FHL2 activates β-catenin/Wnt signaling by complexing with APC and TRIM63 in lung adenocarcinoma

Highlights

• •Highly expressed FHL2 is correlated with poor prognosis in LUAD patients.
• •FHL2 facilitates the progression of LUAD in vitro and in vivo.
• •FHL2 acts as a scaffold for APC and TRIM63 in LUAD.
• •FHL2 enhances TRIM63 mediated ubiquitination of APC and stabilizes β-catenin/Wnt signaling pathway.

Abstract

Objectives

Four and a half LIM domain 2 protein (FHL2) was reported to regulate the progression of various cancers and this study aimed to clarify the intrinsic mechanism of FHL2 facilitating the progression of lung adenocarcinoma.

Methods

In this study, bioinformatic analysis and immunohistochemistry staining were used to confirm the FHL2 levels in patients with lung adenocarcinoma. The potential influence of FHL2 on the biological function of lung adenocarcinoma cells was verified in vitro and in vivo. To uncover the potential mechanism contributing to the advance of lung adenocarcinoma, liquid chromatography‒mass spectrometry and immunoprecipitation assays were performed to detect the partners of FHL2.

Results

FHL2 levels were upregulated in lung adenocarcinoma and contributed to a dismal prognosis. Moreover, in vitro and in vivo assays suggested that genetic inhibition of FHL2 undermined the viability, migration and invasion of lung adenocarcinoma cells, while forced expression of FHL2 showed the opposite trend. Mechanistically, liquid chromatography‒mass spectrometry and coimmunoprecipitation assays revealed that FHL2 could function as a scaffold to enhance TRIM63-mediated ubiquitination of APC. The degradation of APC further stabilized β-catenin and activated Wnt signaling pathway.

Conclusion

Collectively, this study uncovered the underlying mechanism by which FHL2 regulates the biological characteristics of tumors and provided a novel target for lung adenocarcinoma treatment.

Introduction

While lung cancer has the second highest morbidity, it still has the highest mortality and ranks as the leading cause for cancer-related death in both males and females, with a 5-year overall survival of 16 % [1]. It can be mainly classified into two subgroups: non-small cell lung cancer, accounting for approximately 85 % of all cases, and small cell lung cancer, accounting for the remainder. In non-small cell lung cancer, lung adenocarcinoma (LUAD) is the most common subtype, accounting for approximately 40 % of all cases [2]. While radical resection with minimally invasive surgery has shown promising efficacy in early LUAD, the prognosis of patients with advanced LUAD is still dismal [3]. According to the latest guideline, platinum-based chemotherapy has still been recommended as the front-line therapy for LUAD and the efficacy is limited by intrinsic or acquired drug resistance, leading to the treatment failure and poor prognosis. Thus, further study on the oncogenesis and metastasis of LUAD is critical for developing new treatment strategies.

Four and a half LIM domain 2 protein (FHL2) is a multifunctional scaffolding protein and has been found to interact with multiple proteins because of its LIM motifs, including IER3 and MDM2 [4]. The zinc-finger motif contained in the LIM domain facilitated FHL2 transcriptional repression or coactivation, which modulated various cellular activities, including cell proliferation and motility. LIM domain was the platform for multimeric protein complex formation, which enabled FHL2 to regulate transcriptional expression of other genes. Meanwhile the expression of FHL2 was regulated by other different transcription factors including P53, SRF and Sp1 [5]. Moreover, accumulated studies have confirmed that FHL2 regulates the biological function of cancers in a context-dependent manner. It promotes oncogenesis and metastasis in ovarian, cervical and gastrointestinal cancers but inhibits tumor growth in hepatocellular carcinoma [[6], [7], [8]]. Thus, the different roles of FHL2 in tumors might be attributed to the different molecules interacting with it. While previous studies have confirmed the oncogenic role of FHL2 in lung cancer, its role in chemotherapy resistance need further investigation. Thus, this study mainly aimed to uncover the intrinsic mechanism by which FHL2 contributes to the progression of LUAD and further explored its role in chemotherapy.

