FBXL18 activating AKT/CCND1 signaling pathway mediates radioresistance in esophageal squamous cell carcinoma

Abstract

F–box and leucine-rich repeat protein 18 (FBXL18) is closely associated with cancer progression. However, its role in regulating the radioresistance of esophageal squamous cell carcinoma (ESCC) remains unclear. Radioresistant ESCC cells were developed using fractional doses of X–ray irradiation, and validated via cell counting kit–8 (CCK−8) assay. The sensitivity of these radioresistant cells to radiotherapy was also assessed using CCK−8. The expression levels of FBXL18 and Cyclin D1 (CCND1) were analyzed through Western blotting. RNA interference (RNAi) technology was employed to investigate whether silencing FBXL18 could reduce ESCC radioresistance and inhibit the AKT/CCND1 signaling pathway. Co-immunoprecipitation and Western blotting were used to evaluate the polyubiquitination of AKT. Radioresistant ESCC cells were successfully established, and FBXL18 expression was significantly elevated in these cells. Increased levels of phosphorylated AKT (p–AKT) and CCND1 were also observed. Silencing FBXL18 notably reduced the radioresistance of ESCC cells and decreased p–AKT and CCND1 expression levels. Also, FBXL18 was found to interact with AKT, promoting its K63–linked polyubiquitination, and activating the AKT/CCND1 signaling pathway. FBXL18 interacts with AKT and facilitates its K63–linked polyubiquitination, thereby activating AKT/CCND1 signaling while maintaining the radioresistance of ESCC cells.

INTRODUCTION

The mortality rate of esophageal cancer worldwide is ranked seventh (1). Currently, esophageal squamous cell carcinoma (ESCC) accounts for the predominant histological type in esophageal cancer. Although many attempts have been made to develop new therapeutic methods for ESCC, the treatment outcomes remain unsatisfactory. Importantly, the majority of ESCC patients are diagnosed at an advanced stage. For advanced ESCC, the principal treatment approach is chemotherapy or chemoradiotherapy, followed by radical surgery (2). Radioresistance often develops with the accumulation of X–ray doses, leading to failed radiotherapy and cancer metastasis (3). Therefore, understanding the molecular mechanisms of radioresistance and identifying effective targets to overcome radioresistance are crucial.

Numerous studies have revealed the complex underlying mechanism of radioresistance, highlighting the critical role of cancer stem-like cells. Moreover, it was found that the activation of the PI3K/AKT/BAD signaling pathway enhances DNA damage response, in consequence inhibiting irradiation-induced ESCC apoptosis (4), indicating that enhanced DNA damage repair capacity could confer radioresistance of ESCC. Epithelial–mesenchymal transition (EMT) is also closely linked to radioresistance. There is strong evidence that the activation of EMT and PI3K/AKT signaling unequivocally reduces the radiosensitivity of ESCC (5). Significantly, the tumor microenvironment was also correlated to the radioresistance of ESCC, as evidenced by Fiori et al. (6). Furthermore, it was found that irradiation-induced DNA damage response can lead to the release of cytokines and chemokines (7), triggering inflammatory reactions, and ultimately resulting in radioresistance. Recent studies have also demonstrated that the ubiquitination modifications, monoubiquitination and polyubiquitination, are crucial in the DNA damage response (8-10), indicating that polyubiquitination is significantly associated with radioresistance.

The ubiquitin system includes ubiquitin–activating enzyme E1, ubiquitin–binding enzyme E2, and ubiquitin–protein ligase E3. SCF (SKP1/Cullin1/F–box) E3 ligases are a class of multi-subunit RING E3 ligases, where F–box protein is the crucial substrate recognition subunit, and functions in tumorigenesis and progression (11-13). The F–box with a leucine-rich amino acid repeat sequence (FBXL) is a member of the F–box protein family, which contains F–box and leucine-rich repeat domains. FBXL was proven to function in the progression of hepatocellular carcinoma (14). FBXL18, a member of the FBXL subgroup, was demonstrated to be a mediator of proteasomal degradation of FBXL7, with the function of regulating apoptosis in HeLa cells (15). Furthermore, it has been reported that FBXL18 promotes glioma progression by helping ubiquitinate AKT (16). Therefore, our study aims to validate the role of FBXL18 in the radioresistance of ESCC, and clarify whether FBXL18 regulates polyubiquitination to activate AKT in radioresistant ESCC cells.

