Correction of the tumor suppressor Salvador homolog-1 deficiency in tumors by lycorine as a new strategy in lung cancer therapy

Salvador homolog-1 (SAV1) is a tumor suppressor required for activation of the tumor-suppressive Hippo pathway and inhibition of tumorigenesis. SAV1 is defective in several cancer types. SAV1 deficiency in cells promotes tumorigenesis and cancer metastasis, and is closely associated with poor prognosis for cancer patients. However, investigation of therapeutic strategies to target SAV1 deficiency in cancer is lacking. Here we found that the small molecule lycorine notably increased SAV1 levels in lung cancer cells by inhibiting SAV1 degradation via a ubiquitin–lysosome system, and inducing phosphorylation and activation of the SAV1-interacting protein mammalian Ste20-like 1 (MST1). MST1 activation then caused phosphorylation, ubiquitination, and degradation of the oncogenic Yes-associated protein (YAP), therefore inhibiting YAP-activated transcription of oncogenic genes and tumorigenic AKT and NF-κB signal pathways. Strikingly, treating tumor-bearing xenograft mice with lycorine increased SAV1 levels, and strongly inhibited tumor growth, vasculogenic mimicry, and metastasis. This work indicates that correcting SAV1 deficiency in lung cancer cells is a new strategy for cancer therapy. Our findings provide a new platform for developing novel cancer therapeutics.

The SAV1 gene is infrequently mutated in human cancers; however, its expression in cancer tissue is frequently downregulated either epigenetically or post transcriptionally 15,16,27 . Both gene knockout and downregulated expression of SAV1 promote carcinogenesis 17,36,37 . Notably, SAV1 deficiency is closely associated with poor prognosis for cancer patients 38 . In contrast, SAV1 overexpression by transfecting cancer cells with SAV1 cDNA-plasmid inhibits tumorigenesis and improves survival of tumor-bearing mice 30 . However, there are no reported therapeutics that correct SAV1 deficiency in cancer.
In this study, we explored exogenous agents that increase SAV1 levels in cancer cells and found that lycorine can correct SAV1 deficiency in lung cancer.
Lycorine is a small molecule derived from Lycoris radiate and has strong anticancer effects and low toxicity 18,39 . Lycorine markedly increased SAV1 levels in cancer cells by inhibiting SAV1 degradation via the ubiquitin-lysosome system. Lycorine-induced increases of SAV1 levels activated MST1 and triggered degradation of oncogenic YAP. Higher SAV1 levels also inhibited YAP-activated transcription of various oncogenic genes and tumorigenic AKT and NF-κB signal pathways. Together, these effects resulted in strong inhibition of tumor growth, vasculogenic mimicry, and metastasis in tumor-bearing mice. Our findings suggest a new strategy for effective cancer therapy.

Cell culture and lycorine treatment
Human lung cancer cell lines SPC-A-1 and A549 were cultured in DMEM supplemented with 10% fetal bovine serum (the complete medium), at 37°C in a humidified atmosphere of 5% CO 2 and treated with lycorine at the concentrations of 0-10 μM, as previously described.
Cell proliferation assay SPC-A-1, A549, and HBE Cells were seeded in a 96-well plate and incubated with series concentrations of lycorine (0-60 μM). After 3 days, 10 μl of the CTG solution (5 mg/ ml) was added to each well and incubated for 10 min at 37°C, then the dual fluorescence wavelength absorbance at 490 nm was measured using SpectraMax M5 multidetection reader, and the IC50 was assessed by SPSS 16.0.

Cell-invasion assay
The ability of cell invasion was detected via transwell assay as we described before 40 . SPC-A-1 and A549 cells in 200 μl of serum-free DMEM were seeded to the upper chamber, and 500 μl of the complete medium was added to the lower chamber. Then, lycorine at the concentration of 0-10 μM was added to the upper and lower chambers. After incubation for 24 h, the membrane of the chamber was strained with Wright-Giemsa solution, and photographed under a microscope.
Wound-healing assay A549 cells were seeded in six-well plats for 24 h, wounded with a 200-μl plastic tip. Then the cells were incubated in 2% FBS DMEM with lycorine at the concentrations of 0-10 μM for 24 h, stained with Giemsa solution, and photographed under a microscope.