In this study, we found that FHL2 was highly expressed in tumor tissues and was associated with poor prognosis in LUAD patients. The in vitro and in vivo assays confirmed the oncogenic role of FHL2 in LUAD. Mechanistically, the highly expressed FHL2 could act as the scaffold for APC and TRIM63, which enhanced the ubiquitination of APC, stabilized the downstream β-catenin and activated the Wnt signaling pathway. Thus, this study confirmed that FHL2 was an oncogenic regulator and facilitated the progression of LUAD via the destabilization of APC and activation of the Wnt signaling pathway.

Methods

Cell culture and transfection

The LUAD cell Lines A549, H1299, H1395, H1975 and Calu-3 were provided by the cell bank of the Chinese Academy of Sciences. The culture conditions were DMEM or RPMI-1640 (Gibco, USA) with 10 % fetal bovine serum (Gibco, USA) at 37 °C and 5 % CO2. Lentiviruses silencing FHL2 and TRIM63 and overexpressing FHL2 were constructed by Genomeditech (Shanghai, China). Transfection was performed according to the manufacturer's protocol. The transfection efficiency was verified by western blot or quantitative real-time polymerase chain reaction (qRT‒PCR). The targets of shFHL2 and shTRIM63 are described in Table S2.

Tissue samples, tissue microarray (TMA) and bioinformatic analysis

A TMA including 128 LUAD tumor tissues and matched para-tumor tissues was obtained from patients who underwent radical resection of malignant lesions between 2008 and 2010. The tissues were frozen in liquid nitrogen and stored at −80 °C. The histopathology of all the tumor tissues was verified by two pathologists. The clinical characteristics of these patients are recorded in Table S1. The consent forms of all patients were obtained, and the human and animal ethnics were waived by the Ethics Committee of Fudan University Zhongshan Hospital. The bioinformatic analysis of FHL2 was performed in GEPIA 2.0 (http://gepia2.cancer-pku.cn/). The classification of FHL2^low and FHL2^high is based on the cutoff of median expression of FHL2.

RNA extraction and qRT‒PCR

qRT‒PCR was performed according to our previous study [9]. Briefly, total RNA was extracted from the LUAD cell lines using TRIzol (Yeasen, China). cDNA was produced with the Hifair III reverse transcriptase cDNA synthesis kit (Yeasen, China). qRT‒PCR was performed with Hieff SYBR Green Master Mix (Yeasen, China) in an applied biosystem (Thermo Fisher, USA). The parameters were set according to the manufacturer's protocol. The forward and reverse primers are provided in Table S2. The expression of GAPDH was regarded as the internal control.

Western blotting

The total proteins of LUAD cells were extracted via lysis buffer for the WB/IP assay (Yeasen, China) with a proteinase inhibitor cocktail (Yeasen, China). The concentrations were measured by the BCA protein quantification kit (Yeasen, China). SDS‒PAGE was used to separate the total proteins, and then the proteins were transferred to PVDF membranes (Millipore, USA). Next, the membranes were incubated with primary antibodies for 12 h at 4 °C and further incubated with HRP-conjugated secondary antibodies for 1 h. Finally, the target protein was detected with ECL detection reagent in a Tanon imaging system (Shanghai, China). The antibodies used in this study are presented in Table S3.

Colony formation assays

To evaluate the proliferation ability, approximately 1000 cells of each group were added into a 6-well plate and cultured with DMEM and 10 % FBS for 2 weeks. Then, the cells were fixed with 4 % paraformaldehyde for 10 min and stained with 0.1 % crystal violet for 10–20 min. A colony containing more than 50 cells was regarded as positive.

CCK-8 assay

For the cell proliferation assay, 1000 cells per well were seeded in a 96-well plate. Ten hours post-seeding, the supernatants were replaced with fresh culture medium containing CCK8, in accordance with the manufacturer's guidelines. OD value measurements at 450 nm were initially taken after 2 h and subsequently recorded daily over a period of four days.

