Antibodies for phosphor-p38 MAPK (4511), PARP (9542), and LC3 (4108) were from Cell Signaling Technology (Boston, Massachusetts, USA). Antibody for HSPA4 (381501), JNK1 (201001), P44/42 MAPK (ERK1/2) (250222), p38 (220979), Hsp90 alpha (209902), GAPDH (200306-7E4), $\beta$-tubulin (200608) were from Zen Bioscience (Chengdu, China). Anti-caspase 3 (ab32351) and c-FLIP${}_{\text{L}}$ (ab8421) were from Abcam (Cambridge, UK). Anti-phospho-JNK (2679369) and phosphor-MAP kinase 1/2 (ERK1/2) were from Millipore (Boston, MA, USA). Licochalcone A (HY-N0372), U0126 (HY-12031A), SB203580 (HY-10256), MKT-077 (HY-15096), z-VAD-fmk (HY-16658B), Chloroquine (CQ) (HY-17589A) and Thiazolyl Blue (HY-15924) were from MCE MedChemExpress (Monmouth Junction, New Jersey, USA).

NCM460 of normal human colonic epithelial and HT29 cell line was from JENNIO Biological Technology (China). Colon cancer cell lines SW620 and HCT116 were purchased from Procell Life Science & Technology Co., Ltd. (China). NCM460 grew in RPIM 1640 mediums. SW620 cells grew in DMEM mediums. HT29 and HCT116 cells grew in MCCOY’S 5A mediums. These mediums were added with fetal bovine serum (FBS) (1:10), L-glutamine (2 mmol/L), penicillin (100 U/mL), and streptomycin (100 µg/mL). All cells were held at 37 °C with 5% CO${}_2$ and were mycoplasma-free (Yeasen Mycoplasma Detection Kit). MCCOY’S 5A mediums without fetal bovine serum were from Solarbio (Beijing, China). RPIM 1640 medium and DMEM medium without FBS were obtained from Gibco (Grand Island, New York, USA). All cell lines were authenticated shortly before use by the STR technique, carried out by manufacturers.

Thiazolyl Blue (MTT) experiment was used to investigate the cytotoxic activities of LA. CRC cells were divided into three parts and plated in 48-well plates (2.5 $\times$ 10${}^4$/well) and overnight, next cultured with MTT (20 mg/mL) at 37 °C. After 3 h, mediums were discarded, and MTT crystals were treated with 100 µL DMSO for 10 min. Next, the optical densities were measured using a fluorescent plate reader (Olympus CKX41, Tokyo, Japan) at OD${}_{570}$. Then, the inhibitory rates of cell growth were computed as follows: (A control - A treated)/A control $\times$ 100%, where A treated and A control is the absorbance of the treated and control groups, respectively. Each experiment was performed in quadruplicate and repeated at least three times.

The proliferation effects of LA on SW620 cells were investigated with a colony formation experiment. CRC cell line was grown in six-well plates (2 $\times$ 10${}^4$ cells/well). Then, they were incubated with diverse thicknesses of LA. After 7 days (change the mediums with LA every 3 days), mediums were discarded, and cells were treated with paraformaldehyde for 15 min. Then, Colony formation was photographed after being treated using 0.1% crystal violet dye for 10 min.

CRC cell line gown in 48-well plates, next, exposed to DMSO (0.1%) or LA for 8 h at 37 °C. Then, discarded the old culture medium from the 48-well plates and added 200 µL medium containing LA and EdU (medium: EdU solution = 1000:1) to each well for 2 h. Cells were treated with 2 mg/mL glycine and 0.5% TritonX-100 after being fixed with 4% paraformaldehyde. Next, DNA syntheses were measured by the EdU Apollo488 Imaging Kit according to recommendations (RiboBio, Guangzhou, China).

Cells were grown in 48-well plates and, next, cultured with CQ, U0126, and LA for 8 h at 37 °C. Then, cells were treated with Methanol precooled at –20 °C and were exposed to Triton X-100 diluted in PBST (1:200) at 20 °C. After 30 min, they were treated with BSA (1:20 diluted in PBST) for 1 h and held at 4 °C for 12 h with an antibody against LC3 (1:200 diluted in PBST). Then, cells were held with an anti-rabbit antibody which was conjugated to Alexa Fluor 594. After 1 h, DAPI was used to stain the nucleus for 5 min, and cells were examined with a fluorescence microscope instrument (Olympus CKX41, Japan) immediately.