Tube-formation assay
The tumor cell formation of capillary structure in vitro was tested, as we previously described 41 . In brief, viable SPC-A-1 cells were pre-treated with lycorine at the concentrations of 0-5 μM for 24 h and transferred to each well of a 48-well plate containing 0.15 ml Matrigel matrix. After incubation at 37°C, 5% CO 2 for 16 h, the tubes were stained with Wright-Giemsa solution and photographed by OLYMPUS FSX-100 microscope.

Matrigel plug assay
The Matrigel plug assay was performed, as we described before 42 . In brief, 0.5 ml of ice-cold Matrigel was mixed with 50 μl (2 × 10 6 ) of either SPC-A-1 or A549 cells with or without 10 μg of lycorine, and subcutaneously injected into the midventral abdominal region of 6-8-week-old nude mice (n = 7). After 14 days, the plugs in the mice were excised, sectioned, stained by hematoxylin-eosin (H&E) solution, and photographed by an OLYMPUS FSX-100 microscope.

Tumor xenograft mouse model and cancer metastasis model
Tumor xenograft assay was conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Soochow University. In brief, 1 × 10 7 human lung cancer SPC-A-1 cells were subcutaneously injected in 6-week-old female nude mice, and the mice were randomly divided into two groups, then were intraperitoneally injected with either lycorine in PBS (10 mg/kg) or PBS as a control daily for 28 days. Tumor volume was measured and calculated according to the formula: tumor volume = 0.5 × length × width 2 .
The lung cancer cell metastasis model was conducted by the approved protocols of the Institutional Animal Care and Use Committee (IACUC) of Soochow University. A549 cells (3 × 10 6 ) were injected into 6-week-old female nude mice via the lateral tail vein. After 24 h, either lycorine in PBS (10 mg/kg/day) or vehicle PBS was intraperitoneally injected once a day. The body weight of the mice was recorded every other day. After lycorine treatment for 34 days, the mice were sacrificed after anesthetized, and their lungs were collected and analyzed by H&E staining.

RT-PCR and quantitative real-time PCR
The total RNA was extracted from A549 and SPC-A-1 cells incubated with lycorine. The RT-PCR and quantitative real-time PCR were conducted as we described before 43 .

Western blotting
Proteins from SPC-A-1 and A549 cells were extracted using the MPER Mammalian Protein Extraction Kit, and the western blotting was performed as previously described 40 . After the proteins were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with Tris-glycine running buffer and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk and incubated with primary antibodies at 4°C overnight, followed by incubation with HRP-coupled secondary antibody for 1 h at room temperature and the enhanced chemiluminescence detection reagents, and exposed to X-ray film.

Co-immunoprecipitation (Co-IP)
Co-IP was carried out as previously described 40 . About 400 μl of the A549 cell lysates were incubated with primary monoclonal antibody (1:500) or mouse normal IgG as a control at 4°C for 4 h, then further incubation with 20 μl of prewashed protein A/G agarose beads at 4°C overnight with rotation. The immune complexes were released from the beads in SDS-loading buffer. The proteins were detected by western blotting as mentioned above.

Statistical analysis
All results represent the mean ± SD. Differences between the groups were assessed by one-way ANOVA using GraphPad Prism 5. Statistical comparisons were performed using the Student's t test, and the significance of differences was indicated as *P < 0.05 and **P < 0.01, ***P < 0.001.

Results
For this study, we used lung cancer cell lines as a model to explore traditional Chinese medicinal herbs that can raise SAV1 expression levels in cancer cells. We found that a small molecule called lycorine (MW: 287.31) significantly inhibited the growth of several lung cancer cell lines. Lycorine has an IC50 value of 4.75 μM in SPC-A-1 cells and 4.03 μM in A549 cells, respectively. Immortalized normal human bronchial epithelial (HBE) cells were less sensitive to lycorine, and we observed an IC50 value of 15.02 μM in the cells (Supplementary Fig. S1a-c). Lycorine arrested lung cancer cell cycle in the G2/M phase ( Supplementary Fig. S1d, e). In addition, lycorine significantly induced lung cancer cell apoptosis (Supplementary Fig. S1f). We conducted a series of experiments to study the role and mechanism of lycorine as an antilung cancer agent.

