Introduction
Lung cancer is the most common cause of cancer-related death worldwide, accounting for 1.8 million deaths in 2020 [
1]. The development of treatments for non-small cell lung cancer (NSCLC) has been hindered over the past 60 years by disease heterogeneity, complications, and the lack of safe, effective drug therapies [
2]. Although immunotherapy and targeted therapy have significantly improved the efficacy of NSCLC treatment in the past 20 years, most NSCLC patients develop resistance to targeted drugs and disease progression at advanced stages [
3]. Therefore, it is crucial to investigate new molecular markers and develop potential therapeutic targets for the treatment of NSCLC in the future.
Almost all biological processes are controlled by the conjugation of ubiquitin with the post-translational modification of the target protein, so the research on protein ubiquitination continues to expand. Since ubiquitin controls protein stability through the action of hundreds of enzymes, using the ubiquitin system to target specific enzymes and thereby reshape the proteome holds great promise for improving disease prognosis [
4]. USP5 belongs to a ubiquitin-specific protease family of deubiquitinases (DUBs) and is associated with various diseases, including cancer. USP5 prefers to cleave unanchored (not coupled to a target protein) polyubiquitin chains while also being able to remove polyubiquitin chains from protein substrates, so it is essential for free ubiquitin recycling [
5]. In addition, USP5 is also critical for maintaining the steady state of the monoubiquitin pool. Like other USP family deubiquitination enzymes, USP5 is involved in the occurrence and development of various cancers. It has been reported that USP5 overexpressed in many tumors and implicated in its progression, including pancreatic cancer [
6], breast cancer [
7], and glioblastoma [
8]. Wang S et al. found that USP5 protein has five key domains, including the implicit ZnF domain and c-box domain, that interact with c-Maf, demonstrating that c-Maf is a crucial factor in USP5-mediated myeloma cell proliferation and survival [
9]. Liu Y et al. found that the Histidine-rich protein Hpn activates the p14-p53 pathway by inhibiting USP5 and promoting the apoptosis of liver cancer cells [
10]. In addition, studies have shown that USP5 is significantly up-regulated in pancreatic ductal adenocarcinoma (PDAC), and inhibition of USP5 can significantly reduce the growth of PDAC cells, suggesting that USP5 plays a vital role in the occurrence and development of pancreatic cancer [
11]. The above studies show that USP5 plays an essential role in the process of various cancers and is an attractive anticancer target. Therefore, therapies targeting USP5 may play a positive role in cancer treatment.
Several studies have shown the effect of USP5 on NSCLC, including promoting the proliferation of NSCLC cells and stabilizing the expression of PD-L1 [
12,
13]. While in our study, we found that USP5 was highly expressed in NSCLC tissues and was associated with poor prognosis in patients. Further investigations discovered that by inhibiting USP5 with EOAI, NSCLC cells underwent DNA damage and subsequently caused activation of the p53 transcription factor, which caused cell cycle arrest, and induced apoptosis and autophagy in NSCLC cells. We investigated the clinical significance of USP5 in NSCLC and revealed the molecular mechanism of targeting USP5 to treat NSCLC. Our findings provide evidence that reshaping the cancer proteomic landscape using innate degradation mechanisms may impact cancer treatment outcomes and provide new insights into molecularly targeted therapies for NSCLC.
Materials and methods
Antibody and reagents
The primary antibodies used were as follows: total poly ADP-ribose polymerase (PARP) (No. 5625), Bax (No. 5023), Bak (No. 12105), Survivin (No. 2808), cleaved Caspase 3 (No. 9664), cleaved Caspase 9(No.7237), Mcl-1(No. 94296), Bcl-xl (No. 2764), XIAP (No. 2045), c-IAP1(No. 7065), cyclinB1(No. 4138), p-H3(Ser10) (No. 53348), ATF4(No. 11815), c-Myc(No. 18583), LC3(No. 12741), γ-H2AX(No. 9718), Ki-67 (No. 9449), (Cell Signaling Technology, Inc., Boston, MA), P21(No. AP021), P27(No. AP027), Cleaved PARP (No. AF1567), AMPKα(No. AF6195) p-AMPKα (Thr172) (No. AA393), P53(No. AF1162), p-P53(No. AF5893), caspase-9 ( No. AC062), caspase-3 ()(Beyotime, China), Bcl-2 (No. ab32124), USP5 (No. ab154170), Noxa (No. ab13654), FOX3A (No.ab109629), (Abcam Trading Company Ltd), HIF1α (No. 20960–1-AP), c-MYC (No. 67447–1-Ig), p-mTOR (Ser2448) (No. 67778–1-Ig) (Proteintech, Wuhan, China). GAPDH (Hangzhou Goodhere Biotechnology, China) was used as the loading control. All the primary antibodies were diluted in 1:1000 and incubated at 4˚C overnight. The secondary antibodies are peroxidase-conjugated goat anti-mouse IgG and peroxidase-conjugated goat anti-rabbit IgG (diluted in 1:5000, ZGSB. Bio, Inc., Beijing, China). EOAI (No.HY-111408), 3-Methyladenine, (3MA, No. HY-19312), and Pifithrin-β, (PFT, No. HY-16702), Cisplatin, (CDDP, No. HY-17394) were purchased from Med Chem Express (MCE).