Migration and invasion assays

For the migration assay, approximately 5 × 10⁴ cells were seeded in the upper chamber of a 24-well transwell plate with an 8 μm pore size (Corning, USA). The upper chamber contained 200 μl serum-free culture medium, and 500 μl culture medium containing 10 % FBS was added to the lower chamber. After 30 h of incubation at 37 °C with 5 % CO2, the upper chamber was fixed with 4 % paraformaldehyde for 10 min and stained with 0.1 % crystal violet for 10–20 min. For the invasion assay, the upper chamber was precoated with 60 μl Matrigel matrix for 30 min in the incubator. Then, approximately 5 × 10⁴ cells were seeded in the upper chamber of an 8 μm-pore 24-well transwell plate with 200 μl serum-free culture, and 500 μl medium with 10 % FBS was added to the lower chamber. The chambers were incubated for 72 h, and the membranes were fixed with 4 % paraformaldehyde for 10 min and stained with 0.1 % crystal violet for 10–20 min. The images were photographed with a microscope (Olympus, Japan).

Immunofluorescence (IF)

To determine the localization of FHL2 and APC, A549 cells were fixed with 4 % formaldehyde. Then, the cells were penetrated with 0.1 % Triton-X100 and blocked with 5 % bovine serum albumin for 1 h. Next, the cells were incubated with primary antibodies for 12 h at 4 °C. After that, the cells were incubated with Alexa Fluor 488 or Alexa Fluor 594 for 2 h at room temperature. The nuclei were stained with DAPI for 30 min. Finally, the slides were sealed with antifade mounting medium. The slides were scanned by a fluorescence microscope (Leica, Germany).

Immunohistochemistry (IHC)

IHC was performed according to our previous study [10]. After dewaxing and rehydration, the tissues were blocked with 3 % H2O2 for 30 min and subjected to antigen retrieval with citric acid at 100 °C for 20 min. Then, the tissues were incubated with bovine serum albumin and stained with the Streptavidin Peroxidase IHC assay kit (ZSGB-Bio, China). The primary antibodies used for IHC are listed in Table S3. Image-Pro Plus 6.0 software was used for image analysis. The average gray value was used to quantify the expression level.

Coimmunoprecipitation (Co-IP) and protein mass spectrometry

A total of 2 × 10⁷ A549 cells were lysed with Wb/IP lysis buffer (Yeasen, Shanghai) containing a protease and protein phosphatase inhibitor cocktail (Yeasen, Shanghai). Then, 20 μl of protein A/G magnetic beads (MedChemExpress, USA) was added to the lysates and incubated for 2 h. Next, the beads were removed, and 10 μl of FHL2 or IgG antibodies were incubated with the lysates for 12 h at 4 °C. Then, 50 μl of A/G magnetic beads were added and incubated with the lysates for 2–4 h. Finally, the beads were separated by the magnetic separator and washed three times with PBS. The immunoprecipitated protein was eluted from the beads. The proteins were detected by the Easy-nLC1200 chromatographic system (Thermo Fisher, USA) and Q-exactive mass spectrometer (Thermo Fisher, USA).

Subcutaneous xenograft model

The in vivo xenograft assay was performed with 4-week-old BALB/c nude mice raised by GemPharmatech (Nanjing, China). Mice were fed in a pathogen-free environment. To construct the subcutaneous xenograft model, a total of 5 × 10⁶ cells were subcutaneously injected into the right flank of nude mice (n = 5 for each group). The size of the tumors was measured every four days and calculated as ½ length × width [2]. After 4 weeks, the mice were sacrificed, and the tumors were harvested. The in vivo experiments were waived by the Animal Committee of Zhongshan Hospital Fudan University.