Lycorine effectively inhibits lung cancer tumor growth and metastasis in xenograft mice
We investigated the anti-lung cancer effects of lycorine in vivo using three mouse models: (1) human lung cancer cell xenograft, (2) cancer metastasis, and (3) Matrigel plug assay. In the xenograft mouse model, nude mice (n = 7 each group) were subcutaneously injected with human lung cancer SPC-A-1 cells, then administered with daily intraperitoneal injections of lycorine at 10 mg/kg body weight or PBS as a control for 28 days. The results showed that tumor volume notably diminished (Fig. 1a), and tumor weight reduced by more than three-fold in the lycorine treatment group compared with the control group (Fig. 1b, c). Mouse body weight was not significantly different between the two groups (Supplementary Fig. S2a), In the tumor metastasis mouse model, nude mice (n = 7 each group) received intravenous tail injections of A549 lung cancer cells. Mice were then intraperitoneally injected daily with lycorine at 10 mg/kg or PBS as a control for 34 consecutive days. After this time, we examined tumors in the lung tissues. H&E staining showed significantly less lung metastasis in the lycorinetreated group than the PBS control group (Fig. 1d). There Fig. 1 Lycorine effectively inhibits lung cancer tumor growth, vasculogenic mimicry, and metastasis in xenograft mice. The nude mice (n = 7 each group) were subcutaneously injected with 1 × 10 7 human lung cancer SPC-A-1 cells, and then intraperitoneally injected with lycorine at the dose of 10 mg/kg/day or vehicle control daily. The tumor size (a) were recorded for 28 days. The excised tumors were imaged (b) and weighted (c). In the another cancer metastasis model, the nude mice (n = 7 each group) were daily treated with lycorine at the dose of 10 mg/kg/day or vehicle control daily, and the lung metastasis in either lycorine-treated or control mice were accessed by lung tissue H&E staining (d, ×200) arrowheads indicate lung cancer nodules in the lung tissues. LH + represents with lycorine treatment, LH− represents vehicle control. In the in vivo Matrigel plug assay model, lycorine effectively abrogated the formation of tumor blood vessels by SPC-A-1 (e) and A549 cells (f). The in vitro tumor cell tubeformation assay showed that lycorine at the concentration of 5 μM near completely inhibited SPC-A-1 cell formation of capillary like structure (tube) (g, h). Data represent the mean ± SD of triplicates. *P < 0.05. **P < 0.01. ***P < 0.001.
were no significant differences between the lycorinetreated and PBS control groups for mouse body weight or organ coefficients (Supplementary Fig. S2b, c).
Lung cancer has robust vasculogenic mimicry to facilitate tumor growth and metastasis 44,45 . Therefore, we next investigated whether lycorine could inhibit tumor vasculogenic mimicry in mice. Using a Matrigel plug assay in nude mice (a classic method to assess both angiogenesis and vasculogenic mimicry in vivo) with A549 and SPC-A-1 lung cancer cells, we found that after 14 days of treatment, lycorine effectively abrogated the formation of tumor blood vessels in the mice (Fig. 1e, f). Lycorine did not affect mouse body weight (Supplementary Fig. S2d). Because tumor cell-mediated vasculogenic mimicry plays a crucial role in cancer dissemination [46][47][48] , we used an in vitro tube-formation assay to examine whether lycorine inhibited lung cancer cell-mediated tube formation. In this experiment, we observed that the tube-forming ability of lung cancer SPC-A-1 cells was markedly reduced by 5 μM lycorine treatment (Fig. 1g, h), suggesting that lycorine is a potent inhibitor of tumor vasculogenic mimicry. Collectively, these data indicate that lycorine markedly inhibited lung tumor growth, vasculogenic mimicry, and metastasis with low toxicity.
Lycorine activates the tumor-suppressive Hippo signal pathway by increasing SAV1 levels in lung cancer cells Our preliminary experiments of gene expression profile using reverse transcription polymerase chain reaction (RT-PCR) and western blotting showed that lycorine promoted SAV1 expression in the tumors of tumorbearing mice. Because no effects of lycorine on the Hippo pathway in cancer is reported and the SAV1-targeted therapeutics against cancer is lacking, we studied the mechanisms of lycorine's anti-lung cancer effect, focusing on the Hippo pathway. We observed that lycorine treatment markedly increased levels of SAV1 and activated MST1 in the Hippo pathway, but significantly decreased the level of oncogenic YAP in lung cancer tissues (Figs. 2-4). Immunohistochemical (IHC) staining showed that SAV1 expression levels were significantly higher in the tumor samples from the lycorine-treated xenograft mice compared with those of the PBS controls (Fig. 2a, b). Tissue immunofluorescent (IF) staining confirmed the higher SAV1 protein levels after lycorine treatment (Fig. 2c), suggesting that lycorine increased SAV1 levels in the tumors from the xenograft mice. In addition, IHC staining showed that levels of the oncogenic YAP protein were significantly lower in tumor tissues from lycorine-treated mice than in controls (Fig. 2d, e). To further test these findings, we used two different methods to measure SAV1 and YAP expression in the tumor samples. Western blotting indicated that SAV1 protein levels were 5.7 times higher in tissues from the lycorine-treated mice, and YAP protein levels were 4.2 times lower in the tumor tissues from the lycorine-treated mice compared with controls ( Fig. 2f, g, h); whereas RT-PCR (Fig. 2i) and quantitative PCR (qPCR) (Supplementary Fig. S2e) found no difference in mRNA levels between these two groups. These data indicate that lycorine regulates expression of SAV1 and YAP post transcriptionally.
Next, we studied the mechanisms of lycorine-induced increases of tumor-suppressive SAV1 and decreases of oncogenic YAP in the Hippo pathway using a lung cancer cell model. Western blotting showed that lycorine treatment not only markedly increased SAV1 protein levels ( Fig. 3a-d; Supplementary Fig. S3a, b) but also dramatically increased phosphorylation of its interacting protein MST1, indicating MST1 protein activation, in SPC-A-1 (Fig. 3a, b) and A549 lung cancer cells (Fig. 3c, d) in a concentration-dependent manner. Lycorine treatment did not change total MST1 protein levels or SAV1 mRNA levels in these lung cancer cells ( Supplementary Fig. S3c, d). These data suggest that the lycorine-induced increase in SAV1 protein activates tumor-suppressive MST1 in lung cancer cells.
The mechanisms of the effect of lycorine on SAV1 protein levels in lung cancer cells is unclear. We hypothesized that lycorine may prevent SAV1 protein from degradation. We found that treating SPC-A-1 cells with MG132, a ubiquitination-proteasome inhibitor, increased SAV1 levels and that lycorine treatment further increased SAV1 protein levels in cancer cells in the presence of MG132 (Fig. 3e, f). Furthermore, co-immunoprecipitation (Co-IP) and western blotting showed that levels of ubiquitinated SAV1 protein were markedly lower in lycorinetreated SPC-A-1 cells than in cells without lycorine treatment (Fig. 3g, h), suggesting that lycorine-induced SAV1 protein accumulation in lung cancer cells is mediated by inhibiting the ubiquitination-proteasome system that degrades SAV1.

