRP94–CCVs) (Fig. 6d) and different cells (HepG2, 7721, and L02 cells) treated with ER–horse (Fig. 6e). The survival rates of HepG2 cells with different treatment groups illustrated that the cell apoptosis induction by ER–horse was the best among other control groups (with survival rates of 45.66% for ER–horse group, 83.76% for ERVs group and 92.2% for siGRP94 group). In addition, compared with siGRP94–RBCVs and siGRP94–CCVs, ER–horse clearly had the greatest antitumor effects (with survival rates of 46.6% for ER–horse group, 74.77% for siGRP94–RBCVs group, and 83.49% for siGRP94–CCVs group) on HepG2 cells due to their tumor–targeting ability and the presence of STIM1 on the surface. Obviously, the antitumor effects of ER–horse on HepG2 cells were superior to those on the other two cell types due to their different uptake efficiencies of ER–horse (with survival rates of 44.17% for HepG2 cells, 70.63% for 7721cells, and 84.99% for L02 cells). The results suggested that ER–horse decreased HepG2 cell viability and induced cell apoptosis in a time–dependent manner. Furthermore, to verify the effect of the activation of CRAC on the antitumor effect, we assessed HepG2 cell survival after pretreatment with 2–APB, BAPTA, or DIDS, and then treatment with ER–horse (Fig. 6f). The results showed that cell survival increased significantly after 2–APB pretreatment, providing further evidence that apoptosis was elicited by the CRAC channel activation after cellular uptake of ER–horse. Consistent with CCK–8 analysis (Fig. 6c), the flow cytometry results revealed that the treatment with ER–horse led to a significant increase in the percentage of apoptotic tumor cells compared to those in the control groups (with apoptosis rates of 53.85% for ER–horse group, 13.76% for NC–ERVs group, 9.54% for ERVs group, 7.55% for NC group, 6.49% for siGRP94 group) (Fig. 6g).

Tumor homologous targetability and therapeutic efficacy in vivo. Inspired by the distinct therapeutic effects mediated by ER–horse in vitro, we systematically investigated its antitumor performance in vivo. Prior to this analysis, the in vivo biodistribution of ER–horse was inspected in a HepG2 cell xenograft model in BALB/c nude mice (Supplementary Fig. 10). Successful generation of this xenograft model was confirmed by visual observation of tumor nodules 14 d after the inoculation with HepG2 cells. HepG2 tumor–bearing nude mice were intravenously administered the same doses of ER–horse, siGRP94–RBCVs or siGRP94–CCVs labeled with the near–infrared dye DIR. Fluorescence imaging of the treatment groups was performed at designated time points (0, 4, 8, 12, 24, 36, and 48 h) to obtain an initial estimate of vesicle distribution (Supplementary Fig. 10a, c). The images showed that siGRP94–FITC was quickly cleared from the circulation and rarely reached the tumor site. Conversely, a strong and sustained (from 36–48 h) fluorescence signal was observed at tumor sites in mice injected with ER–horse, whereas there was negligible fluorescence at the tumor site at 48 h in the mice treated with the other vesicles. These results suggested that encapsulation by ERVs decreased the clearance of siGRP94 and prolonged its time in circulation and in vivo target residence time; these effects are beneficial for subsequent antitumor effects.

To intuitively analyze the distribution of fluorescence in different organs ex vivo intuitively, the mice in each group were euthanized after 48 h, and major organs (e.g., the heart, liver, spleen, lungs, and kidneys) and tumors were collected and subjected to fluorescence imaging to analyze vesicle accumulation. ER–horse accumulated in tumors to a significantly greater extent than the other three vesicle types (Supplementary Fig. 10b, d); these data demonstrated the strong tumor–targeting ability of ERVs, which showed enhanced binding to liver cancer cells and thus increased siGRP94 accumulation in tumors. The fluorescence intensities in liver and spleen were relatively high in all vesicle groups (ER–horse, siGRP94–RBCVs, and siGRP94–CCVs). We speculated that glycoproteins on biovesicles may be recognized by cognate receptors on cells within these organs, leading to accumulation of fluorescence signal, but the specific mechanism should be further assessed.

To confirm that ER–horse induced severe Ca²⁺ imbalance and ROS production at tumor site, frozen tumor sections from mice treated with PBS (control), siGRP94, ERVs, or ER–horse of 48 h were stained with Fluo–4 AM (calcium) and dihydroethidium (DHE, ROS). In comparison to the PBS, siGRP94 and ERV groups, the ER–horse group showed the stronger fluorescence intensities of Ca²⁺ and ROS in tumor sections, indicating that the large–scale Ca²⁺ influx and ROS production occurred in this group (Fig. 7a).