RESULTS

FBXL18 expression in radioresistant cells of ESCC

To evaluate whether FBXL18 was involved in developing ESCC radioresistance, radioresistant ESCC cell lines, including KYSE150−R and EC109−R, were established. Subsequently, the resistance ability was verified using CCK−8 assay. The result showed that the cell viability of KYSE150−R cells was higher than that of KYSE150−S cells when exposed to the same irradiation dose (Fig. 1A). Similarly, the cell viability of EC109−R was also higher than EC109−S when exposed to the same irradiation dose (Fig. 1B). These findings indicate that radioresistant KYSE150−R and EC109−R cells were successfully constructed. The protein level of FBXL18 in radioresistant cells was then analyzed using western blot, and the result showed that FBXL18 was significantly up-regulated in both KYSE150−R and EC109−R cells, compared with the parental cells (Fig. 1C, D). These findings suggest that FBXL18 may be involved in the radioresistance of ESCC.

Fig. 1.

FBXL18 is increased in radioresistant cells of ESCC. (A) Radiosensitive KYSE150 cells (KYSE150−S) and radioresistant KYSE150 cells (KYSE150−R) were irradiated with the indicated doses of X–ray (IR). Forty-eight hours later, cells were subjected to CCK−8 assay to evaluate cell viability. (B) Radiosensitive EC109 cells (EC109−S) and radioresistant EC109 cells (EC109−R) were irradiated with indicated doses of X–ray (IR). Forty-eight hours later, cells were subjected to CCK−8 assay to evaluate cell viability. (C, D) KYSE150−S, KYSE150−R, EC109−S, and EC109−R cells were lysed for western blot to assess the protein levels of FBXL18 (C), and the relative protein level of FBXL18 was also measured by optical density analyses (D). GAPDH was used as a loading control. **P < 0.01, ****P < 0.0001.

Knock-down of FBXL18 enhances the sensitivity to irradiation

To further verify the involvement of FBXL18 in radioresistance, FBXL18 was knocked down in KYSE150−R and EC109−R cells using siRNAs targeting FBXL18. The results showed that FBXL18 was significantly knocked down (Fig. 2A, B). CCK−8 assay showed that the cell viability of KYSE150−R and EC109−R cells was significantly reduced after FBXL18 knock down, compared with cells transfected with control siRNA when exposed to irradiation (Fig. 2C, D). These findings further prove that FBXL18 is involved in developing radioresistance in ESCC cells.

Fig. 2.

Knockdown of FBXL18 increases radiosensitivity in radioresistant cells of ESCC. (A, B) KYSE150−R and EC109−R cells were transfected with siNC or siFBXL18 for 48 h, followed by western blot to evaluate the protein levels of FBXL18 (A), and optical density analyses were also measured (B). GAPDH was used as a loading control. (C, D) KYSE150−R (C) or EC109−R (D) cells transfected with siNC or siFBXL18 were irradiated with the indicated doses of X–ray (IR). Forty-eight hours later, cells were subjected to CCK−8 assay to evaluate cell viability. **P < 0.01, ***P < 0.001, ****P < 0.0001.

AKT activation in radioresistant cells

Afterwards, the critical radioresistance-associated protein AKT and its phosphorylation form were detected in radioresistant cells. The result showed that the AKT level was not increased, while the p–AKT level was highly elevated in radioresistant cells (Fig. 3A). The quantification of p–AKT also demonstrated that p–AKT was significantly increased in radioresistant cells (Fig. 3B). Simultaneously, the downstream of AKT, CCND1, was also significantly up-regulated (Fig. 3A, C). These results further suggest that AKT/CCND1 signaling may also be involved in the development of radioresistance in ESCC cells.

Fig. 3.

AKT/CCND1 signaling is activated in radioresistant cells of ESCC. (A) KYSE150−S, KYSE150−R, EC109−S, and EC109−R cells were lysed for western blot to assess the protein levels of p–AKT, AKT, CCND1, and GAPDH. (B, C) The relative protein levels of p–AKT (B) and CCND1 (C) were measured by optical density analyses. *P < 0.05, **P < 0.01.