Statistical analysis

The significant difference was analyzed via SPSS 23.0 (IBM, USA) and GraphPad Prism 8.0. Differences between two groups were analyzed by the chi-squared test or Student's t-test. Spearman correlation analysis was performed to verify the correlation coefficients for the expression of FHL2, APC, β-catenin and TRIM63. To analyze the differences between three groups or more, one-way ANOVA was applied. Univariate and multivariate Cox regression analyses were performed to identify the independent factors associated with poor prognosis in LUAD patients. Kaplan‒Meier and log-rank tests were performed to analyze the survival difference between the FHL2^high and FHL2^low groups. A P value <0.05 was determined to be statistically significant. *, P < 0.05; **, P < 0.001; ***, P < 0.0001.

Results

FHL2 was highly expressed in tumor tissues and indicated poor prognosis in LUAD patients

The expression of FHL2 and its role in the prognosis of LUAD patients were initially explored in the GEPIA 2.0 database. The bioinformatic analysis showed that FHL2 was obviously highly expressed in the tumor tissues of LUAD (Fig. 1A). In the FHL2^high group, patients had lower disease-free survival (P = 0.0038) and overall survival (P = 0.0038) (Fig. 1B). To further verify the bioinformatic results, IHC staining of FHL2 was performed in a TMA containing the tumor tissues and matched para-tumor tissues of 128 LUAD patients. Consistently, the results suggested the high expression of FHL2 in the tumor tissues compared with the matched para-tumor tissues (Fig. 1C). Moreover, Kaplan‒Meier and log-rank analyses showed that patients with high FHL2 expression had a higher incidence of postoperative recurrence (P = 0.017) and shorter overall survival (P = 0.009) (Fig. 1D). In addition, the chi-square and Student's t tests indicated that patients in the FHL2^high group were prone to have larger tumor sizes, higher clinical stages and more lymph node metastases (Table S1). The univariate and multivariate Cox regression analyses identified the expression of FHL2 as an independent risk factor for recurrence and overall survival in 128 LUAD patients (Fig. 1E). Thus, the results of bioinformatic analysis and verification in tumor tissues suggested that FHL2 might contribute to the progression of LUAD.

Fig. 1.

FHL2 was highly expressed in tumor tissues and indicated poor prognosis in LUAD patients

A. Bioinformatics analysis of FHL2 expression in tumor (n = 483) and normal tissues (n = 347) via the GEPIA2.0 database.

B. Prognosis analysis was performed in the GEPIA2.0 database based on the expression of FHL2 (classified as FHL2^low and FHL2^high groups).

C. Representative images of FHL2 IHC staining in a TMA containing 128 paired LUAD tumor tissues and paratumor tissues. The mean optical density calculated by ImageJ was regarded as the score for each dot. Paired Student's t-test was applied to determine significant differences.

D. The 128 patients were classified into two groups according to the expression of FHL2 (FHL2^low and FHL2^high). The prognosis discrepancy was compared by Kaplan‒Meier analysis and log-rank test.

E. Univariate and multivariate Cox regression analyses were performed to identify the independent factors predicting poor prognosis in 128 LUAD patients.

FHL2 facilitated the progression of LUAD cells in vitro and in vivo

To further explore the impact of FHL2 on LUAD, the FHL2 level in five LUAD cell lines (A549, H1395, H1975, Calu-3 and H1299) was detected via western blotting and qRT‒PCR, which showed that FHL2 was most highly expressed in A549 cells and least expressed in H1299 cells (Fig. 2A-B). Therefore, we constructed two stably transfected cell lines with FHL2 knockdown in A549 cells (shFHL2#1 and shFHL2#3) and FHL2 overexpression in H1299 cells (H1299-FHL2) and confirmed the efficiency with western blotting and qRT‒PCR (Fig. 2C-E). In colony formation assays, knockdown of FHL2 impaired the viability of A549 cells, while FHL2 overexpression in H1299 cells rescued it (Fig. 2F). Matrigel Transwell assays showed that knockdown of FHL2 impeded the migration and invasion of A549 cells, while overexpression of FHL2 showed the opposite trend (Fig. 2G). In line with the in vitro results, the in vivo assays in BALB-c nude mice showed that knockdown of FHL2 slowed the growth of subcutaneous tumors (Fig. 2H). Taken together, these in vitro and in vivo assays suggested that FHL2 could impose a potentiating influence on the progression of LUAD.

