73 fresh-frozen paired samples of primary lung cancer tissues and normal lung tissues were obtained from the Oncology Department at The Fifth Affiliated Hospital of Sun Yat-sen University. All human sample studies were reviewed and approved by the Ethics Committee of Sun Yat-sen University Cancer Center (GZR2020-154), and informed written consent was obtained from all donors.

All cell lines A-549 (RRID : CVCL_0023), NCI-H1975 (RRID : CVCL_1511), NCI-H292 (RRID : CVCL_0455), HEK293T (RRID : CVCL_0063), NCI-H1993 (RRID : CVCL_1512) were purchase from ATCC and identified by short tandem repeat (STR) profiles from the China Center For Type Culture Collection (CCTCC, Wuhan University, China). All experiments were performed with mycoplasma-free cells. Cells were cultured with Dulbecco’s modified Eagle’s medium media (Thermo Scientific, USA) with 10% fetal bovine serum (FBS, Thermo Scientific, USA). HEK-293T cells were cultured in a six-well plate with complete culture medium the day before and transfected when the cell confluence was approximately 90%. We used a plasmid lentiviral packaging system, and the two packaging plasmids psPAX2 and PMD2.G (Thermo Scientific, USA) and the target gene were mixed at 3 μg:2 μg:1 μg. Then, polyethylenimine (PEI) was added to a quarter of the plasmid. Finally, 200 μL Opti-MEM medium was added to the mixture, and then the mixture was evenly added to the cells. After 6-8 hours, the solution was changed to complete medium. After 48 hours, the cell culture supernatant was collected and filtered with a 0.4-μm filter membrane to obtain virus solution.

The indicated lung cancer cell migration and invasion ability were detected by Transwell assay using 24-well Boyden chambers (BD Inc., USA) with 8-μm pores coated with (invasion) or without (migration) Matrigel. We seeded 5×10⁴ A549 cells or 1×10⁵ (NCI-H1975 and NCI-H292) cells per well on the Transwell inserts and incubated them in 300 μL of serum-free DMEM at 37°C in the top chambers for 12 hours and 24 hours, whereas DMEM containing 20% FBS was added to the lower chamber. Cells that traversed the inserts to the bottom chamber surface were fixed, stained and observed under phase-contrast microscopy.

Briefly, cells were collected and analyzed in lysis buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, 0.1% SDS) supplemented with 1% protease inhibitor (Sigma-Aldrich, USA) after being gently washed 3 times with cold PBS, and then the lysate supernatant was obtained after centrifugation at 13,000 ×g and 4°C for 20 min. Protein (20 μg) was loaded and separated on 10% SDS-PAGE minigels, transferred onto PVDF membranes (Millipore, USA) and blocked with 5% blocking buffer for 1 hour at room temperature. The PVDF membranes were incubated with primary antibody overnight at 4°C with constant shaking, followed by incubation with horseradish peroxidase-conjugated secondary antibody (Promega, USA) for 1 hour at room temperature. The proteins were detected with an enhanced chemiluminescence (ECL) system (Bio-Rad, USA).

Briefly, A549 and NCI-H1975 cells were transfected to express Flag-tagged CTSV. The cells were then lysed in NETN buffer containing 15 mmol/L NaF, 60 mmol/L β-glycerophosphate, and 1 mg/mL each of pepstatin A and aprotinin. The debris was removed and incubated with Flag-conjugated beads at 4°C for 4 hours. Then, the beads were washed with NETN buffer 5 times, the bound proteins were analyzed by SDS-PAGE, and protein mass spectrometry was performed by PTM BioLabs. The immunocomplexes were then washed 4 times with NETN buffer and determined by SDS-PAGE and Western blot.

Lung cancer tissues were fixed in 4% paraformaldehyde for 16 hours at room temperature and then dehydrated in descending concentrations of ethanol. The tumor sections were incubated in 0.3% H₂O₂ solution at room temperature for 20 min and washed with PBS 3 times. Then, the cells were probed with monoclonal anti-CTSV (1:200) at 4°C overnight in a humidified container. After washing with PBS 3 times, the tissue sections were treated with goat anti-rabbit at room temperature for 2 hours. Streptavidin/peroxidase complex and diaminobenzidine were used for immunostaining, and hematoxylin was used for counterstaining.

Antibodies specific for CTSV (WB: Abcam Cat#ab166894, IHC: Theremofish Cat#PA5-47061, ELISA: Theremofish, Cat#PA5-112393, RRID : AB_2261304), Flag-Tag (Cell Signaling, Cat#14793S, RRID : AB_2797401), HA-Tag (Cell Signaling, Cat#3724S, RRID : AB_391833), Tubulin (Genetex, Cat#GTX112141, RRID : AB_1157911) HSP70 (Abcam, Cat#ab2787, RRID : AB_1874830) and GAPDH (Genetex, Cat#GTX100118, RRID : AB_2617427) were purchased from the indicated companies.

