2.1 Cell lines and reagents

CRC cell lines, including LoVo, HT29, and HCT-116, were sourced from the American Type Culture Collection. To establish stable 5-FU resistant CRC cells, HT-29 cell line were treated with gradually increasing concentration of 5-FU (from 10 to 300 μM initially), cells were passaged three times at each 5-FU concentration of 10, 20, 40, 80, 150, 300 μM. We assessed 5-FU resistance at each dose by calculating the IC50 using CCK8 assays. These lines were propagated in DMEM medium from KeyGen Biotechn, Nanjing, China, which was enriched with 10% fetal bovine serum and 1% penicillin–streptomycin, both acquired from Gibco, Eggenstein, Germany. Meanwhile, α-hederin was procured from Chengdu Herbpurify CO., LTD (purity>98%), the chemical structure of α-hederin is shown in Figure S1. α-Hederin was initially dissolved in DMSO before being further diluted using the aforementioned DMEM medium.

2.2 Cell viability assay

For CRC cell viability analysis, 5 × 10³ cells were dispensed into each compartment of a 96-well format. Subsequently, the cells were exposed to varying doses of α-hed ranging from 0 to 56 μM. In another setting, a consistent concentration of 24 μM α-hed was administered for distinct time intervals: 0, 6, 12, 24, 48, and 72 h. Post-treatment, the CCK-8 assay from Beyotime, Nantong, China, was utilized to determine cell viability. Absorbance readings were procured at a wavelength of 450 nm using instrumentation from Tecan Group Ltd., Männedorf, Switzerland.

2.3 Transmission electron microscopy

Exposure of HT-29 cells to DMSO served as the vehicle control, whereas another set experienced 16 μM α-hed intervention for a duration of 24 h. Subsequent procedures involved fixing these cells in 2.5% glutaraldehyde, maintained at 4°C over a 12-h period. This was followed by post-fixation using 1% OsO₄ for an additional hour, staining processes with uranyl acetate, and a dehydration step. The final embedding was done using the Poly/Bed 812 resin, sourced from Pelco. Detailed cellular observations were carried out with an electron microscope, specifically the FEI Tecnai G2 from Thermo Scientific, USA.

2.4 Western blotting

Cell lysates underwent preparation, and their protein concentrations were quantified using the BCA method, a product of Thermo Scientific, USA. A 12% SDS-PAGE allowed the separation of these proteins, followed by their transfer to membranes made of poly-vinylidene difluoride. Once transferred, these membranes underwent a blocking phase for a span of 2 h using 10% bovine serum albumin in TBST that was freshly mixed. Post-blocking, the membranes were exposed to designated primary antibodies for a period of 12 h. For visualization purposes, membranes were treated with HRP-conjugated secondary antibodies sourced from Beyotime, Nantong, China, alongside the ECL substrate from Thermo Scientific, USA. The antibodies used for the experiment included: Alix (CST, 2171S, 1:1000 dilution), ERK (CST, 137F5), p-ERK (CST, 4370S, 1:1000), JNK (proteintech, 66210-1-Ig, 1:3000), p-JNK (proteintech, 80024-1-RR, 1:2000), p38 (CST, 8690S, 1:1000), p-p38 (CST, 4511S, 1:1000), and GAPDH (CST, 5174S, 1:1000).

2.5 Reverse transcription and quantitative real-time PCR

For the synthesis of cDNA, total RNA underwent reverse transcription utilizing the PrimeScript RT Reagent Kit, a product from Vazyme, Nanjing, China. The quantification process for Alix, PLCβ3, IP3R, and PKC was carried out with their expression levels adjusted based on GAPDH. The primers specified in subsequent sections were employed for the RT-qPCR execution (Table 1).

2.6 Swiss target prediction assay

Swiss Target Prediction tool (http://www.swisstargetprediction.ch) analyzed the potential molecule targets of α-hed.

2.7 Molecular docking of α-hed to the GPCRs structural model

Molecular docking methods were used to evaluate binding abilities of α-hed with GPCRs. The 2D structure of α-hed was downloaded from PubChem database, and then transformed into 3D structure with mimimum structural energy by ChemBio3D program. 3D structures of the GPCRs were retrieved from PDB database, and dehydrated by PyMOL 2.5.2. Then, AutoDockTools 1.5.6 performed molecular docking of α-hed with GPCRs. PyMOL 2.5.2 was used to visualize the binding modes of α-hed and GPCRs.