Fig. 2.

FHL2 facilitated the migration, invasion and viability of LUAD cells in vitro and in vivo

A. FHL2 expression in 5 LUAD cell lines (A549, H1395, H1975, Calu-3 and H1299) was detected by western blotting.

B. The mRNA level of FHL2 in 5 LUAD cell lines was measured by qRT‒PCR.

C. The efficiency of FHL2 silencing or overexpression was verified by western blotting.

D. The mRNA level of FHL2 in A549-shFHL2#1 and A549-shFHL2#3 cells was confirmed by qRT‒PCR.

E. The mRNA level of FHL2 in H1299-FHL2 cells was measured by qRT‒PCR.

F. After silencing FHL2 in A549 cells and overexpressing FHL2 in H1299 cells, the viabilities of the two cell lines were detected by colony formation assays.

G. After silencing FHL2 in A549 cells and overexpressing FHL2 in H1299 cells, the migration and invasion abilities were confirmed by Matrigel Transwell assay. Scale bar = 50 μm.

H. In vivo assay in nude mice containing three groups (NC, A549-shFHL2#1 and A549-shFHL2#2, N = 5 for each group). The tumor size was measured every three days.

E used Student's t-test to analyze the differences. D and F-H were compared using one-way ANOVA.

FHL2 enhanced the ubiquitination of APC in LUAD cells

Previous studies reported FHL2 as a multifunctional scaffolding protein that could interact with multiple proteins [4]. Thus, we investigated the partners interacting with FHL2 in LUAD cells. Immunoprecipitation was performed in A549 cells, and mass spectrometry was used to detect potential interacting proteins. Then, the detected proteins were overlapped with the potential candidates predicted in the HitPredict database, and a total of 8 proteins were found (Table S4). Among the 8 candidates, FHL2 was the most enriched protein, followed by APC, which is a tumor suppressor and acts as an antagonist of the Wnt signaling pathway via the degradation of β-catenin (Fig. 3A). Therefore, we suspected that APC was the main partner interacting with FHL2 in LUAD cells. The co-IP assays with FHL2 antibody in A549 cells identified that FHL2 could interact with APC (Fig. 3B). Meanwhile, the co-IP assays with APC antibody in A549 cells also observed the interaction of APC with FHL2 (Fig. 3C). Moreover, we found that the knockdown of FHL2 in A549 cells enhanced the expression of APC, while the overexpression of FHL2 in H1299 cells decreased the level of APC, which was reversed by the administration of the protease inhibitor MG132, as shown by western blotting (Fig. 3D-E). The knockdown of FHL2 also impeded the ubiquitination level of APC in A549 cells (Fig. 3F). In addition, we found that the regulation of FHL2 had no significant impact on the mRNA level of APC (Fig. 3G). Thus, FHL2 mainly destabilized APC at the posttranscriptional level. The IF staining of APC and FHL2 showed that both were mainly present in the cytoplasm of A549 cells (Fig. 3H). IHC staining of FHL2, APC and β-catenin in the TMA containing 128 LUAD tissues confirmed the negative correlation between FHL2 and APC and the positive association between FHL2 and β-catenin (Fig. 4A).

Fig. 3.

FHL2 enhanced the ubiquitination of APC in LUAD cells

A. Overlap of the proteins detected by IP/LC‒MS and the predicted candidates in the HitPredict database.

B. Immunoprecipitation with FHL2 antibody and western blotting were performed in A549 cells to verify the interaction between FHL2 and APC.

C. Immunoprecipitation with APC antibody and western blotting were performed in A549 cells to verify the interaction between APC and FHL2.