Then, immunohistochemistry (IHC) and histological analyses of tumor sections treated above were performed to evaluate preliminary indicators of the potential therapeutic efficacy of ER–horse. In ER–horse group, the levels of ERS marker protein GRP78 were clearly increased, but those of other ERS marker GRP94 were not, probably due to gene silencing by the delivered siGRP94. IHC results suggested that ER–horse induced apoptosis by downregulating a protective UPR–related protein (GRP94) after triggering ERS, as indicated by the increased GRP78 expressions (Fig. 7b). In the hematoxylin and eosin (H&E) staining, tumors treated with ER–horse had the largest necrotic areas, whereas tumors in the siGRP94 group and ERV group were similar to those in the control group, with some indication of normal morphology (Fig. 7c). Sections of the major organs were also analyzed by H&E staining and negligible morphological damage of major organs was observed in any of the groups (Supplementary Fig. 11). Similar to H&E staining of tumors, significant green fluorescence indicating apoptotic cells was observed in the tumors incubated with ER–horse compared with from the other groups in terminal deoxynucleotidyl transferase dUTP nick–end labeling (TUNEL) staining assays, (Fig. 7d). Taken together, the findings indicated the potential of ER–horse to inhibit the growth of tumors derived from HepG2 cells.

To obtain further insight into the antitumor activity, body weight and tumor volume were measured to monitor the therapeutic efficacy (Fig. 7e and f). Balb/c nude mice bearing HepG2 tumors were tail vein injected with ER–horse, ERVs, siGRP94 or PBS (labeled as control group) every other day. Negligible body weight abnormalities were monitored, suggesting the low biological toxicity of these groups. A mild tumor volume variation was observed in ERVs and siGRP94, indicating their weak antitumor performances. As expected, ER–horse exhibited the remarkable inhibition of tumor growth, attributing to its homologous targetability and stress cascade capacity. Consistently, ER–horse led to the highest tumor growth inhibition rate (80.9%) among these groups, indicating that ER–horse drastically inhibited tumor growth (Fig. 7g). Moreover, the excised tumor with ER–horse incubation presented the smallest tumor size (Fig. 7h). H&E staining of tumor tissues also suggested that the ER–horse presented large necrotic regions when compared with the other control groups (Fig. 7i). Taken together, as the homologous biovesicles and stress cascade inducer, ER–horse enabled tumor–specific accumulation and eventually potentiated precise HCC nanotherapy harnessing biovesicle surface functions.

In summary, we constructed a homologous ER–horse boosting stress cascade for precise and potent HCC nanotherapy. The ER membrane endowed ER–horse with mannose, thus enabled homotypic recognition with MR on HepG2 cells. After recognition and binding, CRAC channels were activated by the coupling of Ca²⁺–sensing STIM1 on ER–horse and Ca²⁺–selective ORAI1 on HepG2 cells, causing Ca²⁺ influx to intracellular Ca²⁺ homeostasis disruption. Meanwhile, ER–horse facilitated the endosomal escape of siGRP94 through a highly efficient membrane fusion mechanism, and siGRP94 was released into the ER area without endosome destruction due to the reorganization of Endo–ERVs. With the entry of Ca²⁺ and the downregulation of GRP94 through siRNA–mediated gene silencing, ER–horse successfully triggered the stress cascade (Ca²⁺ homeostasis disruption–ERS–OS), and eventually achieved precise HCC nanotherapy. This work provides a novel and general strategy for construction of versatile ER membrane–based vesicles that can be used to trigger stress cascade with homologous targetability effects for therapy and may suit the needs of diverse biomedical applications.

Materials. All materials can be found in the accompanying Supplemental Information file.

Cell culture. HepG2 human liver cancer cells, SMMC–7721 human liver cancer cells (hereinafter referred to as 7721 cells), L02 human liver cells, SK–BR–3 human breast cancer cells, MCF 10A human breast epithelial cell cells, and human RBCs were provided by the Chinese Academy of Sciences Cell Bank and Chongqing Traditional Chinese Medicine Hospital. The cells were cultured with DMEM/F12 supplemented with 10% fetal bovine serum, 100 units/mL penicillin, and 100 µg/mL streptomycin at 37°C in a humidified atmosphere of 5% CO₂.