AKT activation by FBXL18

To further determine whether FBXL18 activated AKT signaling through direct interaction, the co-immunoprecipitation assay was conducted. It was found that FBXL18 could pull down AKT, and also, AKT could pull down FBXL18 in ESCC cells, indicating that FBXL18 could interact with AKT (Fig. 4A). Subsequently, we wondered whether FBXL18 could activate AKT by promoting its ubiquitination modification. The co-immunoprecipitation experiment demonstrated that the overexpression of FBXL18 could promote the ubiquitination modification of AKT (Fig. 4B). Furthermore, it was found that FBXL18 could induce the K63–linked polyubiquitination of AKT, thereby activating AKT (Fig. 4C). These findings proved that FBXL18 could activate AKT signaling by promoting the K63–linked polyubiquitination of AKT.

Fig. 4.

FBXL18 modulates AKT activation by regulating its K63-induced ubiquitination in radioresistant cells of ESCC. (A) KYSE150 cells were lysed for reciprocal co-immunoprecipitations with an anti-AKT or anti-FBXL18 antibody. (B, C) KYSE150 cells were transfected with FBXL18–overexpressing plasmids (FBXL18−OE), along with wild-type ubiquitin plasmids (HA–Ub–WT) (B), or mutated ubiquitin plasmids containing K63 sites (HA–Ub–K63) (C). Forty-eight hours later, cells were lysed for co-immunoprecipitations with an anti-AKT antibody. IP, immunoprecipitation; WCL, whole cell lysate.

Knockdown of FBXL18 downregulates AKT/CCND1 signaling

To further demonstrate that FBXL18 could regulate the AKT signaling pathway, FBXL18 was knocked down in radioresistant ESCC cells to observe the variation of AKT signaling. The western blot result showed that silencing FBXL18 could significantly decrease the level of p–AKT in radioresistant ESCC cells (Supplementary Fig. 1A, B of the Supplementary Information [SI]). Simultaneously, the level of CCND1 protein was also significantly reduced in radioresistant ESCC cells (Supplementary Fig. 1A, C of the SI). Moreover, overexpression of FBXL18 enhanced the radioresistance of ESCC cells; however, after cells were treated with AKT inhibitor, this effect could be significantly reduced (Supplementary Fig. 1D of the SI). These findings further suggest that FBXL18 may mediate the development of radioresistance of ESCC cells by regulating AKT/CCND1 signaling.

DISCUSSION

The application of adjuvant chemotherapy and radiotherapy has improved the prognosis of ESCC patients, particularly in those with a high potential for lymph node metastasis. However, because of the insensitivity of esophageal cancer to radiotherapy, the 5-year survival rate of ESCC patients ranges (0 to 10)% under conventional doses of radiation alone. Therefore, illustration of the mechanism of radioresistance in ESCC is significant.

Ubiquitination is a highly conserved biological process in eukaryotes, involved in regulating multiple cellular functions (17, 18). Ubiquitination is now widely known to be a crucial mechanism in many biological processes, including the cell cycle, apoptosis, autophagy, and epigenetics, as well as NF–κB and T–cell receptor signaling, and even DNA repair and transcription (19-21). The function of ubiquitination in cancers has attracted more attention over the last two decades, because it could effectively inhibit proteasome in clinical cancer treatment. Overexpression of protein members or their mutant proteins in the ubiquitin system has been found in numerous cancer types. It was reported that the activation of ubiquitin is closely associated with malignant progression, and poor survival and prognosis (22, 23). Accumulating evidence suggests ubiquitination plays a role in oncogenic pathways through the monoubiquitylation and polyubiquitination of various substrates, independent of proteasome activity (24). Deficiencies in the process of deubiquitination can significantly impact NF–κB signaling, AKT signaling, and BRCA signaling (25). For decades, it has been proven that most compounds can suppress tumor progression by targeting the ubiquitin process. For example, Bortezomib, approved by the Food and Drug Administration, was applied to treat relapsed myeloma, and in consequence, was applied to hematological malignancies (26, 27).

F–box proteins, substrate-recognition subunits of the SCF E3 ligase complex, function in multiple cellular processes through ubiquitination modification and the degradation of target proteins (28, 29). Recently, the F–box family has been demonstrated to be involved in tumorigenesis (30). FBXL18 has been proven to play a critical role in apoptosis in cervical cancer (15). Moreover, the elevated FBXL18 could promote RPS15A ubiquitination and SMAD3 activation, and consequently drive HCC (14). Although numerous studies have demonstrated the critical role of FBXL18 in cancer progression, its potential function in resistance remains unclear.

