LINC00669 insulates the JAK/STAT suppressor SOCS1 to promote nasopharyngeal cancer cell proliferation and invasion

All animal experiments were conducted in accordance with the guidelines of Animal Experimentation Ethics Committee of the Third Xiangya Hospital of Central South University. Clinical NPC tumor and para-tumor samples were collected from 16 patients with informed consents being signed by all of the patients.

All the NPC cell lines and normal human nasopharyngeal epithelial cell line NP69 were purchased from the Chinese Academy of Sciences Cell Bank (Shanghai, China). NPC cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco, Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Louis, MO, USA) and 1% antibiotics (100 U/ml penicillin and 100 mg/ml streptomycin sulfates), while NP69 cells were maintained in the RPMI1640 medium containing 10% FBS and 1% penicillin/streptomycin. Cycloheximide (20 mg/ml, Millipore Sigma, #C7698) and MG132 (5 μM, Sigma-Aldrich, #1211877-36-9) were used for studying ubiquitin-proteasome-mediated protein degradation.

Total RNA of NPC tissues and cells was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.), followed by reverse transcription using RevertAidTM H Minus First Strand cDNA Synthesis Kit (Fermentas). The relative mRNA contents were quantified by quantitative PCR (qPCR) using SYBR Green PCR Master Mix (ABI 4309155) in the ABI7900 Realtime-PCR machine. Primers used for qPCR analysis were included in Table 1.

The truncated LINC00669 plasmids were constructed using Phusion Site-Directed Mutagenesis Kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) by following the manual instruction. In vitro loss-of-function studies were performed by pre-designed siRNA from Millipore Sigma using Lipofectamine™ 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). For in vivo xenograft experiment, lentiviral vectors expressing shLINC00669, SOCS1, shSOCS1 and the empty lentiviral vector were purchased from GeneChem (Shanghai, China for the stable infection of CNE-2 cells upon puromycin (2 μg/ml) selection.

Total proteins from NPC tissues and cells were extracted using RIPA buffer (1% NP-40, 0.1% SDS, 50 mM DTT) supplemented with protease inhibitor cocktail containing 2 μg/ml Aprotinin, 2 μg/ml Leupeptin and 1 mM PMSF. An equal amount of total protein from each sample was separated by 8 ~ 15% SDS-PAGE gel and electroblotted onto a polyvinylidene difluoride membrane. Immunoblotting was conducted using antibodies listed in Table 2.

Total proteins from NPC cells cultured in 10-cm culture dishes were used for Co-IP assay. Briefly, primary antibodies against SOCS1 or STAT1 were added in each cell lysate and incubated overnight at 4 °C with rotation. The normal human IgG protein was used as a negative control. Protein A/G agarose beads (Beyotime Biotechnology, #P1012) were then added for affinity binding of primary antibody by a 2 h-incubation at 4 °C with gentle rotation. The unbound antibodies were washed off through the sequential wash with PBS and cell lysis buffer. The agarose beads were resuspended in 20 μl 1x SDS-PAGE loading buffer, and boiled for western blot analysis of the precipitated target proteins.

Time-dependent NPC cell proliferation was characterized by MTT cell proliferation and cytotoxicity detection kit (KeyGEN Biotech, #KGA312) by following the manual instruction.

NPC cell migration was characterized by wound-healing assay. Briefly, a confluent monolayer of NPC cells in 24-well plate was scratched using a sterile 100 μl pipette tip. Cells were then allowed to expand in FBS-reduced growth medium (1%) for another 48 h. The migration was assessed using the scratch ratio by dividing the interval at the starting point (0 h) with the one at the ending point (48 h).

NPC cells were seeded into the Matrigel (BD Biosciences, San Jose, CA)-coated upper chamber of 8.0 μm pore size Transwell apparatus (Corning, NY, USA) with serum-free medium. Growth medium supplemented with 10% FBS was added to the lower chamber as a chemoattractant. The cells were allowed to invade for 48 h at 37 °C under 5% CO₂. Cells invaded to the lower surface of filter were then fixed in 70% ethanol for 30 min followed with staining by 0.1% crystal violet for 10 min at room temperature.

Cellular LINC00669 localization was characterized using the fluorescence in situ hybridization kit from Ribo Bio (Guangzhou, China) according to the manufacturer’s instruction.

NPC cell apoptosis were determined by flow cytometry using Annexin V-PE/7-AAD Apoptosis Kit (Abnova, #KA3809) according to the manual instruction.