CRC cell lines were grown in 6-well plates (6 $\times$ 10${}^6$ cells/well) and, 12 h later, were grown with LA. After 8 h, the collected cells were fixed with ethanol (70%) for 12 h at –20 °C. After that, cells were treated with propidium iodides (PI) for 30 min. Quantitation of the cell cycle ratio was measured by an Accuri C6 flow cytometry instrument (BD Biosciences, Franklin Lakes, New Jersey, USA).

Apoptotic cell death of CRC was measured by Annexin V-FITC Apoptosis Detection Kit (BD Biosciences). CRC cell line grown in 6-well plates. Then, they were grown with DMSO (0.1%) or 20 µM LA. After 48 h, cells were harvested after being treated with 0.25% Trypsin. Next, the binding buffer with PI (5 µL) and Annexin V (FITC-conjugated) (5 µL) was used to treat cells for 15 min in the dark at room temperature. After that, the Accuri C6 flow cytometry instrument (BD Biosciences) was utilized to detect cell apoptosis. The percentages of AnnexinV-FITC/7AAD stained cells were analyzed by Flowjo 10.0 software (Tree Star, San Carlos, California, USA). Experiments were repeated at least three times, and representative results are shown in each figure.

All RNA was separated from CRC cells by Trizol (Ambion, Waltham, Massachusetts, USA). The process of quantification includes two reactions. The first reaction is reverse transcription which used FastKing RT Kit (gDNase) (TIANGEN, Beijing, China) in a reaction system of 20 µL. Reactions were carried out in a C1000 TouchTM Thermal Cycler (BIO-RAD, Hercules, California, USA) at 42 °C for 15 min, then heat inactivation at 95 °C for 3 min. After that, the reverse transcription reaction system was mixed with water without nuclease and held at –80 °C.

The second reaction is Real-time PCR which was carried out using rotorgene Q 2plex HRM (QIAGEN, Hilden, Germany) with SuperReal PreMix Plus (SYBR Green) (TIANGEN) with a 20 µL PCR reaction system. Each sample was divided into three parts to run and analyze. After the PCR cycle reaction, an analysis of the melting curve was carried out to prove the specificity of PCR products. In addition, primer sequences of c-FLIP${}_{\text{L}}$, Hsp70, and Hsp90 were designed and synthesized by Sangon Biotech (Table 1).

Radio immunoprecipitation assay buffer was used to lyse cells. Then, whole cell lysates were centrifugated, and the supernatant fraction was collected for immunoblotting. Protein concentrations of supernatant fraction were investigated by the bicinchoninic acid (BCA) experiment. Then, proteins were incubated at 100 °C for 10 min. After that, equal proteins were extracted from the lytic sample, then distributed by SDS-PAGE. Afterward, protein molecules were metastasized to polyvinylidene difluoride membrane in an electric field. After that, it was exposed to tris buffered saline with Tween-20 (TBST), including non-fat milk powder (1:20). After 2 h; the membrane was exposed to primary antibodies with specificity at 4 °C for 12 h. The next day, it was exposed to secondary antibodies (horseradish peroxidase-conjugated). Immunoreactivities were investigated by the enhanced chemiluminescence (ECL) system (Millipore). Each experiment was carried out 3 times; then representative consequences were shown.

CRC cell lines were grown in 12-well plates (3 $\times$ 10${}^6$ cells/well) and, 12 h later, were incubated with LA or DMSO for 8 h. Next; cells were treated with CHX (200 µg/mL). Radio immunoprecipitation assay buffer was used to lyse cells at 0, 0.5, 1, 2, and 4 h. Then, whole cell lysates were centrifugated, and the supernatant fraction was collected for immunoblotting.

All statistics were shown as mean $\pm$ SD and were assessed using SPSS 25.0 for Windows (SPSS Inc., Armonk, New York, USA). The differences between the two groups were determined using two-tailed t-tests. The differences among three or more groups were determined using one-way analysis of variance (ANOVA). p 0.05 was viewed as statistically significant.