Therefore, to assess whether FBXL18 is involved in the development of ESCC radioresistance, the radioresistant ESCC cell lines were established. Our analysis showed that FBXL18 was significantly up-regulated in radioresistant cancer cells. These results suggest that FBXL18 may be involved in the radioresistance of ESCC. To further verify that FBXL18 can regulate radioresistance, the FBXL18 in radioresistant cells was knocked down, and the subsequent results showed that knocking down FBXL18 reduced the radioresistance ability. These findings further prove that FBXL18 is involved in the development of radiation tolerance in ESCC cells.

Previous studies have shown that FBXL18 can regulate the progression of glioma by promoting the ubiquitination modification of AKT (16). Therefore, it was necessary to study whether FBXL18 can activate the AKT signaling pathway. Our result show that p–AKT was highly increased in radioresistant cells, and its downstream CCND1 was significantly up-regulated, which was consistent with the previous study (31). These results further suggest that AKT/CCND1 signaling may also be involved in the radioresistance in ESCC cells. Our further co-immunoprecipitation showed that FBXL18 could bind to AKT in ESCC cells. Moreover, FBXL18 can enhance the K63–linked polyubiquitination of AKT. Similarly, silencing FBXL18 significantly inhibited AKT activation and CCND1 protein expression. These findings further suggest that FBXL18 may be involved in mediating the development of radioresistance in ESCC cells by regulating AKT/CCND1 signaling.

Although this is the first time that FBXL18 has been used to activate AKT signaling by K63-linked polyubiquitination in radioresistant ESCC, some issues remain to be addressed. It was found that the FBXL18 was overexpressed in radioresistant ESCC cells, while silencing FBXL18 also attenuated the resistance ability of ESCC cells. However, it is still unknown whether the overexpression of FBXL18 could promote the radioresistance of ESCC. Therefore, the overexpression of FBXL18 should be performed in radiosensitive ESCC cells to study whether FBXL18 could convert the radiosensitive cells into radioresistant cells. Moreover, the radioresistant ESCC cells highly expressed FBXL18; however, the expression of FBXL18 in tumor tissues from the ESCC patients receiving radiotherapy was unclear. Therefore, future research should conduct IHC staining for FBXL18 in ESCC tissues after radiotherapy. The validation of FBXL18 in clinical cancer tissues could identify FBXL18 as a potential biomarker for radioresistance in ESCC. Last, the tumor microenvironment—including immune components, stromal interactions, and vascularization—plays a critical role in modulating responses to radiotherapy in vivo. While our study primarily focuses on in vitro models, which allow for precise mechanistic dissection, we acknowledge that due to the complexity of the in vivo environment, in vitro radioresistance does not always directly translate into in vivo outcomes. in vivo studies are necessary to validate the functional relevance of FBXL18 in the context of the tumor microenvironment. Future work will involve the establishment of ESCC xenograft models with FBXL18 modulation (e.g., knockdown or overexpression), to evaluate tumor response to irradiation in vivo. This will help confirm the functional role of FBXL18, while also assessing its potential as a therapeutic target in a more physiologically relevant context.

The expression of FBXL18 was significantly elevated in radioresistant ESCC cells. Knocking down FBXL18 effectively resensitized these cells to radiotherapy. Mechanistic studies revealed that the AKT/CCND1 signaling pathway was activated in radioresistant ESCC cells. FBXL18 facilitated the activation of AKT signaling by promoting the K63–linked polyubiquitination of AKT. Silencing FBXL18 markedly inhibited the AKT/CCND1 signaling pathway in radioresistant ESCC cells.

MATERIALS AND METHODS

Cell culture and chemicals

The KYSE150 cells were cultured using a 1:1 ratio of RPMI−1640 and F−12, with 2% fetal bovine serum (FBS, Gibco) and 1% Penicillin–Streptomycin Solution (P/S). The EC109 cells were cultured using DMEM with 10% FBS and 1% P/S (Gibco). All cells were maintained in an incubator at 37°C. When the cell confluence reached 80%, they were digested using trypsin for passage in a 1:3 ratio. AKT Inhibitor VIII was purchased from Selleck Chemicals (Shanghai, China).