Tissues for IHC were fixed in 10% buffered formalin for 24 h and embedded in paraffin. The deparaffinized and rehydrated sections were blocked for endogenous peroxidase by incubation in 3% hydrogen peroxide followed with antigen retrieval in the boiling citrate buffer (10 mM, pH 6.0) for 10 min. The sections were then blocked with normal goat serum (1:10) and subject to incubation with anti-Ki67 monoclonal antibody (1:100, Dako, Glostrup, Denmark) or anti-Caspase 3 antibody (1:100, Abcam, #ab32351) overnight at 4 °C. Thereafter, the PBS cleaned sides were incubated with the biotinylated secondary antibody at 37 °C for 30 min, and subsequently incubated with a 1:200 streptavidin-biotin-peroxidase complex (Sigma, St. Louis). Reactive products were visualized with 3,3′-diaminobenzidene (DAB) as the chromogen, and nuclei were counter-stained with hematoxylin.

NPC cells were fixed with 4% paraformaldehyde and permeabilized by 0.2% Triton X-100. After 1 h blocking with 1% bovine serum albumin at room temperature, the cells were incubated with anti-STAT1 (1:100) antibody overnight at 4 °C. Next day, the cells were washed off the unbound primary antibody and stained with donkey anti-rabbit IgG Alexa Fluor 488 secondary antibody (ThermoFisher, #A32790) for 1 h at room temperature.

Male BALB/c nude mice (8-week-old, n = 6 mice/group) from Beijing HFK Bioscience Co. Ltd. (Beijing, China) were maintained under pathogen-free conditions. After subcutaneously injected with 10⁶ control or viral shRNA infected CNE-2 cells, the mice were monitored for tumor growth from 1 week after injection. Tumor size was calculated based on the formula of Volume = a x b² × 0.52 (a, long diameter; b, short diameter).

Biotin-labeled full-length LINC00669 and antisense LINC00669 were synthesized in vitro using Biotin RNA Labeling Mix (Roche) and the Riboprobe Systems with T7 RNA polymerase (Promega). After DNase I treatment, the RNA probes were purified with RNeasy Mini Kit (QIAGEN). For each pulldown assay, 30 μg of biotin-labeled RNA probes were incubated with CNE-2 cell lysates at room temperature for 4 h followed by adding pulling down the binding protein partner with streptavidin magnetic beads (TermoFisher, USA) at 4 °C overnight. The proteins were then separated by electrophoresis and visualized with sliver staining. The unique bands pulled down by sense LINC00669 were subject to mass spectrometry and retrieved in human proteomic library.

RIP assay was conducted in wildtype and LINC00669-depleted NPC cells using Magna RIP RNA-binding protein immunoprecipitation Kit (Millipore, MA) according to manufacturer’s instruction. Antibodies against the normal human IgG and SNRNP70 were used as negative and positive controls for RIP, respectively. The relative enrichment of LINC00669 was normalized to the amount of the enriched U1snRNA.

The common signatures in NPC were searched by overlapping the upregulated genes (p ≤ 0.05, log2 fold change ≥2) in databases of GSE12452 and GSE53819 followed by the overall survival correlation analysis using the Signature-based statistics tool in GEPIA 2 (http://​gepia2.​cancer-pku.​cn/​#index). In silico prediction of LINC00669-SOCS1 interaction was performed using online tools from catRAPID (http://​s.​tartaglialab.​com/​page/​catrapid_​group).

Data are expressed as the mean ± standard deviation of three independent experiments. Two-tailed Student’s t-test was performed for detecting difference between two groups, while One-Way ANOV test was conducted for comparing two-group difference among the multiple groups using GraphPad Prism 6 software. P < 0.05 was considered as significant.

Emerging roles of lncRNA in cancer development and therapeutic opportunities have aroused wide research interest and attention [13]. To explore novel lncRNAs that play key roles in NPC, we retrieved and interrogated the expression profiles of lncRNAs in two independent NPC mRNA expression profiling databases of GSE12452 [14] and GSE53819 [15] to have identified 153 commonly upregulated signatures (Fig. 1a). None of the other lncRNAs showed significant difference in the survival curve analysis except for three classical lncRNAs of LINC00669, AFAP1-AS1 and CT75, whose upregulation are negatively correlated to the overall survival of NPC patients (Fig. 1b, c). Since LINC0069 was a novel lncRNA whose function was completely unknown, we made it the focus in our current study. Consistently, LINC00669 was confirmed by qPCR to be significantly upregulated in the patient-derived NPC tumors (Fig. 1d) and a panel of the general NPC cell lines (Fig. 1e) in comparison to their corresponding controls. Due to the highest expression of LINC00669 in CNE-2 and 5-8F among the tested cell lines, they were chosen for the subsequent functional studies. FISH assay showed that LINC00669 is dominant in the cytoplasm in both CNE-2 and 5-8F cells (Fig. 1f), indicating the potential roles involved in translational or signaling regulations.