Immunohistochemistry (IHC)
USP5 expression was assessed in neoplastic cells and peritumoral tissue. The human NSCLC tissue array (HLugA180Su07) was bought from Shanghai Outdo Biotech Co. Ltd. Tumor tissue was dehydrated and incubated with peroxidase. Antigen retrieval was then performed in a pressure cooker using 0.01 mol/L citrate buffer (pH 6.0). Incubate USP5 antibody (dilution: 1:200), γ-H2AX (dilution: 1:200), Ki-67 (dilution: 1:1000) overnight at 4 °C according to the instructions. The slides were then stained with a Histostain-Plus kit (SP-9000) and 3,3-diaminobenzidine tetrahydrochloride (DAB) (ZLI-9032) (ZSGB-BIO, Beijing, China). Finally, nuclei were stained with hematoxylin. The IHC results were scored by two senior pathologists in a double-blind manner. The following proportion scores were assigned to the sections: 0 if 0% -5% of tumor cells exhibited positive staining, 1 for 6%-25% positive cells, 2 for 26%–50% positive cells, 3 for 51%–75% positive cells, 4 for 76%–100% positive cells. In addition, the staining was scored on a scale of 0–3: 0, negative; 1, weak; 2, moderate; and 3, strong. The positive grade of IHC was determined according to the positive proportion of tumor cells and staining intensity: 0 was classified as negative (-), 1–3 as weakly positive ( +), 4–5 as positive (+ +), and 6–7 as strongly positive (+ + +). Survival analysis was performed according to USP5 expression, with low USP5 expression including (-) and ( +), and high expression including (+ +) and (+ + + +).
Cell lines and culture conditions
NSCLC cell lines (A549, H460, Cslu-3, H3122) and Normal lung epithelial cell line (BEAS-2B) were purchased from the Shanghai Cell Resource Center Academy of Sciences. All cell lines were tested by STR (QuiCell Biological, Shanghai, China) and showed no cross-contamination of cell lines. Cells were cultured in a complete medium containing 90% DMEM (No. 10–013-CVRV, CORNING, China) with 4 mM glutamine (No. 2323081, Glibco) and 1% penicillin–streptomycin (No.C0224, Beyotime, China) and 10% fetal bovine serum (No. FSP500, ExCell Bio) in a 37 °C cell incubator with 5% CO2. Cells were passaged every 2–3 days, and cells in the logarithmic growth phase were selected for experiments.
Immunoblotting
Cells were incubated with DMSO or EOAI. After 48 h, cells were lysed in the RIPA lysis buffer containing 150 mM NaCl, 50 mM Tris–HCl pH 7.5, 1% NP-40, 1 mM permethylsulfonate, 1 mM orthovanadate, 1% SDS, 1% protease inhibitors at 4 C. Cell extracts were subject to centrifugation to remove insoluble materials. Total protein concentration was determined by BCA assay (No. P0011, Beyotime, China), and the protein was boiled at 95℃ for 5 min. Proteins were submitted to SDS-PAGE and then transferred onto a polyvinylidene fluoride (PVDF) membranes (No. IPVH00010, Millipore, Germany) by electroblotting. The saturation was done in TBS-Tween 0.1% containing 5% bovine serum albumin, 3% skimmed milk (for anti-TrkA), or 0.2% casein (for anti-sortilin) for 1 h at room temperature. Membranes were incubated with primary antibodies in saturation buffer (overnight, 4 °C). After being washed with TrisBuffered Saline Tween-20 (TBST), the membrane was incubated with secondary antibodies for 2 h at room temperature. ECL Kit (No. P0018, Beyotime, China) and Automatic chemiluminescence image analyzer (Tanon5200, Shanghai, China) visualized the proteins.
