Downregulation of pro-surfactant protein B contributes to the recurrence of early-stage non-small cell lung cancer by activating PGK1-mediated Akt signaling

Cycloheximide, MG132, 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI), Dulbecco`s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), crystal violet, agar, cell proliferation and cytotoxicity assay kit (MTT), and actin antibody were purchased from Sigma-Aldrich (St.Louis, MO, USA). H1299, and H522 cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA). PC-9 cell line was kindly provided by Dr. Shen (Jilin University, China). A cell counting kit-8 (MTT kit) and transwell were obtained from Biosharp (Wuhan, China) and BD Biosciences (Franklin Lakes, NJ, USA), respectively. VECTASTAIN ABC Kits (HRP) from Vector Laboratories (Burlingame, CA, USA). Antibodies against PGK1, ubiquitin, ADRM1, Flag, pro-SFTPB, HRP-conjugated anti-mouse IgG, FITC-conjugated anti-mouse IgG, and Texas red-conjugated anti-mouse IgG were purchased from Proteintech (Rosemont, IL, USA). Antibodies against to p-Akt (Ser 473), Akt, UCH37, hRpn2, p-mTOR, and mTOR were obtained from Abcam (Cambridge, UK). Pro-SFTPB elisa kit was purchased from Shanghai Baiyi Biotechnology Co., Ltd. (Shanghai, China). siRNAs of PGK1, ADRM1 and UCH37 genes were obtained from Shanghai GeneBio Co., Ltd (Shanghai, China) and shRNAs of pro-SFTPB were obtained from Hanbio Biotechnology Co., Ltd. (Shanghai, China). The target sequences of siRNA and shRNA were summarized in Additional file 1: Table S2.

All cells were cultured in DMEM supplemented with 10% FBS at 37 ℃ in an atmosphere of 95% air and 5% CO₂. Human specimens were collected at Daping Hospital of Army Medical University during surgery under a protocol approved by the ethical review committees. Characteristics of the patients were summarized in Additional file 1: Table S3 and S4.

Analysis was performed 24 h after transfection of cells with indicated constructs or nucleotides. For cell viability assay, cells were seeded into 96-well plates at a density of 5 × 10³ cells per well, and cell viability was measured at indicated time using MTT assay kit according to the manufacturer`s protocol. For transwell assay, cells in growth medium without FBS were plated in the upper wells of 24-well chamber at density of 1 × 10⁴ cells per well. The lower wells of chamber contained cell growth medium supplemented with 10% FBS. After 24 h of seeding, cells in the lower side of the chamber were fixed, stained and counted. For soft agar assay, cells were mixed in 0.5 ml 0.35% agar in growth medium and plated on the top of a solid layer of 0.8% agar in growth medium at a density of 3000 cells per well (6-well plate). Colonies were counted 10 days later.

Western blot, Co-IP, IHC, and IF were performed as described previously [15]. After IF analysis, the colocalization between proteins were measured using Image J 1.53t (National Institutes of Health, USA).

H1299 cells were transfected with indicated plasmids. After 72 h of transfection, cells were harvested and extracted proteins, and then subjected to analysis. Proteomics analysis was performed as described [16].

6-weeks old male nude mice were used in this study. For tumorigenesis experiment, 1 × 10⁶ indicated cells in 1 ml PBS were injected subcutaneously on the back of mice. After, 3 weeks, mice were sacrificed and the tumors were weighed. For lung metastasis experiment, 1.5 × 10⁶ indicated cells in 1 ml PBS were injected into mice via tail vein. After, 3 weeks, mice were sacrificed and the lungs were collected and the tumor nodules on the lung surface were counted under a microscopy.

Data were presented as mean ± standard deviation and the differences between groups were considered statistically significant at a p value of less than 0.05. The differences between two groups were analyzed by a student t test using SAS statistical software (SAS Institute). The survival rate of patients with early-stage NSCLC was calculated by Kaplan–Meier survival analysis.