Establishment of radioresistant cell lines

The cell line was seeded into a 25 cm² flask, and irradiated with X–ray when the confluence reached 60%. The irradiation dose was 10 Gy from a 6−MV X–ray linear accelerator at a 3 Gy/min rate. Simultaneously, a 1 cm thick tissue-equivalent bolus was placed on the plate for homogeneity. The irradiated cells were placed into the incubator until the cell confluence reached about 60%. Afterward, the cells were irradiated with 10 Gy of X–ray again, and allowed to recover in the cell incubator until reaching 60% confluence. The fractional irradiations were performed until the total dose reached 80 Gy. The radioresistant cell subline was then identified using the cell-counting kit 8 (CCK−8) assay.

CCK−8 assay

The cells were digested using trypsin, and seeded in a 96–well plate. After overnight incubation, the cells were transfected with indicated siRNAs. After 48 h, CCK−8 reagent (MCE, Shanghai, China) was added to fresh medium in a 1:10 ratio. The fresh medium containing CCK−8 reagent was added to a 96–well plate, and incubated at 37°C for 3 h. The absorbance of the reacted production in the 96–well plate was detected using a microplate reader at 450 nm (Thermo Scientific).

Western blot

Total proteins were extracted using Radio Immunoprecipitation Assay lysis buffer (Yeason, Shanghai, China) containing protease inhibitor cocktail (Merck Millipore, Darmstadt, Germany). Protein concentration was determined using a BCA kit (Yeason), followed by denaturation at 100°C for 10 min. Thirty microgram proteins were loaded and separated by (8 or 10)% SDS polyacrylamide gels, then transferred into a polyvinylidene fluoride (PVDF) membrane (Millipore, Billerica, MA, USA). In consequence, the PVDF membrane was blocked using 5% nonfat dry milk in tris buffered saline–tween 20 (TBST) for one hour at room temperature (RT). Afterwards, PVDF membranes were incubated with primary antibodies against FBXL18 (1:1,000) (H00080028−M01, Abnova, Walnut, CA), AKT (1:1,000) (4691s, CST, MA, USA), p–AKT (1:1,000) (4060, CST), HA (1:5,000) (3724, CST), and Cyclin D1 (1:1,000) (#2922, CST) overnight at 4°C. Incubation with mouse antibodies against GAPDH (1:5,000) (sc–32233, Santa Cruz, Calif., USA) was conducted for internal control. After washing three times using 1× TBST, membranes were again incubated with HRP–conjugated secondary antibodies (anti-mouse and anti-rabbit) for 1 h at RT. Finally, the membrane was incubated with ECL buffer (Thermo, USA) for 5 min, followed by imaging using Bio-Rad image lab (Bio-Rad, CA, USA). The relative quantification of the protein band was determined using Image J.

Plasmids and siRNAs

The HA−Ubiquitin (HA−Ub) plasmid was constructed according to the work reported by Dr. Mizuho Hasegawa (32). The siRNAs for FBXL18 were synthesized by Sangon Biotech with the sequence (5’-CUGCAGCACAUGAAAUUCAdTdT-3’). The siRNAs were transfected into parental and resistant cells using Lipofectamine Regent (13778030, Thermo Fisher, Waltham, MA), according to the manufacturer’s instructions.

Immunoprecipitation and ubiquitination assay

The cells transfected with plasmids were lysed using the lysis buffer (Yeason, Shanghai, China). The whole cell lysate was incubated with protein A/G beads (Thermo Fisher) for pre-clearing. Afterwards, the pre-cleared supernatant was incubated with anti-FLAG beads (Sigma, MO) at 4°C overnight. The FLAG beads were collected after centrifugation at 5,000 rpm for 1 min, and washed three times using the lysis buffer. Subsequently, the immune-bound complex was eluted using 3× FLAG peptide incubation at 4°C for 30 min, and subjected to Western blot assay.

Statistics

All statistical data were analyzed using a two-tailed t-test in GraphPad software. The statistical data are represented using the mean ± standard deviation (SD). The symbols *, **, and *** indicate P-values of less than (0.05, 0.01, and 0.001), respectively.