Detailed synthesis procedure of ER–horse. The ER membrane was isolated with an ER isolation kit according to the manufacturer’s protocol. Briefly, the cells were washed with 10 volumes of PBS (0.01 M, pH 7.4) and centrifuged at 600 × g for 5 min three times, and the supernatant was removed by aspiration. Then, the packed cell volume (PCV) was measured, and the cells were suspended in a volume of 1× Hypotonic Extraction Buffer (an aliquot of 10× Hypotonic Extraction Buffer was diluted 10–fold with water. Just before use, Protease Inhibitor Cocktail for mammalian cells was added to the 1× Hypotonic Extraction Buffer at a concentration of 1% (v/v)) equivalent to 3 times the PCV and incubated for 20 min at 4°C, which allowed the cells to swell. Next, the cells were centrifuged at 600 × g for 5 min, the supernatant was removed by aspiration, and the “new” PCV was measured. Then, a volume of 1× Isotonic Extraction Buffer (created by aseptically removing an aliquot of 5× Isotonic Extraction Buffer and diluting it 5–fold with water. The 1× Isotonic Extraction Buffer was kept at 4°C before use. Just before use, Protease Inhibitor Cocktail for mammalian cells was added to the 1× Isotonic Extraction Buffer at a concentration of 1% (v/v)) equivalent to 2 times the “new” PCV was added, and the mixture was transferred to a 7 mL Dounce homogenizer. The homogenizer was used to break the cells with 10 strokes. Finally, the homogenate was centrifuged at 1,000 × g for 10 min at 4°C. The thin floating lipid layer was carefully removed by aspiration. The supernatant was transferred to another centrifuge tube using a pipette, and the pellet was discarded. The supernatant was centrifuged at 12,000 × g for 15 min at 4°C. The thin floating lipid layer was carefully removed by aspiration. The supernatant was transferred to another tube using a pipette, and the pellet was discarded. The obtained product was stored at 4°C. According to the calculations, 1.2 × 10^–2 mg of ER was extracted per disk. After mixing with siGRP94 (dissolved in 0.01 M PBS; the final concentration was 10 µM), the suspension was kept at 4°C and extruded five times through a polycarbonate filter of 1.0 µm, five times through a polycarbonate filter of 0.5 µm, and five times through a polycarbonate filter of 0.2 µm using an Avanti Hand Extruder (Avanti Polar Lipids). The final products were characterized by TEM, DLS, CLSM and WB.

Immunofluorescence. Cells (5 × 10⁵) were overlaid on sterilized coverslips in 24–well plates with 2 mL of medium. After 24–48 h, the cells were washed three times with 1× PBS and then fixed with 4% (v/v) paraformaldehyde for 10 min in the dark. The cells were then permeabilized with 0.1% (v/v) Triton X–100 for 10 min and subsequently blocked using 2% (w/v) bovine serum albumin for 1 h. The cells were incubated with a CNX primary antibody at a 1:200 dilution. Cells were washed three times with 1× PBS and then stained with goat anti–rabbit IgG–DyLight 549 (1:400 dilution) and Conanavalin A–FITC (Con A–FITC, 1 µg/mL, containing 1 mg/mL Ca²⁺ and Mn²⁺) for 60 min. DAPI (10 µg/mL) was used as the nuclear stain. A Zeiss LSM510 confocal microscope was used to visualize the cells under a 40× /1.3 oil DIC M27 objective. ImageJ and Leica software were used for image analysis.

Characterization of ER–horse. ER–horse (1.2 × 10^–2 mg/mL, quantified by ERVs) and ER (treated with 10 µM siGRP94, 1.2 × 10^–2 mg/mL ERVs and 1.2 × 10^–2 mg/mL ER–horse for 0 h and 36 h) were characterized by using TEM. Briefly, samples were examined using a Jeol JEM–2100 transmission electron microscope operated at 200 kV. The samples were applied to copper grids for 2 min. The samples were then air–dried and stained using a 2% w/v uranyl acetate solution (Electron microscopy services). A NanoSight NS300 machine (Malvern Instruments, Malvern, UK) equipped with a sample chamber and 640 nm laser was used to analyze the size distribution of ER–horse. ERVs samples (1.2 × 10^–2 mg/mL) were disaggregated using a needle and syringe before injection into the NanoSight sample cuvette. The frame rate was 30 frames per second. Data analysis was performed using NanoSight NTA 3.2 software.

WB analysis was performed to identify the protein profiles of ER–horse. Briefly, equal amounts of the total proteins from different samples quantified with a BCA Protein Assay Kit were added to a 10% SDS–polyacrylamide gel electrophoresis (SDS–PAGE) gel to separate proteins of different molecular weights. Subsequently, the gel was stained with Coomassie brilliant blue and imaged. For the WB assay, the proteins on the gel were then transferred to a polyvinylidene fluoride membrane, and then the membrane was blocked with 5% skimmed milk and incubated with primary antibodies against CNX, CRT, and STIM1. The membrane was incubated with an anti–goat IgG secondary antibody corresponding to the primary antibodies before imaging was performed with ECL Western Blotting Substrate.

The integrative hybrid nanostructure of ER–horse was further determined by the colocalization of the ER membrane and siGRP94. Briefly, siGRP94 was labeled with FITC, and the ER membrane was labeled with antibodies (an anti–CNX primary antibody and a DyLight 549–labeled secondary anti–rabbit IgG antibody). Then, two–fluorescence dye–labeled ER–horse (FITC–siGRP94: 10 µM; ERVs: 1.2 × 10^–2 mg/mL) was assembled. Four microliters of ER–horse was dropped onto the slide, and fluorescence images were obtained using a Leica SP8 CLSM (Wetzlar, Germany) at different excitation wavelengths (DyLight 549: 549 nm; FITC: 488 nm).