Lycorine promotes the degradation of oncogenic YAP and decreases expression of tumorigenic genes and cancer cell invasion
Oncogenic YAP, a key downstream protein in the Hippo pathway, promotes transcription of many tumorigenic genes and activates multiple signaling pathways [49][50][51] . We tested whether lycorine treatment affected YAP expression. Western blotting results showed that total YAP protein levels were notably reduced, but phosphorylated YAP protein (p-YAP) levels increased in a lycorine concentration-dependent manner in both SPC-A-1 (Fig. 4a) and A549 lung cancer cells (Fig. 4b). In addition, western blotting and Co-IP results showed that lycorine treatment increased YAP protein ubiquitination and degradation in the cancer cells via the ubiquitin-proteasome system (Fig.  4c, d, e). Furthermore, dynamic analysis of YAP degradation indicated that lycorine shortened the half-life of YAP in both SPC-A-1 (Fig. 4f, g) and A549 cells (Fig. 4h, i). Intracellular YAP protein localization analysis using nuclear protein extracts and western blotting showed that lycorine profoundly reduced the levels of nuclear YAP protein (Fig. 4j, k). These data indicate that lycorine reduced YAP protein levels by triggering phosphorylation and degradation of YAP, thereby reducing levels of oncogenic YAP in Fig. 2 Lycorine elevates the tumor suppressor SAV1 levels and reduces the oncogenic YAP amounts in the tumors from xenograft mice. The levels of SAV1 and YAP proteins in the tumors from xenograft mice were detected by IHC and IF, respectively. Lycorine raised SAV1 levels (IHC: a, b, ×200; IF: c, ×1000), while YAP amounts were decreased by lycorine treatment (IHC: d, e). The amounts of SAV1, YAP, and β-actin in the tumor tissues from xenograft mice were detected by western blotting (f-h) and RT-PCR (i), respectively. LH + represents with lycorine treatment, LH− represents vehicle control. Data represent the mean ± SD of triplicates. *P < 0.05. **P < 0.01. ***P < 0.001. the cell nucleus. Nuclear YAP promotes the transcription of various oncogenic genes 10,52-54 . Therefore, we investigated the effect of lycorine treatment on the expression of YAP-activated transcription of oncogenic genes. qPCR showed that lycorine treatment significantly decreased the mRNA levels of two YAP-regulated tumor angiogenic genes, Fig. 3 Lycorine significantly elevates SAV1 protein levels through degradation of the protein via ubiquitination-lysosome system. After A549 and SPC-A-1 cells were treated with lycorine (LH) at the concentrations of 0-10 μM for 72 h, the protein levels of SAV1, total MST1 (T-MST1), and phosphorylated MST1 (p-MST1) were determined by Western blotting. SAV1, p-MST1 protein levels were increased in lung cancer SPC- A-1 (a, b) and A549 (c, d) cells by lycorine in a dosage-dependent manner. Treatment of the lung cancer cells with the ubiquitin inhibitor MG132 significantly increased SAV1 protein levels, combination of lycorine with MG132 further increased SAV1 protein levels (e, f). Co-IP and western blotting analysis displayed that lycorine significantly decreased SAV1 protein ubiquitination (g, h). Data are shown as mean ± SD of three independent replicates. Fig. 4 Lycorine promoted the phosphorylation and ubiquitin-mediated degradation of oncogenic YAP in lung cancer cells. Western blotting showed the at lycorine notably decreased total YAP (T-YAP) protein levels, but markedly increased the phosphorylation of the protein (p-YAP) in a dose-dependent manner in lung cancer SPC-A1 (a) and A549 cells (b). Ubiquitination inhibitor MG132 significantly reversed the lycorinemediated YAP protein degradation in SPC-A-1 (c) and A549 cells (d). Lycorine significantly increased YAP protein ubiquitination in SPC-A-1 cells (e). The half-life of YAP was been shorten with lycorine treatment in the presence of the protein synthesis inhibitor CHX in SPC-A-1 (f, g) and A549 cells (h, i). The intracellular protein distribution analysis showed that lycorine decreased total and plasma YAP levels, and markedly reduced nuclear YAP levels (J, k). Data are shown as mean ± SD of three independent replicates.