2.8 CRC xenografts

The Institutional and Local Committee on the Care and Use of Animals at Nanjing University of Chinese Medicine granted permission for all described experiments (Ethical approval number, 202110A025). Nude male mice, 5 weeks old and weighing between 18 and 22 g, were procured from Vital River Laboratories, Beijing, China. These mice (N = 30) were distributed randomly across five distinct groups. Each mouse received a subcutaneous injection on their right flank with HT-29 cells, where the injection volume was 100 μL PBS containing 1 × 10⁶ cells. Following a period of 14 days post-injection, when tumors approached an approximate volume of 100 mm³, a regimen of treatments commenced. The mice underwent intraperitoneal injections either with 5-FU at a dose of 25 mg/kg or α-hed at dosages of 1.5, 1, or 0.5 mg/kg, administered biweekly. This treatment spanned 21 days. On Days 7, 14, 21, 28, and 35, tumor dimensions, including length and width, were recorded. The volume was then computed using the formula V = 0.5lw². Concluding the experiments on the 35th day, the mice underwent euthanasia, post which tumor tissues were extracted and subsequently weighed.

2.9 Tissue staining

Tumor samples underwent fixation in a 10% formalin solution prior to being embedded in paraffin. Subsequent sectioning yielded slices with a thickness of 3 μm. For the purpose of observing histopathological changes, staining was conducted using hematoxylin and eosin (H&E). The immunohistochemical (IHC) methodology implemented included the use of primary antibodies targeting Alix and ki-67, both at a dilution ratio of 1:200. Post incubation with the respective secondary antibody, visualization of positive staining was achieved through the application of diaminobenzidine (DAB). Moreover, the TUNEL assay, relying on a colorimetric approach, was executed in accordance with guidelines provided by Beyotime.

2.10 Statistical analysis

For the purposes of data analysis, the GraphPad Prism version 8 platform was employed. Utilizing one-way ANOVA, variations of significance were assessed. A threshold of p < 0.05 served as the criterion for deeming differences to hold statistical significance.

3.1 α-Hed inhibits cell viability and induces cytoplasmic vacuolation in CRC cells

In assessing the potential anti-tumor capabilities of α-hed against CRC cells, three distinct CRC cell strains (LOVO, HT-29, and HCT-116) and one normal human colon epithelial cell (NCM-460) underwent treatment with varying α-hed concentrations for a 24-h period, followed by viability testing using the CCK-8 method. A marked inhibition in cell proliferation was observed across all strains, manifesting in a dose-responsive manner. The corresponding IC50 values were approximately 48 μM for LOVO, 23 μM for HT-29, and 19 μM for HCT-116, 40 μM for NCM-460, as indicated in Figure 1A. When exposed to 24 μM of α-hed, a time-sensitive suppression of proliferation was apparent in each cell strain. The viability was reduced by half at around 14 h for HCT-116, 24 h for HT-29, 46 h for LOVO, and 48 h for NCM-460 (Figure 1B).

Interestingly, cytoplasmic vacuolation was observed in LOVO, HT-29, and HCT-116 cells upon α-hed exposure (Figure 1C). Examination via electron microscopy highlighted deviations in the morphology of both ER and mitochondria. Specifically, cells subjected to α-hed demonstrated pronounced ER swelling, coupled with subtle mitochondrial enlargement (Figure 1D). Previous literature has underscored the role of Alix downregulation in augmenting paraptosis,^{11, 16} suggesting that Alix functions as a potent indicator for paraptosis. In line with this, both protein and mRNA concentrations of Alix in HT-29 cells treated with α-hed were seen to reduce in a dose-sensitive manner (Figure 1E–G).

However, α-hed may also induce the caspase-dependent apoptosis in CRC cells. We have conducted western blot analysis of cleaved caspase-3 in HT-29 cells at 24, 48 and 72 h after treatment with α-hed (Figure S2A,B). The results showed that α-hed activated the cleaved caspase3 expression in CRC cells. To determine if apoptosis is the decisive form of α-hed induced cell death, we inhibited apoptosis by a caspase inhibitor, Z-VAD fluoromethyl ketone (Z-VAD-FMK).¹⁷ Previous study has shown that 5-FU induces caspase-dependent apoptosis,¹⁸ and it could be blocked by Z-VAD-FMK,¹⁹ so we used 5-FU as a positive control to induces apoptosis. Our study showed that both α-hed and 5-FU increased the expression of cleaved caspase3, however, Z-VAD-FMK can't rescue the elevated cleaved caspase3 expression that induced by α-hed. In contrast, Z-VAD-FMK rescued the elevated cleaved caspase3 expression that induced by 5-FU (Figure S2C,D). These results indicated that apoptosis is not the only main form of cell death following α-hed exposure, it would support the induction of paraptosis may exert decisive influence to CRC cell death.