D. After silencing FHL2 in A549 cells, the level of APC was detected via western blotting.

E. After the overexpression of FHL2 and the administration of MG132 in H1299 cells, the level of APC was explored via western blotting.

F. FHL2 was silenced in A549 cells, and immunoprecipitation was performed with IgG and APC antibodies. The ubiquitination level of APC was verified by western blotting.

G. FHL2 was knocked down or overexpressed in A549/H1299 cells, and the mRNA level of APC was measured by qRT‒PCR. One-way ANOVA was applied.

H. IF detection of the localization of FHL2 and APC in A549 cells. Scale bar = 100 μm.

Fig. 4.

Verification of FHL2, APC and β-catenin expression in LUAD tissues

A. IHC staining of FHL2, APC and β-catenin in a TMA containing 128 LUAD tissues. Spearman correlation analysis was performed to detect the association within the three molecules.

FHL2 acted as a scaffold for TRIM63 in LUAD cells

Since FHL2 facilitated the ubiquitination of APC at the posttranscriptional level, we suspected that FHL2 might also act as a scaffold for ubiquitination-related partners. Among the 8 candidates identified by the overlap of mass spectrometry data and the targets predicted in the Hitpredict database, TRIM63 is an E3 ligase (Fig. 3A) that has been reported to function as an oncogene in various cancers [11,12]. Therefore, it was hypothesized that FHL2 mediated the destabilization of APC via interaction with TRIM63. The Co-IP assay with FHL2 antibody in A549 cells confirmed the interaction between FHL2 and TRIM63. In addition, another Co-IP assay with TRIM63 antibody verified the interaction of TRIM63 and FHL2 (Fig. 5A). The knockdown of FHL2 in A549 cells impeded the interaction between TRIM63 and APC, which suggested that the complex of the two proteins were FHL2 dependent (Fig. 5B). Moreover, knockdown of FHL2 had no obvious impact on the level of TRIM63 (Fig. 5C). Then, TRIM63 was knocked down in H1299 cells overexpressing FHL2 (H1299-FHL2) (Fig. 5D), and western blotting showed that the APC level was enhanced while the expression of β-catenin was dramatically decreased. The downstream of β-catenin like c-Myc, Cyclin D1 and MMP7 was also reduced (Fig. 5E). Additionally, the silencing of TRIM63 relieved the ubiquitination of APC in H1299-FHL2 cells (Fig. 5F). The CCK-8 assay presented the impaired proliferation of H1299-FHL2 after the knockdown of TRIM63 (Fig. 5G). Matrigel Transwell assays confirmed that silencing TRIM63 impeded the migration and invasion of H1299-FHL2 cells (Fig. 5H). Consistently, colony formation assay showed impaired viability of H1299-FHL2 cells after TRIM63 knockdown (Fig. 5I). Thus, these in vitro results verified that FHL2 acted as a scaffold for TRIM63 to facilitate the ubiquitination of APC and activation of the Wnt signaling pathway in LUAD cells.

Fig. 5.

FHL2 acted as a scaffold for TRIM63 in LUAD cells

A. Immunoprecipitations with FHL2 and TRIM63 antibodies were performed in A549 cells to verify the interaction of FHL2 and TRIM63.

B. Co-IP assay with the FHL2 antibody and western blotting was performed in A549 cells with the knockdown of FHL2 (shFHL2#1) to detect the interaction of APC and TRIM63.

C. The impact of FHL2 knockdown on the expression of TRIM63 was verified by immunoblotting.

D. The efficiency of TRIM63 silence in H1299-FHL2 cells was verified by qRT-PCR.

E. After the silencing of TRIM63 in H1299-FHL2 cells (shTRIM63#1 and shTRIM63#2), the levels of APC, β-catenin, c-Myc, Cyclin D1 and MMP7 were detected by western blotting.

F. In H1299-FHL2 cells, immunoprecipitation with APC and IgG antibodies was performed after the knockdown of TRIM63. The ubiquitin level was detected by western blotting.