Sema4D and Ang2, in SPC-A-1 (Fig. 5a, c) and A549 cells (Fig. 5b, d). Lycorine treatment also decreased expression of S-phase kinase-associated protein 2 (SKP2), a YAP-controlled E3 ligase that is important for regulating ubiquitination and cell cycle (Fig. 5e, f). In addition, protein levels of Sema4D and SKP2 were significantly lower after  lycorine treatment (Supplementary Fig. 4a-f). Because there is cross talk between the Hippo signaling pathway and the AKT and NF-κB signal pathways 43,55 , we investigated the effect of lycorine on these two signaling pathways. Our results showed that lycorine treatment significantly inhibited the phosphorylation of AKT and NF-κB in both SPC-A-1 (Fig. 5g, h) and A549 cells (Fig. 5i, j). Collectively, these data suggest that lycorine activates the tumor-suppressive Hippo pathway and inhibits the transcription of oncogenic genes and tumorigenic AKT and NF-κB signaling pathways.
Malignant tumor cells are highly motile and easily metastasize to multiple organs 56,57 . Given our finding that lycorine inhibited lung cancer cell angiogenesis and dissemination ( Fig. 1d-h), we next studied the effect of lycorine on lung cancer cell motility. A wound-healing assay showed that lycorine significantly decreased A549 cell migration (Fig. 6a, b) and markedly inhibited the invasion of both SPC-A-1 and A549 cells in a concentration-dependent manner (Fig. 6c-e). Furthermore, the invasion of SPC-A-1 cells was significantly reduced by YAP shRNA (Fig. 6f-i), and combining YAP shRNA with lycorine treatment further reduced YAP expression (Fig. 6f-i) and decreased the invasion capability of the cancer cells (Fig. 6j, k), suggesting that lycorine treatment was synergistic with a YAP inhibitor.
Together, these data indicate that lycorine treatment suppresses tumorigenicity, angiogenesis, invasion, and metastasis of lung cancer cells primarily by increasing Fig. 7 Schematic representation of correction of the SAV1 deficiency in lung cancer cells by lycorine as a new strategy for effective cancer therapy. Lycorine increases SAV1 levels through inhibiting SAV1 protein degradation via ubiquitination-proteasome system, induced MST1 phosphorylation, consequently reduces oncogenic YAP amounts by promoting the degradation of YAP protein and inhibits the transcription of YAP downstream oncogenic genes in lung cancer. In addition, elevation of SAV1 levels by lycorine inhibits AKT signaling and reduction of YAP suppresses NF-κB signaling, resulting in effective inhibition of lung cancer cell proliferation, motility, vasculogenic mimicry, and metastasis. SAV1 levels. We hypothesize that SAV1 increase is due to decreased SAV1 protein ubiquitination and degradation, which activates downstream MST1. MST1 activation promotes the ubiquitination and degradation of oncogenic YAP and inhibits transcription of the YAP-target genes. As a result, AKT and NF-κB signal pathways are suppressed, which effectively inhibits proliferation, motility, vasculogenic mimicry, and metastasis of lung cancer cells, resulting in strong anticancer effects (Fig. 7).