3.2 α-Hed targets GPCRs to activate PLCβ3/IP3R/PKCα signaling pathway

Swiss Target Prediciton tool predicted 100 proteins with known ligands that are highly similar to α-hed, 60% of which are Family A GPCRs (Figure 2A), such as α2-adrenoceptor subtypes (α2A, α2B, and α2C) (ADRA2A, ADRA2B, ADRA2C), dopamine D1 receptors (DRD1), and so forth. We then analyzed molecular simulation docking of α-hed and GPCRs complex. For instance, α-hed forms six hydrogen bonds with ADRA2B. Hydrogen bond is a intermolecular force that kept molecules together, the amino acids TYR-40, LYS-280, PHE-234, SER-191, LEU-190, PHE-151 in α-hed forms hydrogen bonds, the molecular docking yields a binding energy of −10.2 kcal/mol (Figure 2B). Besides, other GPCRs, such as ADRA2A, ADRA2C, DRD1, and so forth, also exhibit molecular docking binding target (Figure S3). These findings are certainly suggested a direct binding between α-hed and GPCRs.

GPCRs are links of extracellular signals to the control of cell fate, such as survival, proliferation and death.²⁰ Some downstream regulators of GPCRs, such as PLCβ3, IP3R, PKCα were able to ignite Ca²⁺ signaling.²¹ CRC Cells were treated with 12 μM α-hed for the indicated time points, at 6 h of α-hed treatment, the protein and mRNA expression levels of PLCβ3 were markedly increased. Coincidentally, IP3R started to elevate at 6 h. At 20 h of α-hed treatment, expression of both IP3R and PKCα were significantly increased (Figure 2C–E). CCG258747, a GPCR kinases (GRKs) subfamily–selective inhibitor, was used to impair efficacy on GPCRs-dependent process in cells.²² We subsequently explored if CCG258747 could block the activation of PLCβ3/IP3/ Ca²⁺/PKCα pathway by α-hed. CCG258747, given alone, did not significantly modify the expression of PLCβ3, IP3R, PKCα, but upregulated Alix. It significantly diminished the promoting effects of α-hed on both the proteins and mRNA expression of PLCβ3, IP3R, PKCα while counteracted the inhibitory effect of α-hed on Alix (Figure 2F–H), confirming the presumption that α-hed targets GPCRs to activate PLCβ3/IP3R/PKCα signaling pathway.

3.3 Elevated intracellular Ca²⁺ levels are critical for α-hed-induced paraptosis

Cytosolic Ca²⁺ signals mediate diverse cellular responses, it generated through the coordinated translocation of Ca²⁺ across the ER membrane. ER and mitochondria are the main reservoirs of Ca²⁺.²³ We found both ER and mitochondria swelled after α-hed exposure, dysregulation of ion homeostasis may result in abnormal osmotic pressure, which in turn lead to morphological changes of organelles, so we presumed that α-hed may disrupt the concentration of cytosolic Ca²⁺ level.

Utilizing flow cytometry and the cell-permeable Ca²⁺-indicator dye, Fluo-3, it was observed that when HT-29 cells underwent treatment with α-hed, there was a pronounced enhancement in the levels of intracellular Ca²⁺. This augmentation was noted to initiate around the 10-h mark, escalating sharply by 14 h, and culminating in a peak at the 24-h interval post the administration of α-hed (Figure 3A,B). Interestingly, CCG258747 significantly diminished the promoting effects of α-hed on Fluo-3 fluorescence intensity (Figure 3C), confirming the presumption that α-hed targets GPCRs to elevate intracellular Ca²⁺ levels. We then observed the localization of Ca²⁺ to the ER, mitochondria. As shown in Figure S4, the untreated HT-29 cells exhibited evenly distributed Ca²⁺, and no obvious localization commonality with ER or mitochondria was observed. After treatment with α-hed, the Fluo-3 fluorescence intensity increased significantly, and aggregated in clusters, which adjacent to ER-tracker red, coinciding with mitochondrial-tracker-red. These results indicate that α-hed elevated intracellular Ca²⁺ levels, and the release of Ca²⁺ from ER is subsequently uptaken by mitochondria.