For size distribution evaluation, nanoparticle tracking analysis (NTA) was performed with a NanoSight NS300 (Malvern, UK) equipped with a 405 nm laser and an sCMOS camera. Resuspended ER–horse (1.2 × 10^–2 mg/mL, quantified by ERVs) were diluted 50–fold in filtered PBS to achieve a final concentration of 2.4 × 10^–4 mg/mL. The capture length was 60 s with the camera level set to 14 and the detection threshold set to 3. An image of filtered PBS was taken to verify that the diluent had no particles. A total of 1498 frames were captured and analyzed. NTA 3.2 Dev Build 3.2.16 software (Malvern, UK) was used for capturing and data analysis.

The hydrodynamic size and zeta potential of ERVs (1.2 × 10^–2 mg/mL, quantified by ERVs), ER–horse (1.2 × 10^–2 mg/mL, quantified by ERVs) and siGRP94 (10 µM) diluted in PBS (0.01 M, pH 5.5, pH 6 and pH 7.4) were measured by DLS using a Malvern Zetasizer Nano ZS. The particle size versus the number of particles per mL and the average zeta potential were plotted using Origin 9.0 software (average of 3 biological replicates).

To investigate the encapsulation of siGRP94 in ER–horse, the protective effect of ER–horse on siGRP94 in serum–containing medium was detected. The ER–horse (siGRP94: 10 µM, labeled with FITC) and siGRP94 (10 µM, labeled with FITC) were incubated with DAPI–labeled HepG2 cells for 1 h in serum–containing medium (containing 10% FBS). The cells were fixed with 4% paraformaldehyde. The protective effect of the ERVs was analyzed by CLSM following mounting with antifade polyvinylpyrrolidone mounting medium.

Homologous–targeting Ability Of Er–horse

Determination of the abundance of MR in different cell lines. To explore the homologous targeting ability of ER–horse in different cell lines, it was necessary to detect the abundance of MR on the cell membranes. Briefly, different cell types (HepG2 liver cancer cells, 7721 liver cancer cells, L02 liver normal cells, MCF 10A normal breast epithelial cells and SK–BR–3 breast cancer cells) were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured overnight. The cells were then blocked in 5% serum for 30 min and probed with primary antibodies (CD206, diluted at 1:400) overnight at 4°C. After primary antibody incubation, the cells were washed 3 times with PBST (0.01 M PBS containing 0.05% (v/v) Tween–20) and incubated with the appropriate DyLight 549–conjugated secondary antibody (anti–mouse IgG antibody, diluted at 1:400) for 1 h at room temperature. Then, the cells were washed 3 times with PBST, and the nuclei were counterstained with DAPI. Photographs were taken with a confocal microscope (FV1000, Olympus, Japan) and analyzed using LAS X software.

Interaction of mannose and MR. ER–horse was incubated with Con A–FITC (10 µg/mL) at 37°C for 1 h in the dark and added to HepG2 cells. After incubation for 30 min, the HepG2 cells were fixed with 4% paraformaldehyde and then labeled with a primary antibody against CD206 and a secondary antibody conjugated with DyLight 549. The nuclei were counterstained with DAPI. CLSM (Olympus FV 1000) was used to observe the fluorescence signals. Pearson’s coefficient was calculated using the JACoP plugin of ImageJ 1.46r software.

Blocking with free mannose. To block the capturing effect of MR on the surfaces of HepG2 cells, cells were incubated with excessive free mannose. First, HepG2 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured overnight. The cells were then incubated with free mannose (2 mg/mL) at 37°C for 1 h. Then, the blocked cells were incubated with ER–horse (working concentration 10 µM) at 37°C for 30 min, washed with PBS to remove the unbound free mannose and used for the indicated assays. The cells were fixed with 4% paraformaldehyde. The ability of HepG2 cells to capture ER–horse was analyzed by CLSM following mounting with antifade polyvinylpyrrolidone mounting medium.

Comparison of the homologous targeting of ER–horse to different cells. ER–horse was incubated with Con A–FITC (10 µg/mL) at 37°C for 1 h in the dark. Cells (including HepG2, 7721, and L02 cells) were incubated with ER–horse (labeled with FITC) for 0, 2, 5, 10, and 30 min. The cells were fixed with 4% paraformaldehyde. The homologous targeting effect of ER–horse was analyzed by CLSM following mounting with antifade polyvinylpyrrolidone mounting medium.

Interaction of STIM1 and ORAI1. ER–horse was labeled with a primary antibody against STIM1 (diluted at 1:100) and a secondary antibody conjugated to DyLight 549 (anti–rabbit IgG antibody, diluted at 1:400) and added to HepG2 cells. After incubation for 30 min, the HepG2 cells were fixed with 4% paraformaldehyde and then labeled with a primary antibody against ORAI1 (diluted at 1:100) and a secondary antibody conjugated with Cy5 (anti–mouse IgG antibody, diluted at 1:400). The nuclei were counterstained with DAPI. CLSM (Olympus FV 1000) was used to observe the fluorescence signals. Pearson’s coefficient was calculated using the JACoP plugin of ImageJ 1.46r software.