The condition medium from indicated cultured cells were collected, and centrifuge at a centrifugal force of 2,000g per minute for 20 minutes, collected the supernatant for centrifuge at a centrifugal force of 10,000g per minute for 1 hour, collected the supernatant and repeated the centrifugation with 100,000g for 1 hour. Finally subjected the supernatant to concentration column and obtained that concentrated medium(CM).

All data were obtained from two or three independent experiments and are presented as the mean ± SD. We conducted the analysis in version 7.00 of GraphPad Prism using the unpaired, two-tailed Student’s t-test module, where ns represents no statistical significance, * represents P value<0.05, ** represents P value<0.01, *** represents P value<0.001, and **** represents P value<0.0001. Graphs were drawn with GraphPad Prism software. Kaplan-Meier survival analyses were used to compare survival among lung cancer patients based on CTSV expression. Western blot bands were quantified with Image-Pro Plus, and the results were normalized to those for the control GAPDH. Statistically significance was defined as P<0.05.

Cysteine cathepsins are pivotal in disease development and progression; they have been reported to be overexpressed in a number of cancers and to lead to increased cancer cell invasion and metastasis (25–27), however, many facets of CTSV activity in lung cancer remain uncharacterized. To explore the role that the cysteine cathepsin family plays in lung cancer pathogenesis, we investigated The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets and compared the gene expression of cathepsins family (CTSB, CTSC, CTSF, CTSH, CTSK, CTSL, CTSO, CTSS, CTSV, CTSW and CTSX) changes in lung cancer tissues and normal tissues. The RNA sequence data from the TCGA demonstrated that the expression of CTSW, CTSH, CTSO and CTSS was downregulated in lung cancer tissues; notably, only CTSV was found upregulated in lung cancer tissues (n=1037), approximately 2.6-fold higher than that in normal tissues (n=108) ( Figures 1A, B ), and no significant changes were observed in the remaining cathepsin family members. In addition, analysis of GEO datasets (GSE31210, GSE19188 and GSE3269) showed that lung cancer exhibits high CTSV levels ( Supplementary Figures S1A–C ). Next, we performed microarray gene expression profile analysis of lung cancer tissues from patients (n=73) with newly diagnosed lung cancer with paired adjacent normal tissues. Immunohistochemical staining of CTSV showed that CTSV expression in lung tumor tissues was higher than that in normal tissues (P<0.0001) ( Figures 1C, D ). We then analyzed the association of CTSV expression with the prognosis of 73 lung cancer patients. Our Kaplan-Meier analysis revealed that high expression of CTSV were correlated with poor overall survival (P=0.028) ( Figure 1E ). Further exploration of the Kaplan-Meier Plotter database (28) showed that elevated CTSV expression was correlated with poor survival in cancer patients ( Figure 1F ). Collectively, these results showed that CTSV is upregulated in lung cancer and is correlated with poor overall survival, implying the oncogenic role of CTSV.

While detecting the protein expression of CTSV in cancer cell lines, we noticed that the CTSV detection by Western blotting revealed more than one band, and the majority of the protein was found at 43 kDa. To determine whether the higher band is responsible for CTSV glycosylation, we treated NCI-H292 cells lysates with peptide-N-glycosidase F (PNGase F) and endoglycosidase (Endo H) to remove the structure of N-glycan and then analyzed the cell lysates by Western blotting. A significant reduction in 39- and 43-kDa CTSV and a remarkable increase in 37-kDa CTSV were observed ( Figure 2A ). The results generally revealed that glycosylated CTSV is present and that the higher band was indeed the glycosylated form of CTSV.

Yuki Niwa et al. observed that cathepsin V is glycosylated at both Asn²²¹ and Asn²⁹² in human fibrosarcoma HT1080 cells (24). To further investigate the glycan locus in lung cancer cells, we used liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) to determine certain sites of glycosylation for CTSV. Glycopeptides carrying N-glycans were identified for 2 N-glycosylation sites, which is consistent with a previous study ( Figures 2B, C ). The substitution of asparagine (N) to glutamine (Q) (N221Q, N292Q or both [2NQ]), caused a certain level of decrease in glycosylation when compared with that in the wild-type (WT) cells ( Figures 2D, E ). When treated with the N-linked glycosylation inhibitor tunicamycin, the glycosylation of CTSV was completely suppressed ( Figure 2F ). Similarly, hemagglutinin (HA)-tagged CTSV had a molecular weight shift from its regular size when treated with Endo H and PNGase F ( Figure 2G ). Together, the results suggested that CTSV is exclusively N-glycosylated at N221 and N292 in lung cancer cells.