To interrogate the anticancer effects of LA in CRC in vitro, SW620, HCT116, HT29, and normal colonic epithelial cell line NCM460 grown with diverse thicknesses of LA for 48 h, and next cell viabilities were investigated by MTT experiment. Then, LA was proved to suppress the viability of CRC cells (Fig. 1A). The 50% inhibition concentrations (IC50) for 48 h of LA treatment were 26.4 µM, 30.6 µM, 41.5 µM, and 89.4 µM for the SW620, HCT116, HT29, and NCM460 cells respectively, indicating that LA selectively inhibited CRC cells. It seems that KRAS-mutated CRC cells (SW620 and HCT116) were more sensitive to LA compared with KRAS wild HT29 cells. Similarly, SW620 cells were exposed to diverse thicknesses of LA, then the cell viabilities were detected by MTT experiment at 24, 48, 72, and 96 h (Fig. 1B). Then, LA was proved attenuated the viabilities of CRC cells dose- and time-dependently. Consistently, colony formation assay demonstrated that LA inhibited single CRC cells proliferation (Fig. 1C), and EdU staining showed that LA inhibited DNA synthesis of CRC cells (Fig. 1D). Moreover, LA-induced cell viability inhibition was significantly enhanced under starvation stimulation, indicating that LA might effectively inhibit CRC cells under stress conditions (Fig. 1E).

Fig. 1.

The suppression of licochalcone A (LA) on colorectal cancer (CRC) cell proliferation is significantly greater than its toxicity to normal cells. (A) different thickness of LA was used to treat SW620, HCT116, HT29 and NCM460 cells. After 48 h, cell proliferations were detected by Thiazolyl Blue (MTT) experiment. (B) SW620 cells were cultured with diverse thickness LA, and then cell viability was tested by MTT experiment at 24, 48, 72, and 96 h. (C) SW620 cells were cultured with LA (0, 5, and 10 µM) for 7 days; then crystal violet staining was carried out. (D) SW620 cells were cultured with 20 µM LA for 8 h; next, EdU staining was performed. Scale bar 100 µm. (E) SW620 and HCT116 were cultured with LA (0, 10, 20, 30, 40, and 50 µM) for 48 h in cultures with serum or not, and then cell viabilities were measured by MTT experiment. (F) SW620 and HCT116 cells were cultured with LA (20 µM); next, flow cytometric assay was utilized to examine the cell cycle at 8 h. N = 3. *p 0.05, **p 0.01, ***p 0.001.

To further investigate the effects of LA in CRC, we detected the cell cycle under LA-induced anticancer activity of SW620 and HCT116 cells. Flow cytometry analysis discovered that significant increase (from 12.1% to 26.5% in SW620 cells and from 18.9% to 33.1% in HCT116 cells) in the ratio of G2 phase in cells after being treated with LA, indicating that the CRC cells cycle was arrest in G2/M phase by LA (Fig. 1F). In summary, LA selectively suppressed CRC cells by inducing arrest in G2/M cell cycle.

Next, we investigated the relationship of apoptosis in LA-triggered cell death by Annexin V/PI experiment. Flow cytometric results indicated that LA enhanced apoptosis of CRC cells (Fig. 2A). Meanwhile, LA greatly augmented the level of hallmark proteins of apoptosis (cleaved caspase-3 and PARP) (Fig. 2B). Furthermore, pan-caspase inhibitor (z-VAD-fmk) greatly compromised LA-triggered cytotoxicity (Fig. 2C), further supporting that LA induced cell apoptosis of CRC.

Fig. 2.

LA triggered CRC cell apoptosis. (A) SW620 and HCT116 cell lines were cultured in 20 µM LA or DMSO, then flow cytometric assay was utilized to detect apoptotic cells at 48 h. (B) SW620 cell line grown with diverse thicknesses of LA. Next, the level of proteins was detected using immunoblotting at 48 h. (C) SW620 cell line was pretreated with 20 µM z-VAD-fmk. After 1 h, cells were cultured with 20 µM LA. Next, cell viabilities were measured using an MTT experiment at 48 h. (D) SW620 cells were grown with diverse thicknesses of LA; next, the level of c-FLIP${}_{\text{L}}$ was measured by immunoblotting at 48 h. (E) SW620 cells were cultured with 20 µM LA. After that, all RNA was isolated, and c-FLIP${}_{\text{L}}$ mRNA was detected by q-RT-PCR at 8 h. (F) SW620 cells were cultured with 20 µM LA or DMSO and, after 8 h, cultured with 200 µg/mL CHX for different times. After that, protein content was tested using immunoblotting. N = 3. *p 0.05, **p 0.01, #p $>$ 0.05.