Uptake Of Er–horse In Hepg2 Cells

Homologous Fusion of ER–horse with the PM in HepG2 cells. HepG2 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured at 37°C for 24 h. The cells were incubated with Hoechst 33258 (10 µg/mL, blue) and DIL (10 µM, green) for approximately 15 min to stain the cell nucleus and PM. Afterward, the cells were washed twice using PBS and then incubated with ER–horse (labeled with Con A–FITC, final concentration of 1.2 × 10^–2 mg/mL, red) for another 5 min. Finally, the cells were washed, incubated with fresh medium and then observed under a CLSM every 5 min.

Fusion of ERVs and endosomes. HepG2 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured at 37°C for 24 h. The cells were incubated with DAPI (10 µg/mL, blue) for 5 min. Afterward, the cells were washed twice using PBS and then incubated with FITC–labeled dextran (1 mg/mL, green) and ER–horse (red, labeled with DIL, final concentration of 1.2 × 10^–2 mg/mL) for 1 h. Finally, the cells were washed, incubated with fresh medium and then observed under a CLSM.

Transport and release of siGRP94 by Endo–ERVs. HepG2 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured at 37°C for 24 h. Afterward, the cells were washed twice using PBS and incubated with ER–horse (the siGRP94 was labeled with FITC, final concentration of 10 µM) for 0.5–1.5 h. Then, after washing twice with 1× PBS, the cells were fixed with 4% (v/v) paraformaldehyde for 10 min in the dark, permeabilized with 0.1% (v/v) Triton X–100 for 10 min and subsequently blocked using 2% (w/v) bovine serum albumin for 1 h. The cells were incubated with primary antibodies (EEA1 and CNX, diluted 1:200) and secondary antibodies (anti–mouse IgG–DyLight 649, purple; anti–rabbit IgG(H + L)–DyLight 549, red). The nuclei were counterstained with DAPI. CLSM (Olympus FV 1000) was used to observe the fluorescence signals.

Endosomal escape process and mechanism. HepG2 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured at 37°C for 24 h. The cells were incubated with Hoechst 33258 (10 µg/mL) for approximately 15 min to stain the cell nucleus. Afterward, the cells were washed twice using PBS and then incubated with Alexa 647–labeled dextran (1 mg/mL, purple) and ER–horse (the siGRP94 was labeled with FITC, green; the ERVs were labeled with DIL, red; final concentration of 10 µM, quantified by siGRP94) for 1 h at 37°C. Finally, the cells were washed, incubated with fresh medium and then observed by CLSM with a Leica SP8 every 15 min.

Mechanism Of Induced Stress Cascade

Determination of the intracellular Ca ²⁺ concentration and ROS production in the cells. HepG2/L02/7721 cells were incubated with ER–horse (1.2 × 10^–2 mg/mL), siGRP94–RBCVs (10 µM), siGRP94–CCVs (10 µM), ERVs (1.2 × 10^–2 mg/mL), NC–ERVs (10 µM), or siGRP94 (10 µM) for 0–12 h. The levels of [Ca²⁺] i were measured with Fluo–4 AM (excitation wavelength of 488 nm and emission wavelength of 525 nm), a visible–wavelength Ca²⁺ probe. The intracellular generation of ROS was inspected with DHE (excitation wavelength of 435 nm and emission wavelength of 610 nm). DHE is the most commonly used fluorescent probe to detect the levels of superoxide anions in cells. The dye (containing 5 µM DHE and 5 µM Fluo–4 AM) was added to the cells, and the cells were incubated for 1 h at 37°C under 5% CO₂ in the dark according to the protocols. Then, the cells were washed three times with PBS (pH 7.4). The nuclei were counterstained with DAPI. Fluorescence was measured by CLSM and flow cytometric analysis.

Determination of the intracellular Ca ²⁺ concentration and the production of ROS in HepG2 cells by incubation with ER–horse after pretreatment with different drugs. Before incubation with ER–horse, HepG2 cells were pretreated with different drugs that affect intracellular Ca²⁺ homeostasis, including BAPTA (12.5 µM), 2–APB (25 µM), and DIDS (30 µM). After 20 min of pretreatment at 37°C, the HepG2 cells were incubated with ER–horse (1.2 × 10^–2 mg/mL) for 1 h at 37°C. The levels of Ca²⁺ and ROS were measured with Fluo–4 AM and DHE, respectively. Then, the cells were washed three times with PBS (pH 7.4). The nuclei were counterstained with DAPI. Fluorescence was measured by CLSM and flow cytometric analysis.