G. A CCK8 assay was performed in H1299-FHL2 cells to measure proliferation ability after the knockdown of TRIM63.

H. Matrigel transwell assays were performed in H1299-FHL2 cells with TRIM63 silencing to detect the discrepancy in migration and invasion ability. Scale bar = 50 μm.

I. Colony formation assay was performed in H1299-FHL2 cells to measure viability after the knockdown of TRIM63.

D and G-I were compared with one-way ANOVA.

Silencing TRIM63 relieved the FHL2-mediated oncogenesis in LUAD

To further confirm the role of TRIM63 in LUAD cells highly expressing FHL2, TRIM63 was knocked down in H1299-FHL2 cells, and an in vivo xenograft model in nude mice was constructed. Obviously, the silencing of TRIM63 relieved the oncogenic role of FHL2 in promoting the growth of subcutaneous tumors (Fig. 6A-B). We also verified the association among TRIM63, APC and β-catenin in LUAD tissues. IHC staining showed a positive correlation between APC and TRIM63 but a negative correlation between β-catenin and TRIM63 (Fig. 6C). In this study, we confirmed FHL2 as a scaffold for both TRIM63 and APC, which facilitated the destabilization of APC. The degradation of APC stablized the β-catenin/Wnt signaling pathway and contributed to the progression of LUAD (Fig. 6D).

Fig. 6.

Silencing TRIM63 relieved FHL2-mediated oncogenesis in LUAD

A and B. In nude mice, subcutaneous tumors with H1299-FHL2 and the silencing of TRIM63 in H1299-FHL2 were constructed. Tumor growth was measured every three days. After dissociation of tumor tissues, the tumor weight was measured. The discrepancies between the three groups were compared with one-way ANOVA.

C. IHC staining of TRIM63, APC and β-catenin in a TMA containing 128 LUAD tissues. Spearman correlation analysis was performed to analyze the correlation among the three molecules.

D. Graphical abstract of this study.

Discussion

Many studies have confirmed the different roles of FHL2 in the progression of various cancers. In ovarian granulosa cell tumors, FHL2 induced progression via regulation of AKT1 transcription [6]. Additionally, FHL2 could interact with TAB182 to enhance the nuclear translocation of β-catenin and aggravate the progression of esophageal squamous cell carcinoma [13]. However, FHL2 could inhibit the growth of hepatocellular carcinoma via a TGF-β-like signaling pathway [14]. Thus, the different partners interacting with FHL2 determine the oncogenic or tumor suppressor role in various cancers. While FHL2 has been discovered to be associated with poor prognosis and facilitate the viability, migration and invasion of NSCLC [15], the underlying mechanism is still elusive. In this study, bioinformatic analysis and IHC staining in 114 LUAD patients confirmed the high expression of FHL2 in the tumor tissues of LUAD. In addition, the expression of FHL2 was an independent risk factor for recurrence of LUAD and overall survival of LUAD patients. Furthermore, we found that FHL2 could act as a scaffold for APC and the E3 ligase TRIM63. The enhanced expression of FHL2 aggravated the degradation of APC mediated by TRIM63, which activated Wnt/β-catenin signal pathway.