Recently, there was growing evidence that c-FLIP${}_{\text{L}}$ could be used as a target to repair the apoptotic reaction of cancer cells [24]. Therefore, we examined the level of c-FLIP${}_{\text{L}}$ in CRC cells. Then, LA was discovered inhibited c-FLIP${}_{\text{L}}$ expression (Fig. 2D) rather than its mRNA level (Fig. 2E) in SW620 cells. Furthermore, the cycloheximide chasing assay suggested that LA did not accelerate the degradation of c-FLIP${}_{\text{L}}$ (Fig. 2F). This indicated that the inhibition of c-FLIP${}_{\text{L}}$ by LA was independent of transcription and degradation and might be related to the translation of its mRNA. Collectively, LA induced cell apoptosis of CRC.

MAPKs are associated with LA-triggered anticancer effects. However, the roles of MAPKs are elusive in CRC. Thereby, we explored mechanisms of MAPKs in LA-induced anticancer activity. Then, LA was found indeed to activate ERK and p38 time-dependently in SW620 and HCT116 cells. Whereas the impact of LA on JNK activity was not obvious in SW620 and HCT116 cells (Fig. 3A). After that, we interrogated the role of ERK and p38 in LA-triggered anticancer activity. The results discovered that ERK inhibitor U0126, not p38 inhibitor SB203580, remarkably promoted LA-induced inhibition of SW620 cell viability (Fig. 3B), indicating that ERK activation resisted LA-triggered anticancer activities in SW620 cells, not HCT116 cells. Therefore, we explored the role of ERK in the LA-induced apoptosis of SW620 cells. The results showed that U0126 enhanced cell apoptosis (Fig. 3C), augmented cl-caspase-3 and cl-PARP expression, and further promoted the downregulation of c-FLIP${}_{\text{L}}$ in LA-induced anti-CRC activity (Fig. 3D). This indicated that the activation of ERK resisted LA-induced cytotoxicity in SW620 cells.

Fig. 3.

ERK-mediated autophagic induction resisted LA-triggered anti-CRC activities. (A) SW620 and HCT116 cells were cultured with 20 µM LA; next, the level of proteins was measured by immunoblotting at different times. (B) SW620 and HCT116 cell line was pretreated with 20 µM U0126 and 20 µM SB203580. After 1 h, it was cultured with 20 µM LA for 48 h. After that, the viability of CRC cells was tested by MTT experiment. (C and D) SW620 cell line was cultured with 20 µM U0126. After 1 h, it was cultured with 20 µM LA; the apoptotic cell death was tested using flow cytometry (C), and the level of proteins was measured by immunoblotting (D) at 48 h. $\beta$-tubulin was measured as an input control. N = 3. **p 0.01, ***p 0.001.

Previous studies have demonstrated LA induced autophagy in diverse types of cancers [13, 14, 15, 17, 19]. Whereas mechanisms about autophagy in LA-triggered anticancer activity were controversial. Therefore, we investigated mechanisms of autophagic induction in LA-triggered anti-CRC effects. Then, LA was discovered to induce LC3-II accumulation (Fig. 4A), indicating LA might increase autophagic activities in SW620 cells. These results were further supported by the autophagic flux assay (Fig. 4B). Consistently, immunofluorescence staining also discovered that CQ blocked LA-induced autophagic flux (Fig. 4C). To uncover the role of autophagic induction in LA-triggered anticancer effects, CQ was used to block the activity of autophagy. As shown in Fig. 4D and 4E, CQ significantly augmented LA-triggered apoptosis of CRC cells and enhanced the level of cl-caspase-3 and cl-PARP, meanwhile downregulated the level of c-FLIP${}_{\text{L}}$ (Fig. 4F). Which discovered that autophagic induction played a protective role by maintained c-FLIP${}_{\text{L}}$ expression in LA-triggered cytotoxicity.

Fig. 4.