Protein extraction and WB analysis. Cells were washed three times with cold PBS and lysed on ice using 100 µL of RIPA lysis reagent supplemented with 2 µL of PMSF for 30 min. Cell debris was removed by centrifugation at 14,000 × g for 20 min at 4°C. The protein concentration was measured by BCA assay. The clarified protein lysates from each experimental condition were boiled for 5 min. Protein samples (30 µg) were subjected to 10% (SDS–PAGE and transferred onto polyvinylidene difluoride membranes. After blocking with 5% nonfat milk in 1× TBST containing 0.05% Tween–20 for 2 h at room temperature, the membranes were incubated with specific primary antibodies overnight at 4°C. Anti–mouse IgG and anti–rabbit IgG were used as secondary antibodies. Specific proteins were detected with enhanced chemiluminescence reagent. The antibodies used were as follows: anti–GRP94 (1:1000), anti–GRP78 (1:1000), anti–Bcl–2 (1:1000), anti–Bax (1:1000), anti–Casepase–12 (1:1000), anti–CHOP (1:1000), anti–TRAF2 (1:1000) and anti–Cl–casepase3 (1:1000).

Analysis of cell survival via CLSM. HepG2/L02/7721 cells were seeded into 24–well plates (1.0 × 10⁴ cells per well) and cultured at 37°C for 24 h. Afterward, the cells were washed twice using PBS and incubated with ER–horse (10 µM) for 36 h before being centrifuged at 1,000 × g for 5 min. Then, the cell culture medium was aspirated, and the pellet was washed twice using PBS and incubated with 200 µL of staining solution (containing 10 µL of PI) for 20 min at room temperature. After washing twice with 1× PBS, the apoptosis results were observed by CLSM (excitation at 535 nm and emission at 617 nm)

Cytotoxicity assay. The cytotoxicity of these formulations was evaluated using a CCK–8 assay. HepG2/L02/7721 cells were seeded in 96–well plates at 5,000 cells per well and cultured overnight. The cells were then incubated with different formulations in the different pretreatments (siGRP94, ERVs, ER–horse, siGRP94–RBCVs, siGRP94–CCVs) for 0–36 h. PBS diluted with DMEM was used for the NC group. CCK–8 reagent was added to each well (10 µL per well), and the cells were incubated for another 4 h. After incubation, the absorbance at 490 nm was measured by a microplate reader (BioTek Synergy H1, USA). The cell viability was calculated.

Flow cytometric analysis of apoptosis. Cells incubated with different formulations for 36 h, including siGRP94, NC, ERVs, NC–ERVs, and ER–horse (1.2 × 10^–2 mg/mL), were harvested, washed in cold PBS, centrifuged three times at 1,000 × g for 5 min and then resuspended in PBS. A flow cytometry assay was performed using double staining (PI and annexin V) at the Academy of Life Sciences (Chongqing Medical University, China).

Animal Experiments

In vivo biodistribution. Four–week–old male BALB/c–nude mice purchased from the Laboratory Animal Services Center of Chongqing Medical University were maintained under pathogen–free conditions. All procedures for the mouse experiments were approved by the Ethics Committee of Chongqing Medical University. The mice were subcutaneously inoculated with a HepG2 cell suspension (1.0 × 10⁶ cells in 100 µL of PBS) and maintained until measurable tumor masses (2–3 mm) formed. To investigate the tumor–targeting effect and distribution of ER–horse in tumor tissues, ER–horse (labeled with DIR), siGRP94–RBCVs (labeled with DIR), siGRP94–CCVs (labeled with DIR), and siGRP94–FITC were intratumorally injected into mice via the tail vein at a siGRP94 concentration of 1 mM per mouse. At certain times post–injection, the mice were anesthetized with isoflurane, and fluorescence images of the mice were acquired using an in vivo imaging system (IVIS Spectrum; PerkinElmer, USA). After 48 h, the mice were sacrificed, and the tumors together with other major organs (heart, liver, spleen, lungs, and kidneys) were collected for ex vivo imaging. The fluorescence signals from different organs were analyzed with IVIS software.

Tissue section Ca ²⁺ and ROS staining. To assess the increases in Ca²⁺ and ROS in vivo, the tumor–bearing mice were then randomly divided into 4 groups (the PBS group, siGRP94 group, ERV group, and ER–horse group). The treatment formulations were peritumorally injected. The mice were euthanized after 36 h, and their tissues were dissected. The tissues were snap–frozen in liquid nitrogen, embedded in O.C.T. compound, and cut into frozen sections (10 µm). The sections were stained with Fluo–4 AM and DHE and then imaged by confocal microscopy.

IHC study and histological analysis. Immunostaining was completed on paraffin–embedded tumor tissues (in the PBS group, siGRP94 group, ERV group, and ER–horse group). A GRP94 antibody (1:1000) and a GRP78 antibody (1:1000) were used in IHC with the streptavidin peroxidase–conjugated method. The sections were visualized under a microscope at 400× magnification. The tumors and other major organs (heart, liver, spleen, lungs, and kidneys) were sectioned for H&E staining. Finally, the tumors were stained with TUNEL and imaged by confocal microscopy.