Cell proliferation and cell colony-forming unit assay
For cell proliferation assay, A549, H460, Calu-3, and H3122 cells in the logarithmic growth phase were seeded in 96-well with 3000–5000 cells per well. Cells were treated with EOAI dissolved in DMSO and prepared at different concentrations (1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5 μmol/L) for 48 h. DMSO-treated cells (0.1%) were used as the control group. Then, CCK-8 reagent (No.IV08, Invigentech, USA) was added to each well and incubated at 37 °C for 1 h, and a microplate reader measured the absorbance at 450 nm to quantify cell viability. For the colony-forming unit assay, cells were seeded in 6-well plates at 500–1000 cells per well. DMSO or EOAI was added 24 h after the cells adhered and cultured for another 7–10 days. After washing with phosphate-buffered saline (PBS; No. D8537, Sigma), the cells were fixed with 4% paraformaldehyde (No. P1110, Solarbio, China) for 15 min, then stained with crystal violet solution (No. G1063, Solarbio, China) for 15 min, and counted after washing.
Cell cycle analysis
Cells after EOAI treatment for 24 h were collected with 0.05% trypsin without EDTA solution and fixed overnight at 4 °C by adding 70% ethanol. After centrifuging, cell suspension with PBS containing 50 μg/ml RNase A and 10 μg/ml PI was kept at 37 °C in the dark. The cell cycle was evaluated by fluorescence activating cell sorter (Becton Dickinson FACScan; Becton–Dickinson, San Jose, CA, USA). The results of flow cytometry were analyzed by ModFit LT 3.1 software.
Cell apoptosis assay
The NSCLC cells treated with DMSO and EOAI for 48 h were collected and stained with Annexin V-FITC/PI at RT for 15 min in the dark, and then the apoptosis ratio was detected by flow cytometry. The cells were collected similarly, and the Casp-GLOW Fluorescein Active Caspase 3 Staining kit (BioVision, Inc., Milpitas, CA, USA) was selected to detect the activity of Caspase 3.
Observation of mitochondrial membrane potential
The cells (1 × 106/ml) were inoculated into 6-well plates and treated with different concentrations of EOAI and 0.1% DMSO for 24 h. Mitochondrial membrane depolarization was detected with the mitochondrial membrane potential assay kit with JC-1 according to the manufacturer's protocol (Yeasen, Shanghai, China). The data were acquired and analyzed by flow cytometry. Cells with intact mitochondria displayed high red fluorescence and appeared in the upper right quadrant of the scatterplots.
Gene knockdown using siRNA
Cells plated in 6-well plates were transfected with two different sequences of siRNAs targeting Noxa and Control using jetPRIME transfection reagent (No. 23Y2307M9, Polyplus-transfection, SA) when 50% confluent. Cells were then incubated with a transfection mixture of 100 nM siRNA and jetPRIME for 24 h. Cells transfected with Noxa and control were used for Cell apoptosis assay or Immunoblotting. All siRNAs were synthesized in Shanghai GenePharma Co. Ltd, China: siNoxa: 5’-GUAAUUAUUGACACAUUUC-3’; siControl:5’- UUCUCCGAACGUGUCACGU-3’.
Immunofluorescence
The cells were seeded in a petri dish, and after the cells adhered, they were treated with DMSO or EOAI for 24 h. Discarded the medium, washed twice with PBS, and added 4% paraformaldehyde to fix the cells for 20 min, then added anhydrous methanol and placed at -20 °C for 10 min to infiltrate. After blocking with 5% BSA for 20 min at room temperature, LC3/γ-H2AX (1:200) primary antibody was added separately at 4 °C overnight. After that, Alexa Fluor® 488 Goat anti-Rabbit secondary antibody (green, dilution:1:500) (Beyotime, China) was added for 2 h at RT. Finally, the nuclei were labeled with DAPI (5 μg/ml, Beyotime, China, blue) and placed at RT for 20 min. Photographs were taken with a fluorescence microscope (Olympus CKX53).
Comet assay
Cells were incubated with 0.1% DMSO or EOAI for one day. Mix 20uL of the cell suspension with 100uL of 0.5% low melting point agarose solution and spread it on a solidified 0.5% standard melting point agarose-coated glass slide. Then cover with a cover slip and cure at 4 °C for 30 min. The slides were placed in lysis buffer (2.5 mol/L NaCl, 100 mmol/L EDTA, 10 mmol/L Trizma base, 10% DMSO, and 1% Triton X-100) at 4 °C for 2 h. Under dark conditions, place the slide in the electrophoresis tank, 300 mA, 20 min. The microscope slides were rinsed with 0.4 M Tris HCl and stained with PI (25 μg/ml) for 10 min, and observed the comets and photographed under a fluorescence microscope (Olympus CKX53).