Next, we conducted gain- and loss- of function assays to explore the functions of LINC00669 in NPC. Consistent in both CNE-2 and 5-8F cell lines, forced expression of LINC00669 dramatically promoted cell proliferation (Fig. 2a), invasion (Fig. 2b) and migration (Fig. 2c), while its depletion by small interfering RNA (siRNA) exhibited the opposite effects. Meanwhile, LINC00669 is essential for the survival of NPC cells as indicated by the strikingly induced spontaneous apoptosis upon loss of LINC00669 (Fig. 2d). Together, these data indicated that LINC00669 profoundly endows carcinogenic properties of NPC cells.

Inspired by the cytoplasmic localization of LINC00669 (Fig. 1e), we speculated that it might participate in the translational or signaling regulations. To identify its target proteins, we incubated CNE-2 cell lysates with a biotin-labeled LINC00669, and proceeded with electrophoresis followed by silver staining. Mass spectrum analysis of the unique protein bands detected only in the sense but not anti-sense LINC00669-incubated cell lysates by silver staining indicated SOCS1 as a potential LINC00669-targeting protein (Fig. 3a). The direct interaction between LINC00669 and SOCS1 was further confirmed by western blot on LINC00669 pulldown cell lysates using an antibody against SOCS1 (Fig. 3b). Vice versa, RNA-immunoprecipitation (RIP) assay using SOCS1 antibody followed by qPCR analysis demonstrated that SOCS1 specifically precipitated LINC00669 in both CNE-2 and 5-8F cell lysates, and the enrichment of LINC00669 was largely decreased upon siRNA-mediated depletion of LINC00669 (Fig. 3c).

Next, we conducted an in-silico prediction through catRAPID [16] to map the domain in LINC00669 that mediates its interaction with SOCS1. The analytical results of both interaction profile (Fig. 3d) and interaction matrix (Fig. 3e) suggested that RNA region of 190–283 nucleotides (nt) and the protein domain ranging from 76 to 187amino acids in SOCS1 were mostly likely to mediate the mutual binding (Table 3). Accordingly, we made a series of the truncated LINC00669 constructs flanking the site of 190–283 nt for testing their binding potentials with SOCS1 (Fig. 3f). Consistent with the predication by catRAPID, only the constructs encompassing region of 190–283 nt were able to pulldown SOCS1 (Fig. 3g). Taken together, these data indicated SOCS1 is an immediate target of LINC00669 for executing its oncogenic functions in NPC cells.

The SOCS family proteins are the negative-feedback inhibitors of the JAK/STAT signaling pathway [17]. As one of the most potent family members, SOCS1 can inhibit the JAK/STAT pathway either as a ubiquitin ligase by recruiting Cullin5 to signaling components [18] or inhibit directly the kinase activity of JAK [19]. Therefore, we determined the influences of gain- and loss- of SOCS1 (Fig. 4a) on the expression of the JAK/STAT pathway component proteins, respectively. As the western blot results displayed, STAT1 protein was dramatically decreased in the SOCS1-overexpressing NPC cells, while the opposite was true when SOCS1 was silenced (Fig. 4b). Instead, the amounts of other JAK/STAT pathway component proteins including JAK2, JAK3 and STAT3 were barely affected by SOCS1. Mechanistically, SOCS1 specifically interacts with STAT1 but not other proteins in the pathway like JAK2 (Supplementary Figure 1A). The proteasome inhibitor MG132 significantly postponed degradation of both JAK2 and STAT1 in the presence of cycloheximide, indicating they are subject to a control by the ubiquitin-proteasome system. Because JAK2 did not directly interact with SOCS1 (Supplementary Figure 1A), we restricted the mechanistic investigations to STAT1 in the following studies. As expected, STAT1 was undergone a ubiquitination modification in NPC cells as evidenced by the co-immunoprecipitated ubiquitin by STAT1 (Supplementary Figure 1B). Forced expression of SOCS1 not only drastically increased total amount of the ubiquitin-modified STAT1 (Supplementary Figure 1C), but also accelerated STAT1 protein degradation (Supplementary Figure 1D) in both of the tested NPC cell lines.