Discussion
The tumor suppressor SAV1 expression is downregulated in various malignant tumors, including lung cancer 58,59 , and results in robust tumorigenicity 12,18 . In the Hippo pathway, SAV1 activates MST1/2-LAST1/2, a downstream complex that stimulates phosphorylation, ubiquitination, and degradation of oncogenic YAP 34,60 . The deficiency of SAV1 in cancer cells causes the MST1/2 proteins to be inactive. Therefore, upregulation of SAV1 expression and subsequent activation of the MST1/2-LAST1/2 complex in cancer cells are important treatment strategies to be explored. In this study, we observed that lycorine treatment increased SAV1 levels, which activated MST1 and promoted YAP degradation. We hypothesize that YAP degradation then downregulated the YAP downstream tumorigenic genes and suppressed AKT and NF-κB signaling pathways in lung cancer cells. Increasing SAV1 levels to activate the tumor-suppressive Hippo pathway and inhibition of these oncogenic pathways resulted in robust inhibition of tumorigenicity, invasion, and metastasis of lung cancer cells, suggesting that use of exogenous agents to increase SAV1 levels in cancer presents a new strategy for discovering novel anticancer drugs.
SAV1 has long been regarded as an adaptor protein in the Hippo pathway, but has not been well-studied, although emerging evidence indicates that SAV1 is essential for tumor suppression 16 . Our study found that SAV1 is an excellent target for cancer therapy, and this approach has several advantages. First, the tumor suppressor SAV1 is downregulated in a variety of malignant tumors 58,59 ; thus, the strategy of increasing SAV1 expression in malignant tumors may be applied broadly to prevent and treat various types of cancer. Second, SAV1 induces the protein kinase-phosphorylation cascade (MST1/2-LAST1/2-YAP/TAZ) in the Hippo pathway and simultaneously inhibits several other tumorigenic signal pathways, such as the AKT and NF-κB signal pathways in cancer cells, therefore amplifying SAV1-mediated tumor suppression. Third, SAV1 overexpression inhibits cancer cell proliferation and motility and induces tumor cell apoptosis 13,61 without obvious side effects in tumorbearing mice. Thus, SAV1 is an excellent target for cancer therapy.