Xenograft model in C57 mice
Xenograft model in C57 mice 6–8 weeks-old specific pathogen free (SPF) male mice (Beijing Vital River Laboratory Animal Technology Co. Ltd, China) weighing 17-19 g, were randomly divided into four groups with 5 mice in each group. Under the pathogen-free environment, all animals were cared for and anesthetized before experimentation following the Guide for the Care and Use of Laboratory Animals. A mouse subcutaneous lung cancer model was established with Lewis lung cell (LLC) suspension (5 × 105 cells in 100ul PBS), and drug (EOAI:15 mg/kg and CDDP:2.5 mg/kg) treatment was given 3 days after the tumor was successfully implanted. Mice were weighed every 3 days, and tumor size was measured using a vernier caliper. The volume of the tumor is calculated based on its long diameter and short diameter. Tumor volume was obtained by the formula: (length × width2)/2. Mice were sacrificed by cervical dislocation on the 21st day after cell implantation. The tumors were collected for IHC analysis.
Quantification and statistical analysis
All data were performed in at least three independent experiments. To quantify the cell colony-forming unit assay, count the number of cells in the dish using ImageJ software. Data were shown as mean SD, and error bars represented the standard deviations from three independent experiments. Data graphics and descriptive statistics were presented by using Graphpad prism 8, SPSS 25.0 and the Microsoft Excel data analysis package. The significance between the two groups was obtained using the Student’s t-test. Overall survival (OS) was defined as the time from the data of diagnosis to death or the last follow-up examination. Survival curves were performed by the Kaplan–Meier method, and groups were compared using log-rank tests. ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Discussion
The bottleneck of cancer treatment is the gradual development of drug resistance during treatment. Therefore, it is necessary to continuously explore the molecular mechanism of lung cancer progression and explore new drug therapeutic targets. There is much evidence that USP5 is involved in tumor progression, including promoting the process of tumor epithelial-mesenchymal transition (EMT), stabilizing cyclins to promote proliferation, activating inflammasome, etc. [
23‐
25]. Therefore, USP5 is enormously appealing as a strategy for targeted tumor therapy. Our results suggested that USP5 is overexpressed in NSCLC tissues and cell lines, and this high expression is significantly correlated with tumor size and patient survival. This result provides a basis for the inhibition of USP5 to shrink the tumor for therapeutic purposes.
Based on the results of Kapuria V et al., drugs with DNBs inhibitory activity have shown attractive antitumor effects [
26]. It is generally accepted that cancer cells exhibit more genomic instability and are more dependent on the DDR pathway to process endogenous and exogenous DNA damage [
27]. When the DDR process is compromised, lethal DNA damage occurs, leading to cell growth arrest and apoptosis. Studies by Nakajima S et al. have shown that USP5 is involved in DNA repair and affects the efficiency of double-strand break (DSB) repair in homologous recombination by eliminating ubiquitin signaling at sites of DNA damage [
28,
29]. These results demonstrate that tumor cells may have a greater need to maintain USP5 activity and, thus, DNA stability than normal cells. Consistent with these reports, our results showed that inhibition of USP5 cells showed a longer tail in the comet assay. The immunofluorescence results showed that inhibition of USP5 resulted in more γ-H2AX-positive cells and a significant upregulation of γ-H2AX after inhibition of USP5 was detected at the protein level, indicating significant DNA damage in NSCLC cells.
Ubiquitination is involved in the modification, degradation and functional regulation of proteins and has been extensively studied in the context of DNA damage response (DDR). Different ubiquitination pathways have various molecular structural features and biological functions, including ubiquitin chains that play essential functions in the DDR process [
30]. In recent years, it has become apparent that altering such ubiquitination by DUBs plays a crucial role in regulating DNA damage-related events and the activities they control [
31]. EOAI, an inhibitor of USP5 deubiquitinating enzymes, showed a dose-dependent tumor growth inhibition in our results. Mechanistically, EOAI can trigger DNA damage in NSCLC cells, causing a significant upregulation of the appearance of γ-H2AX expression. At the molecular level, we found that EOAI induced DNA damage, activated p53, and effectively inhibited NSCLC cell growth through cell cycle arrest and apoptosis.