Subsequently, the research focus shifted to identifying potential evidence that suggests the role of Ca²⁺ in facilitating the paraptosis triggered by α-hed. We tested the effect of the SEA0400, a specific Na⁺/Ca²⁺ exchange inhibitor,²⁴ on α-hed induced cytotoxicity. As shown in Figure 3D, SEA0400 effectively prevented vacuolization formation in α-hed-treated cells. Besides, it also blocked the inhibitory effect of α-hed treatment in HT-29 cells viability, while counteracted the promoting effect of α-hed on cell death. (Figure 3E,F). Via Western blot analysis, when HT-29 cells underwent concurrent treatment with SEA0400, the diminished Alix protein and mRNA levels caused by α-hed were reverted to normal (Figure 3G–I).

The data gathered implies the significant elevation in intracellular Ca²⁺ levels is related to GPCRs signaling. Moreover, obstructing the Ca²⁺ pathway attenuates the paraptotic effects instigated by α-hed in CRC cells.

3.4 α-Hed activates MAPK cascade by Ca²⁺ signaling

Cytosolic Ca²⁺ have been shown to trigger MAPK signaling, which has implications for paraptosis and vacuolation-related cell demise.²⁵ This study probed the phosphorylation status of JNK, ERK, and p38 to evaluate MAPK activation dynamics. Results from Western blotting revealed that while α-hed exposure to HT-29 cells had minimal effect on JNK, ERK, and p38 expression, there was a notable augmentation in the phosphorylation of MAPK proteins, demonstrating a dose-responsive relationship (Figure 4A,B). Moreover, concurrent treatment with SEA0400 in HT-29 cells thwarted the α-hed-triggered phosphorylation states of JNK, ERK, and p38 (Figure 4C,D). Such findings substantiate the crucial role of the MAPK signaling cascade in the Ca²⁺ driven paraptosis in CRC cells.

3.5 Inhibition of chemo-resistant CRC cell line by α-hed may be effective compared to chemotherapeutic agents

Most chemotherapeutic agents, such as 5-FU, exert inhibitory activity in cancer by inducing apoptosis, thus non-apoptotic cell death may suggest strategies that could offer alternatives to chemo-resistant CRC therapeutic approaches. We then determined whether α-hed, which induces paraptosis, is able to target resistant CRC cancer cells. To test this hypothesis, we set up 5-FU-resistant CRC clones by a step-wise dose escalation study as reported previously.²⁶ Cell viability in the presence of 5-FU clearly showed that derived HT-29 clones are resistant to 5-FU, the IC50 of 5-FU-resistant and parental HT-29 clones are being at approximately 200, 75 μM respectively (Figure 5A). Remarkably, reduction of viability by exposure of α-hed in 5-FU-resistant and parental HT-29 cells showed barely difference (Figure 5B). α-hed also induced cytoplasmic vacuolation in 5-FU resistant HT-29 cells, indicating initiation of paraptosis-like features (Figure 5C). Besides, α-hed also decreased expression of Alix in 5-FU-resistant HT-29 cells in a does-dependent manner (Figure 5D). Collectively, α-hed may effectively inhibit apoptosis-based 5-FU-resistant CRC cells via inducing paraptosis-like cell death.

3.6 α-Hed inhibits 5-FU-R HT-29 xenograft growth in mice

In order to validate whether α-hed possesses inhibitory effects on chemo-resistant CRC in vivo, a study was conducted in nude mice with chemo-resistant HT-29 tumor xenografts to assess the therapeutic potential.

These mice underwent treatment via intraperitoneal injection of either 5-FU (25 mg/kg) or varying dosages of α-hed (0.5, 1, 1.5 mg/kg). Observations indicated that 5-FU administration barely led to a suppression in CRC progression relative to the control. Furthermore, intermediate α-hed doses (0.5, 1 mg/kg) exhibited marginal tumor growth deceleration. The dose of 1.5 mg/kg of α-hed marked a significant decline in tumor volumes and weights (Figure 6A–C), which seemed to surpass the inhibitory capability of 5-FU. Analysis of excised tumor samples further elucidated α-hed's impact on CRC expansion. Histological evaluation via H&E staining displayed cytoplasmic vacuolation, mirroring attributes of paraptosis-resembling cellular fatality (Figure 6D). Yet, it remains imperative to highlight observed fluctuations in body weight measurements across both 5-FU and α-hed administered groups (Figure S5).

Through IHC staining, it became evident that administering α-hed led to a notable suppression in cellular proliferation, as illustrated by the diminished expression of Ki67. Moreover, signs pointing towards paraptosis as the decreased expression of Alix (Figure 6E–H). In contrast, the PLCβ3/IP3R/PKCα pathway was increased (Figure S6). Additionally, cell death was witnessed (Figure S7). All these observations collectively hint that α-hed induces paraptosis and contributes to an observable restraint on chemo-resistant CRC growth when assessed in vivo.