Next, we explored the clinical relevance of CTSV glycosylation in the lymph node metastasis of lung cancer. Western blot analysis of 24 paired lung samples showed that glycosylated CTSV was expressed at different levels in lung cancer tissues ( Figures 3A–D ). Importantly, three bands were observed in most lung cancer tissues but one or two bands were observed in normal tissues; therefore, we suspected that the glycosylation of CTSV might be associated with the progression of lung cancer. In this regard, we further assessed the relationship of glycosylation level with lymph node metastasis. Our data showed that the protein levels of the first and second bands of CTSV (the glycosylated forms of CTSV; bands at 43 and 39 kDa) were significantly higher in tumor tissues than in normal tissues, and no significant differences were observed in the third band ( Figures 3E, G, I ). In addition, we found that a higher protein level of the first glycosylation band (band 43 kDa) in lung cancer patients was correlated with lymph node metastasis (P=0.0173), while no significant changes were observed in the second and third bands (39 kDa and 37 kDa) of CTSV ( Figures 3F, H, J and Supplementary Table 1 ). Furthermore, the data also showed that the protein level of total CTSV was significantly higher than that in normal tissues and was correlated with lymph node metastasis as well ( Figures 3K, I ). Together, the results suggested that the level of glycosylation CTSV (band 43 kDa) might be a more sensitive prognostic marker for lung cancer patients with metastasis.

CTSV acts as an intracellular proteolytic enzyme, we suspected that CTSV is secreted from the cells to the intercellular space. To this end, we collected concentrated medium (CM) with a concentration column (10kDa) from the condition medium of indicated cells, which removed microvesicles (MV) and exosome (Exo), and Western blotting was conducted to determine the protein level of CTSV and exosome markers. As a result, we have excluded MV and Exo from condition medium, and 43 kDa CTSV was detected in CM, but not in MV and Exo ( Figures 4A, B ). Additionally, CTSV overexpression led to higher protein levels of CTSV in CM ( Figure 4C ). We then performed migration and invasion assays in A549 cells pretreated with 5μg CM of CTSV-HA-medium or CTSV-HA-high cells(Overexpression effect). We found that the CM of the indicated cells promoted the migration and invasion of A549 cells ( Figure 4D ).

Glycosylation is vital for maintaining protein structure and function. Secreted cathepsins have emerged as potent effectors that modify the tumor microenvironment by degrading the ECM (29). However, the role of N-glycosylated CTSV in determining its pro-metastatic behavior remains unclear. To determine the contribution of glycosylation to CTSV distribution, we performed Western blotting assays in supernatant and whole cell lysate and found that N221Q, N292Q and 2NQ suppressed the secretion of CTSV from intracellular to extracellular matrix ( Figure 4E ). Next, overexpression of CTSV-N221Q, CTSV-N292Q, CTSV-2NQ or wild type CTSV in A549 cells, respectively, we found N221Q, N292Q, both [2NQ] CTSV did not change the migration and invasion ability, but wild-type CTSV significantly increased the migration and invasion of A549 cells ( Figure 4F ). Consistently, we performed migration and invasion assays in A549 cells pretreated with 5μg CM of CTSV-N221Q, CTSV-N292Q, CTSV-2NQ, or CTSV-WT cells, the data revealed that CM from the CTSV-N221, CTSV-N292 or CTSV-2NQ cells showed no significant change in migration and invasion, but greatly promoted cell migration and invasion in wild-type CM group ( Figure 4G ). Interestingly, different from a previous study (24), we found that both glycosylation sites N221 and N292 of CTSV, rather than N292 alone, are important for its secretion, this regulating its promoting effect on tumor metastasis in lung cancer. Collectively, these results supported that glycosylation of CTSV is required for secretion, thereafter determine the metastasis behavior of lung cancer cells.

Our previous evidence support that glycosylated CTSV (band 43 kDa) is particularly expressed in lung cancer samples, and the first glycosylation band (band 43 kDa) of CTSV, which has been secreted outside of the cells, we reasoned that the glycosylated CTSV might serve as a more sensitive prognosis marker for lung cancer. To this end, we isolated serum from the plasma of lung cancer patients and healthy donors, the level of CTSV in circulating serum was significantly higher in patients with lung cancer than that in healthy donors ( Figure 5 ). Together, the data suggested that the level of CTSV in serum distinguished lung cancer patients from healthy donors, revealing that the level of CTSV in serum (glycosylation CTSV) might be a biological marker for lung cancer.