Autophagy played a protective part for CRC cells in LA-triggered anticancer activities. (A) SW620 cells were cultured with 20 µM LA at different times; next, LC3 content was detected by immunoblotting. (B) SW620 cell line grown with 20 µM chloroquine (CQ). After 1 h, cells were cultured with 20 µM LA, then the level of LC3 was detected by immunoblotting at 8 h. (C) SW620 cell line grown with 20 µM CQ. After 1 h, cultured with 20 µM LA, then the level of LC3 was measured by immunofluorescence experiment at 8 h. Scale bar 100 µm. (D, E, and F) SW620 cell line grown with 20 µM CQ. After 1 h, cultured with 20 µM LA, then the viability of CRC cells was tested by MTT experiment (D), the apoptotic cell death was determined with flow cytometric assay (E), and the level of protein was determined by immunoblotting (F) at 48 h. (G and H) SW620 cell line was cultured with 10 µM U0126. After 1 h, it was cultured with 20 µM LA. After that, the level of LC3 was detected by immunoblotting (G) and immunofluorescence experiment (H) at 8 h. Scale bar 100 µm. $\beta$-tubulin and GAPDH were measured for input control. N = 3. **p 0.01, ***p 0.001.

Thereby, we explored the relationship between ERK and autophagy in LA-triggered anticancer effects. The Western blot (Fig. 4G) and immunofluorescence staining (Fig. 4H) results showed that U0126 suppressed LA-induced LC3-Ⅱ accumulation in SW620 cells, which indicated that LA triggered autophagic induction by ERK activation. Therefore, autophagy activation mediated by ERK played a protective part in LA-triggered anti-CRC effects.

HSPs have been reported to inhibit apoptosis by assisting in the synthesis of apoptosis-related proteins [23]. Therefore, we detected the expression of Hsp70, Hsp90, HSPA4 (Hsp70 Family Member), and Hsp90$\alpha$ (Hsp90 Family Member) in CRC. Surprisingly, our results indicated that LA greatly enhanced the expression of Hsp70 rather than its mRNA level in SW620 cells (Fig. 5A and 5B). However, LA only marginally affects the expression of Hsp90, HSPA4, and Hsp90$\alpha$. Furthermore, the cycloheximide chasing assay suggested that LA only marginally affected the degradation rate of Hsp70 and Hsp90 (Fig. 5C). This indicated that the enhancement of Hsp70 by LA was independent of transcription and degradation and could be involved in the mRNA translation.

Fig. 5.

Hsp70 expression maintained c-FLIP${}_{\text{𝐋}}$ expression and resisted LA-triggered apoptosis. (A) SW620 cell line was cultured with different thicknesses of LA. After 48 h, the level of protein was measured using immunoblotting. (B) SW620 cell line was cultured with 20 µM LA. After 8 h, all RNA in cells was isolated, and mRNA of Hsp70 and Hsp90 were detected by q-RT-PCR. (C) SW620 cell line grown with 20 µM LA or DMSO. After 8 h, it was cultured with 200 µg/mL CHX; next, the proteins were measured using immunoblotting at diverse times. (D) SW620 cell line grown with 20 µM CQ. After 1 h, it was cultured with 20 µM LA. After that, the level of Hsp70 was measured by immunoblotting at 8 h. (E, F, and G) SW620 cell line was cultured with 6 µM MKT-077 for 1 h, next, cultured with 20 µM LA for 48 h. After that, the viability of CRC cells was tested by MTT experiment (E), the apoptotic cell death was tested using flow cytometric analysis (F), and proteins were measured by immunoblotting (G). $\beta$-tubulin and GAPDH were measured for input control. N = 3. *p 0.05, ***p 0.001, #p $>$ 0.05.

Next, we found that CQ significantly inhibited Hsp70 expression (Fig. 5D), indicating that LA enhanced Hsp70 expression through autophagy activation. Then, the results of the MTT experiment and flow cytometry analysis indicated that MKT-077 (Hsp70 inhibitor) enhanced LA-triggered apoptosis (Fig. 5E and 5F). Moreover, MKT-077 enhanced cl-caspase-3 and cl-PARP and inhibited c-FLIP${}_{\text{L}}$ (Fig. 5G). This indicated that Hsp70 resisted LA-triggered anticancer effects. Overall, autophagy activation mediated by ERK played a protective part in LA-triggered anti-CRC effects by enhancing the level of Hsp70 and sustaining the expression of c-FLIP${}_{\text{L}}$in vitro.