In DNA damage, p53 is considered a decision-making transcription factor that selectively activates genes as part of specific gene expression programs to determine cellular outcomes [
32]. Like most post-translational modifications, the ubiquitination of p53 can be reversed by the counteracting action of deubiquitinases. Some USP family members have been shown to regulate the p53 pathway [
33,
34]. Studies have shown that multiple genes involved in cell cycle regulation are repressed in a p53-dependent manner [
35]. The p53 tumor suppressor protein has a vital role in protecting genome integrity. In unstressed cells, p53 is maintained at low levels through the ubiquitin–proteasome pathway, and activated p53 transcriptionally regulates multiple biological processes, including DNA damage repair, cell cycle arrest, apoptosis, and autophagy. Das S et al. showed that the p53 target gene encoding zinc finger protein 16 binds to p53 and increases its binding and transactivation to the cell cycle arrest genes CDKN1A and SFN, thereby causing cell cycle arrest [
36]. Our study found that after inhibition of USP5, NSCLC cells were arrested in the G2/M phase, and further protein level analysis found that cells were mainly arrested in the G2 phase. Following the addition of the p53 inhibitor PFT, G2/M arrest was reduced, and there was also a callback in cyclin levels. The above results suggest that the cycle arrest caused by EOAI may be caused by the activation of p53 by DNA damage.
Studies have shown that P21 can increase the levels of pro-apoptotic proteins such as Bax, and change mitochondrial permeability, thereby playing a pro-apoptotic role [
37]. Previous studies have reported increased apoptosis in tumor cells, and the upregulation of the apoptotic protein Noxa depended on the activation of p53 [
17]. Similarly, after inhibition of USP5, we found that NSCLC cell apoptosis was increased, while the expression of pro-apoptotic proteins was up-regulated, and anti-apoptotic proteins were down-regulated. Inactivation of p53 reduced EOAI-induced apoptosis and the expression of the apoptotic protein Noxa after adding the p53 inhibitor PFT. Furthermore, we demonstrated that blocking endogenous Noxa induction can lead to the inhibition of NSCLC cell apoptosis. This is consistent with the results of Zhendong Yu et al., which showed that cell DNA damage could promote apoptosis of tumor cells through the p53-PUMa /NOXA/ BCl2-Bax pathway, and kill cancer cells by inducing cell cycle arrest through the p53-P21 pathway [
38].
The reports of DUBs stably participating in autophagy-related proteins by deubiquitination and their involvement in autophagy are not isolated cases. The study by Xiao W et al. found that E2F4 (E2F transcription factor 4) directly regulates the transcription of autophagy-related proteins (autophagy-related 2A and unc-51 like autophagy activating kinase 2). At the same time, USP2 stabilizes the E2F4 protein through deubiquitination, reducing its transactivation in tumor cells and thereby inhibiting its protective autophagy [
39]. Similarly, Cai B et al. demonstrated that USP5 could promote autophagic degradation [
25]. Our results also prove that inhibition of USP5 can promote autophagy in NSCLC cells, and this promotion effect is inhibited after the inhibition of p53. Mechanistically, after toxic gene stress, p53 activates AMPK while inhibiting mTOR, translation and ribosome biosynthesis are inhibited, and autophagy is activated [
40]. Adenosine monophosphate-activated protein kinase (AMPK) was demonstrated by Egan DF et al. in 2011 to induce autophagy in mammalian cells by activating ULK1 and ULK2 [
41]. Our study confirmed that EOAI triggered autophagy in cells, and the addition of autophagy inhibitors enhances EOAI-induced apoptosis in NSCLC cells. In addition, the activity of AMPK was down-regulated after adding PFT, while EOAI-induced autophagy in NSCLC cells was also inhibited.
After a series of in vivo and in vitro studies, EOAI has been shown to have a promising antitumor effect in various tumors [
42‐
44]. In vivo experiments, EOAI has been shown to reduce tumor volume and, when combined with cisplatin, to have a better antitumor effect than a single agent. Inactivation of USP5 by EOAI can effectively inhibit the growth of NSCLC cells in vitro or in mice. These results suggest targeting USP5 as a potential anti-NSCLC therapeutic strategy.
Collectively, we observed significant DNA damage in NSCLC treatment with the USP5 inhibitor EOAI, as well as cell cycle arrest, apoptosis, and autophagy induced by p53 transcription factor activation. These results enrich the epigenetic relationship of NSCLC, suggesting that EOAI may be a new therapeutic agent targeted to kill NSCLC cells, providing theoretical support for EOAI to inhibit tumor growth in vivo. In addition, we provided information on the expression levels of USP5 in NSCLC tissues and cell lines. From the perspective of expression, the high expression of USP5 indicates a poor prognosis, suggesting that USP5 can not only be used as a therapeutic target of NSCLC but also may have the potential as a diagnostic marker of NSCLC.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.