In vivo antitumor efficacy using a hepatocellular carcinoma tumor Model. The antitumor mechanism was studied based on the following experiments. BALB/c mice were injected subcutaneously with 1 × 10⁶ HepG2 tumor cells in the right flank to prepare the tumor model. When the tumor size reached ~ 100 mm³, mice were randomly divided into 4 groups (n = 4) and injected peritumorally with PBS (labeled as Control group), siGRP94 group, ERVs group, and ER–horse group at a siGRP94 dose (0.5 mg/kg) and ERVs dose (1.2 mg/mL) in 150 µL of PBS buffer once every 2 days. All the mice were administered seven consecutive injections, the tumor growth was monitored by measuring perpendicular diameters using a calliper, and tumor volume was calculated as follows: V = L × W²/2, where W and L represent the shortest and longest diameters, respectively. The tumor volumes and mice body weights were monitored every day. Two days after the last injection, the animals were sacrificed to harvest tumors and the image was captured by camera (Nikon D5100, Japan) and weighted. The pathological changes of tumors were analyzed by H&E staining. All animal testing and research were complied with the relevant ethical regulations. All animal procedures were approved by the Research Ethics Committee of Chongqing Medical University (reference number: 2022-k54).

Statistical Analysis. All data are represented as mean ± standard deviation (SD) (n = 3), and statistical differences between two groups were determined using unpaired two–tailed Student’s t–test (*p < 0.05, **p < 0.01, and ***p < 0.001). Statistical significance was set at p < 0.05.

Reporting Summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1. Schroder, M. & Kaufman, R.J. ER stress and the unfolded protein response. Mutat. Res. 569, 29–63 (2005).

2. Shen, X., Zhang, K. & Kaufman, R.J. The unfolded protein response: a stress signaling pathway of the endoplasmic reticulum. J. Chem. Neuroanat. 28, 79–92 (2004).

3. Rao, R.V., Ellerby, H.M. & Bredesen, D.E. Coupling endoplasmic reticulum stress to the cell death program. Cell Death Differ. 11, 372–380 (2004).

4. Breckenridge, D.G. et al. Regulation of apoptosis by endoplasmic reticulum pathways. Oncogene. 22, 8608–8618 (2003).

5. Hetz, C., Chevet, E. & Harding, H. P. Targeting the unfolded protein response in disease. Nat. Rev. Drug Discovery 9, 12, 703719 (2013).

6. Kaufman, R. J. Orchestrating the unfolded protein response in health and disease. J. Clin. Invest. 110, 1389–1398 (2002).

7. Wang, M. & Kaufman, R. J. The impact of the endoplasmic reticulum proteinfolding environment on cancer development. Nat. Rev. Cancer. 14(9), 581597 (2014).

8. Walter, P. & Ron, D. The unfolded protein response: from stress pathway to homeostatic regulation. Science 334, 1081–1086 (2011).

9. Lee, A. H., Iwakoshi, N. N., Anderson, K. C. & Glimcher, L. H. Proteasome inhibitors disrupt the unfolded protein response in myeloma cells. Proc. Natl Acad. Sci. USA 100, 9946–9951 (2003).

10. Schroder, M. & Kaufman, R. J. The mammalian unfolded protein response. Annu. Rev. Biochem. 74, 739–789 (2005).

11. Garg, A. D., Maes, H., van Vliet, A. R. & Agostinis, P. Targeting the hallmarks of cancer with therapyinduced endoplasmic reticulum (ER) stress. Mol. Cell. Oncol. 2(1), 975089 (2015).

12. Ojha, R. & Amaravadi. R. K. Targeting the unfolded protein response in cancer. Pharmacol. Res. 120, 258266 (2017).

13. Healy, S. J. M. et al. Targeting the endoplasmic reticulumstress response as an anticancer strategy. Eur. J. Pharmacol. 234246 (2009).

14. Walter, P. & Ron, D. The unfolded protein response: from stress pathway to homeostatic regulation. Science 334, 10811086 (2011).

15. Hetz, C., Zhang, K. & Kaufman, R. J. Mechanisms, regulation and functions of the unfolded protein response. Nat. Rev. Mol. Cell Biol. 21, 421–438 (2020).

16. Cao, B., Li, J. & Mao, X. Dissecting Bortezomib: Development, Application, Adverse Effects and Future Direction. Curr. Pharm. Des. 19, 31903200 (2013).

17 Peinelt, C. et al. Amplification of CRAC current by STIM1 and CRACM1 (Orai1). Nat. Cell Biol. 8, 771–773 (2006).

18 Waldherr, L., Tiffner, A., Mishra, D., Sallinger, M. & Schindl, R. Blockage of store–operated Ca²⁺ influx by synta66 is mediated by direct inhibition of the Ca²⁺ selective Orai1 pore. Cancers 12, 2876 (2020).

19. Luik, R. B., Wang, B., Prakriya, M. & Wu, M. Oligomerization of STIM1 couples ER calcium depletion to CRAC channel activation. Nature 454, 538–542 (2008).

20. Saheki, Y. & De Camilli, P. Endoplasmic reticulum–plasma membrane contact sites. Annu. Rev. Biochem. 86, 659–684 (2017).