Based on their immunohistochemistry (IHC) results, early-stage lung adenocarcinoma (LUAD) patients who showed no difference in pro-SFTPB expression in the tumor and its adjacent tissue were placed in the pro-SFTPB normal expression group, and patients in whom pro-SFTPB expression was lower in the tumor than in the adjacent tissue were placed in the pro-SFTPB low expression group (Fig. 1a). We then analyzed the correlation between pro-SFTPB expression and progression of early-stage LUAD. The results showed that the group with low pro-SFTPB expression had a significantly higher recurrence rate (Fig. 1b), shorter recurrence-free survival (RFS) (Fig. 1c) and shorter overall survival (OS) (Fig. 1d) than the group with normal pro-SFTPB expression. Specifically, the median time of recurrence (MTR) and median survival time (MST) of the pro-SFTPB expression group were 77 months and 107 months, respectively, while the MTR and MST of the low pro-SFTPB expression group were 49 months and 83 months, respectively. Consistent with this, RNA sequencing data showed that early-stage LUAD patients who experienced recurrence within 24 months after surgery had lower levels of expression of SFTPB than patients who experienced no recurrence within 5 years after surgery (Additional file 1: Fig. S1). In addition, we investigated the clinical significance of low levels of serum pro-SFTPB in the progression of early-stage LUAD. Patients with early-stage LUAD whose serum pro-SFTPB concentration was within the range of pro-SFTPB concentrations in healthy people (Mean ± SE) were placed in the normal pro-SFTPB group, and patients whose serum pro-SFTPB concentration was lower than the pro-SFTPB concentration in healthy people were placed in the low pro-SFTPB group (Fig. 1e). Our data show that, among these patients with early-stage LUAD, the low pro-SFTPB group had both lower RFS (p < 0.0001) (Fig. 1f) and lower OS (p = 0.0029) (Fig. 1g) than the normal pro-SFTPB group. Taken together, our findings suggest that low expression of pro-SFTPB is associated with poor prognosis and that it may be involved in the stimulation of recurrence in early-stage LUAD patients.

Next, we investigated whether downregulation of pro-SFTPB directly promotes metastasis and tumorigenicity in NSCLC cells. NSCLC cell lines that highly expressed pro-SFTPB were selected (Additional file 1: Fig. S2), and pro-SFTPB expression in these cells was silenced using shRNAs of pro-SFTPB (Fig. 2a). Our data show that downregulation of pro-SFTPB expression significantly promoted NSCLC cell invasion (Fig. 2b), migration (Fig. 2c), viability (Fig. 2d), and colony formation in soft agar (Fig. 2e). These in vitro results were confirmed in animal models. Animal experiments showed that whereas subcutaneous injection of 1 × 10⁶ H1299 cells resulted in tumor formation in 67% of mice within 3 weeks, injection of H1299 cells in which pro-SFTPB had been downregulated resulted in tumor formation in 100% of the animals (Fig. 2f). In addition, downregulation of pro-SFTPB dramatically stimulated NSCLC growth in subcutaneous xenograft models (Fig. 2f, g). Furthermore, an experiment in which a lung metastasis model was used showed that downregulation of pro-SFTPB significantly increased the number of tumor colonies in the lungs (Fig. 2h). In contrast, ectopic expression of pro-SFTPB significantly inhibited growth and metastasis of NSCLC with low expression of pro-SFTPB (Additional file 1: Fig. S3). Together, these findings suggest that pro-SFTPB negatively contribute to NSCLC cell metastasis and tumorigenicity.

To investigate the mechanism by which downregulation of pro-SFTPB promotes NSCLC progression, we performed gene set enrichment analysis (GSEA) using an early-stage NSCLC dataset from the TCGA database. The GSEA results showed that SFTPB expression correlates negatively with activity of the Akt signaling pathway in early-stage NSCLC (Fig. 3a). Consistent with this, in vitro experiments showed that downregulation of pro-SFTPB expression significantly increased the phosphorylation of both Akt and its downstream protein mTOR in NSCLC cells (Fig. 3b). The promoting effect of pro-SFTPB downregulation on Akt phosphorylation was further confirmed by IHC in tumors from xenografted mice and lung metastasis animal models (Fig. 3c). Notably, inhibition of Akt activity by the Akt inhibitor MK2206 (Fig. 3d) significantly suppressed the increase in colony formation in soft agar (Fig. 3e) and the increased invasion by NSCLC cells (Fig. 3f) that were otherwise induced by downregulation of pro-SFTPB expression. These findings indicate that downregulation of pro-SFTPB expression promotes NSCLC progression by activating the Akt pathway.

To investigate how downregulation of pro-SFTPB activates the Akt pathway, we performed a proteomics analysis of control cells and NSCLC cells in which pro-SFTPB had been downregulated. As shown in Fig. 4a and Table S1, many proteins were upregulated after downregulation of pro-SFTPB expression in NSCLC cells. We chose PGK1 for further analysis because it is an activator of the Akt pathway [17] and stimulates tumor progression by promoting glycolysis [18]. The effect of downregulation of pro-SFTPB expression on PGK1 expression was further examined at the mRNA and protein levels. Our results show that downregulation of pro-SFTPB upregulated the protein level of PGK1 (Fig. 4b) but that it did not affect the mRNA level of PGK1 in NSCLC cells (Fig. 4c). In addition, downregulation of pro-SFTPB expression increased the half-life of PGK1 protein when the cells were treated with the protein synthesis inhibitor cycloheximide (CHX) (Fig. 4d) but did not affect PGK1 level when the cells were treated with the proteasome inhibitor MG-132 (Fig. 4e). Downregulation of pro-SFTPB expression also reduced PGK1 ubiquitination, while overexpression of pro-SFTPB increased it (Fig. 4f). These findings suggest that downregulation of pro-SFTPB expression upregulates PGK1 expression at the posttranscriptional level by inhibiting its ubiquitination. Notably, downregulation of PGK1 expression by siRNAs (Additional file 1: Fig. S4) inhibited the stimulation of Akt phosphorylation (Fig. 4g), invasion (Fig. 4h), and soft agar colony formation (Fig. 4i) by NSCLC cells that otherwise occurred after downregulation of pro-SFTPB expression, suggesting that downregulation of pro-SFTPB promotes Akt pathway-regulated NSCLC progression by upregulating PGK1 levels.