As an evolutionarily conserved pathway that is responsible for multiple biological processes, including organogenesis, homeostasis and cell proliferation, the Wnt/β-catenin signaling pathway has been confirmed to facilitate the oncogenesis and progression of various cancers [16]. Many studies have depicted a comprehensive profile of the Wnt/β-catenin pathway, including extracellular ligands, transmembrane receptors and intercellular compounds. The stabilization of β-catenin is controlled by degradation compounds (Gsk3β, Ck1α and APC) [17]. The roles of FHL2 in the Wnt/β-catenin pathway have been reported in previous studies. In fibroblasts, FHL2 could enhance fibroblast activation and kidney fibrosis via the TGF-β1-induced Wnt/β-catenin pathway [18]. Another study on ovarian cancer verified that miR-340 inhibited the Wnt/β-catenin pathway by acting as a sponge for FHL2 mRNA [19]. However, the intrinsic mechanism by which FHL2 regulates the activation of the Wnt/β-catenin pathway is still elusive. In this study, since APC ranked the most enriched protein among the 7 candidates apart from FHL2 and it regulated the activation of Wnt pathway, we focused on this protein potentially interacted with FHL2. The Co-IP assays further identified the enrichment of these two proteins and further studies presented that FHL2 could impact the post translational modification. We identified that FHL2 could activate the Wnt pathway via the destabilization of APC, which is an important mediator of β-catenin degradation. Mechanistically, FHL2 could form a complex with APC and TRIM63 and facilitate the ubiquitination of FHL2 mediated by TRIM63.

Ubiquitination is a critical mechanism for the posttranslational modification of proteins and participates in extensive cellular processes. Ubiquitin contains 7 residues (K6, K11, K27, K29, K33, K48 and K63), which can form polyubiquitin chains to regulate the activation or silencing of different signaling pathways via the degradation of substrates [20]. It is orchestrated by the following process composed of E1 ubiquitin activating enzyme, E2 ubiquitin conjugating enzyme and E3 ubiquitin ligase. Notably, E3 ubiquitin ligases are the core component of the ubiquitination process for substrate identification and combination [21]. Various mouse models have identified the oncogenic or tumor suppressive function of E3 ligases, as they can regulate stabilization of a diverse set of substrates. Due to substrate diversity, blockade of one E3 ligase might silence multiple processes related to the malignant phenotype [22]. A previous study reported that ubiquitination was an important mechanism for the activation or silencing of the Wnt/β-catenin pathway. CK1 and GSK3, which are responsible for the phosphorylation of β-catenin, are ubiquitinated by β-Trcp and degraded by the 26S proteasome system [23]. In addition, β-Trcp could also facilitate the ubiquitination of β-catenin, which acted as a negative regulator in the activation of Wnt-related targets [24]. TRIM63 has been confirmed as an oncogene in various malignancies. In thyroid carcinoma, TRIM63 aggravates progression via activation of the AKT and ERK pathways [25]. Another study confirmed that TRIM63 could stabilize β-catenin via the phosphorylation of GSK3 [11]. However, the substrates for TRIM63 in LUAD are still elusive. In this study, APC was identified as a novel substrate for TRIM63 in LUAD cells. The enhanced ubiquitination of APC mediated by TRIM63 stabilized the Wnt/β-catenin signal pathway.

In conclusion, this study confirmed the high level of FHL2 in LUAD, which was also an independent predictor for poor prognosis in LUAD patients. FHL2 could act as a scaffold to destabilize APC mediated by the ubiquitination of TRIM63 and further activate the Wnt/β-catenin signaling pathway. Thus, this study identified the oncogenic role of FHL2 in LUAD and uncovered the underlying mechanism.

Abbreviations

LUAD, lung adenocarcinoma; FHL2, Four and a half LIM domain 2 protein; qRT‒PCR, real-time polymerase chain reaction; TMA, tissue microarray; IF, Immunofluorescence; IHC, Immunohistochemistry; Co-IP, Coimmunoprecipitation;

Data Availability Statement: All the data in our study are available upon request.

Funding

This work was supported by the following grants: National Natural Science Foundation of China (81972168), The clinical trial program of Zhongshan Hospital (ZSLC035).

CRediT authorship contribution statement

Jian Gao: Conceptualization, Methodology, Software, Writing – original draft. Yong-Qiang Ao: Methodology, Software. Jie Deng: Data curation, Methodology, Validation. Miao Lin: Data curation, Investigation. Shuai Wang: Investigation, Supervision. Jia-Hao Jiang: Conceptualization, Validation, Writing – review & editing. Jian-Yong Ding: Conceptualization, Funding acquisition, Validation, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.