21. Loncke, J. et al. Balancing ER–mitochondrial Ca fluxes in health and disease. Trends Cell Biol. 31, 598–612 (2021).

22. Gómez–Valadés, A. G. et al. Mitochondrial cristae–remodeling protein OPA1 in POMC neurons couples Ca homeostasis with adipose tissue lipolysis. Cell Metab. 33, 1820–1835 (2021).

23. Wang, C. et al. Surfactin–induced apoptosis through ROS–ERS–Ca²⁺–ERK pathways in HepG2 cells. Cell Biochem. Biophys. 67, 1433–1439 (2013).

24. Sari, F. R. et al. Sex differences play a role in cardiac endoplasmic reticulum stress (ERS) and ERS–initiated apoptosis induced by pressure overload and thapsigargin. Cardiovasc. Pathol. 20, 281–290 (2011).

25. Wang, R. et al. Total saponins of aralia elata (Miq) seem alleviate calcium homeostasis imbalance and endoplasmic reticulum stress–related apoptosis induced by myocardial ischemia/reperfusion injury. Cell Physiol. Biochem. 50, 28–40 (2018).

26. Nakano, H. et al. Reactive oxygen species mediate crosstalk between NF–kappaB and JNK. Cell Death Differ. 13, 730–737 (2006).

27. Liu, H., Dai, L., Wang, M., Feng, F. & Xiao, Y. Tunicamycin induces hepatic stellate cell apoptosis through calpain–2/Ca–dependent endoplasmic reticulum stress pathway. Front. Cell Dev. Biol. 9, 684857 (2021).

28. Veit, C., König, J., Altmann, F. & Strasser, R. Processing of the t erminal alpha–1,2–linked mannose residues from oligomannosidic N–glycans is critical for proper root growth. Front. Plant Sci. 9, 1807 (2018).

29. Von Figura, K., Gieselmann, V. & Hasilik, A. Mannose 6–phosphate–specific receptor is a transmembrane protein with a C–terminal extension oriented towards the cytosol. Biochem. J. 225, 543–547 (1985).

30. Guan, C. et al. Glycosylated liposomes loading carbon dots for targeted recognition to HepG2 cells. Talanta 182, 314–323 (2018).

31 Jiang, M. et al. Preparation of CdSe QDs–carbohydrate conjugation and its application for HepG2 Cells labeling. B. Korean Chem. Soc. 33, 571–574 (2012).

32. Gai, X., Tu, K., Lu, Z. & Zheng, X. MRC2 expression correlates with TGFβ1 and survival in hepatocellular carcinoma. Int. J. Mol. Sci. 15, 15011–15025 (2014).

33. Das, R. & Mohapatra, S. Highly luminescent, heteroatom–doped carbon quantum dots for ultrasensitive sensing of glucosamine and targeted imaging of liver cancer cells. J. Mater. Chem. B 5, 2190–2197 (2017).

34. Alfadda, A. A. & Sallam, R. M. Reactive oxygen species in health and disease. J. Biomed. Biotechnol. 2012, 936486 (2012)

35. Luk, B. & Zhang, L. Cell membrane–camouflaged nanoparticles for drug delivery. J. Control. Release 220, 600–607 (2015).

Acknowledgements

The authors acknowledge financial support from the projects of National Science and Technology Major Project of the Ministry of Science and Technology of China (2018ZX10732202), the National Natural Science Foundation of China (82002254, 81873972 and 81873980), the Natural Science Foundation of Chongqing (cstc2021jcyj–msxmX0119), the Chongqing Science Fund for Distinguished Young Scholars (cstc2019jcyjjqX0028), the Foundation for Innovative Research Groups of Chongqing Higher Education Institutions (CXQT20013), and Chongqing Medical University Graduate Talent Training Program (BJRC202023).

Author Contributions

X. T. B, N. K. F and M. L contributed equally to this study. W. C, S. Q. L and F. Z. S designed and supervised the study. X. T. B, N. K. F and M. L performed the major experiments. D. B. H, J.L, and X. Y. L performed the experiments. L. S. K and S. J. D wrote the paper. All authors read and approved the final manuscript.

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Competing Interest

The authors declare no conflict of interest.

Additional information

Supporting Information The online version contains supplementary material available at: http://www.xxx,xxx.

Correspondence and requests for materials should be addressed to Wei Cheng, Siqiao Li or Fangzhou Song.

Peer review information Nature Communications thanks xxx and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Reprints and permission information is available at http://www.nature.com/reprints

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

There is NO Competing Interest.

SupportingInformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

An ER–Horse Detonating Stress Cascade for Hepatocellular Carcinoma Nanotherapy

Status:

Version 1

Abstract

Figures

Introduction

Results

Conclusion

Methods

Homologous–targeting Ability Of Er–horse

Uptake Of Er–horse In Hepg2 Cells

Mechanism Of Induced Stress Cascade

Animal Experiments

Data Availability Statement

References

Declarations

Additional Declarations

Supplementary Files

Status:

Version 1