To investigate how pro-SFTPB regulates PGK1 ubiquitination, we collected proteins bound to pro-SFTPB by coimmunoprecipitation (Co-IP) and identified them by mass spectrometry (Fig. 5a). The results of the mass spectrometry analysis showed that ADRM1 may be a binding partner of pro-SFTPB in NSCLC cells (Fig. 5a). The interaction (Fig. 5b) and colocalization (Fig. 5c) of ADRM1 and pro-SFTPB in NSCLC cells were further confirmed by co-IP and immunofluorescence (IF), respectively. ADRM1 is an ubiquitin receptor that is involved in protein deubiquitination by activating the deubiquitinating enzyme UCH37 (ubiquitin carboxy-terminal hydrolase 37) [19]. Importantly, we found colocalization of ADRM1 and PGK1 (Fig. 5d), and silencing of ADRM1 or UCH37 decreased PGK1 protein levels (Fig. 5e) and increased PGK1 ubiquitination in NSCLC cells (Fig. 5f), suggesting that the ADRM1/UCH37 axis is involved in the regulation of PGK1 deubiquitination. Silencing of ADRM1 or UCH37 also suppressed the downregulation of pro-SFTPB-induced upregulation of PGK1 protein (Fig. 5g) and downregulation of PGK1 ubiquitination (Fig. 5h). Together, these findings suggest that the ADRM1/UCH37 axis is involved in the regulatory effect of pro-SFTPB on the ubiquitination of PGK1 in NSCLC.

To investigate the mechanism by which pro-SFTPB regulates PGK1 deubiquitination through the ADRM1/UCH37 axis, we examined whether pro-SFTPB affects the formation of complexes between ADRM1 and its partner proteins. Previous studies have shown that hRpn2 activates ADRM1 by binding directly to ADRM1 and that activated ADRM1 interacts with the deubiquitinase UCH37 and then exerts its deubiquitination effect [20]. As shown in Fig. 6a and b, pro-SFTPB negatively regulated the binding of hRpn2 and UCH37 to ADRM1. In addition, mimetic molecular docking between pro-SFTPB and ADRM1 by HDOCK showed that pro-SFTPB may bind to the N-terminal Pru (pleckstrin-like receptor for ubiquitin) domain (Fig. 6c) of ADRM1; this domain is important in the activation of ADRM1 by hRpn2 through physical binding (Fig. 6c) [20]. Based on these results, we hypothesized that binding of pro-SFTPB to ADRM1 suppresses hRpn2 binding to ADRM1 and thereby inhibits ADRM1 activation and the interaction between ADRM1 and UCH37. As expected, unlike overexpression of wild-type pro-SFTPB, overexpression of pro-SFTPB mutated at the ADRM1 binding site did not affect the interaction of ADRM1 with hRpn2 or the interaction of ADRM1 with UCH37 (Fig. 6d). Additionally, overexpression of mutated pro-SFTPB did not inhibit ADRM1-induced upregulation of PGK1 (Fig. 6e) and downregulation of PGK1 ubiquitination (Fig. 6f). Taken together, these findings suggest that pro-SFTPB inactivates ADRM1 by blocking the binding of hRpn2 and UCH37 to ADRM1, thereby inhibiting the deubiquitination of PGK1 by ADRM1 and ultimately leading to ubiquitination and degradation of PGK1.

Finally, we used clinical samples to examine whether low expression of pro-SFTPB in tumors correlates with the expression of PGK1 and p-AKT in patients with early-stage LUAD. The expression of pro-SFTPB, PGK1 and p-Akt in clinical samples was examined by IHC (Fig. 7a). As shown in Fig. 7a, c, a correlation study of 160 early-stage LUAD clinical samples showed that low expression of pro-SFTPB correlated significantly with high expression of PGK1 and p-Akt in early-stage LUAD. Specifically, 70% (62 cases) and 66% (58 cases) of the samples that showed low pro-SFTPB expression showed high expression of p-Akt (Fig. 7b) and PGK1, respectively (Fig. 7c). In addition, consistent with the results of in vitro experiments, pro-SFTPB not only decreased the expression of PGK1 but also blocked the binding of PGK1 to ADRM1 in early-stage LUAD tissues (Fig. 7d).