The JAK/STAT signaling contributes to multiple tumor malignancies by promoting the survival, self-renewal, and metastasis of cancer stem cells [20]. We wondered how does the SOCS1/STAT1 regulatory network affect the NPC cell phenotypes. Obviously, SOCS1-depleted NPC cells exhibited a series of the cancerous propensities including enhanced proliferation (Fig. 4c), invasion (Fig. 4d), migration (Fig. 4e) coupled with decreased spontaneous apoptosis (Fig. 4f) in comparison to the control cells. Of note, simultaneous silencing of STAT1 completely abolished all the stimulative effects in the SOCS1-depleted NPC cells. Taken together, these data indicate SOCS1 plays a role of tumor suppressor in NPC cells by directing the ubiquitin-proteasome-mediated degradation of STAT1 in the JAK/STAT signaling pathway.

Aberrant activation of the JAK/STAT pathway has been found in many tumors [10]. Indeed, most of the component genes of this signaling pathway were upregulated in human NPC tumors at the transcription level (Fig. 5a). Unexpectedly, this was concurrent with upregulation of the negative feedback regulator SOCS1. We further examined the expression of SOCS1 and STAT1 proteins in NPC tumors and para-carcinoma tissues by IHC. Although SOCS1 levels were comparative between the normal and cancerous tissues, it was NPC tumors instead of para-carcinoma tissues that exhibited a strong immunostaining signal of STAT1 (Fig. 5b), indicating the interrupted Inhibition of SOCS1 on STAT1 in NPC tumors. Given the fact that SOCS1 strongly interacts with LINC00669, which is highly expressed in NPC tumors, we investigated how changes of LINC00669 and then its interaction with SOCS1 would affect SOCS1/STAT1 network. Obviously, transcription of SOCS1 and STAT1 did not vary with the expression of LINC00669 (Fig. 5c). Nor did SOCS1 protein change accordingly (Fig. 5d). Instead, STAT1 protein was increased when LINC00669 was overexpressed and was decreased when LINC00669 was depleted (Fig. 5d). Concomitantly, loss of LINC00669 also led to an increase in ubiquitination of STAT1 (Fig. 5e), which could be reversed upon forced expression of LINC00669 (Fig. 5f).

As SOCS1 destabilizes STAT1 through the ubiquitin-proteasomal protein degradation, we postulated that changes in LINC00669 expression might also lead to the altered STAT1 protein stability in view of the apparent change in the ubiquitination modification it brought to STAT1 (Fig. 5e, f). Indeed, depletion of LINC00669 largely shortened the half-life of STAT1 as evidence by the accelerated degradation of STAT1 protein in both of the cycloheximide-treated CNE-2 and 5-8F cells (Fig. 6a). Interestingly, not only the total amount of STAT1, but also the phosphorylated STAT1 was greatly increased as a consequence of overexpressing LINC00669 (Fig. 6b), which expectedly led to a nuclear translocation of STAT1 in NPC cells (Fig. 6c).

Now that the establishment of LINC00669/SOCS1/STAT1regulatory cascade, we wondered whether this could be the mechanism that underlies the profound influences of LINC00669 and SOCS1 on the featured NPC cell phenotypes. As expected, silencing LINC00669 suppressed NPC cell proliferation (Fig. 7a), invasion (Fig. 7b) and migration (Fig. 7c), and induced spontaneous apoptosis (Fig. 7d). Simultaneous depletion of SOCS1 completely abolished all the detrimental effects observed in the LINC00669-silenced NPC cells.

Finally, we accessed the role of LINC00669/SOCS1/STAT1 regulatory network in NPC carcinogenesis in vivo. CNE-2 cells with stable knockdown of LINC00669 or concomitant depletion of LINC00669 and SOCS1 were subcutaneously injected into nude mice, respectively. Comparing to the mice injected with control CNE-2 cells, silencing LINC00669 dramatically suppressed xenografted tumor growth (Fig. 8a, b). Instead, tumor growth retardation shown in mice bearing tumors derived from LINC00669-deficient CNE-2 cells was completely reversed in the recipients injected with LINC00669 and SOCS1 double knockdown cells. In accordance, the proliferative cells in xenograft tumors derived from LINC00669 silencing CNE-2 cells were largely diminished compared to the control tumors, while tumors grown from LINC00669 and SOCS1 double knockdown CNE-2 cells were overwhelmed with massive of Ki67 positive cells (Fig. 8c). In contrast, apoptosis as indicated by Caspase 3 immunostaining appeared in an opposite pattern to that featured in proliferation (Fig. 8c).

As a whole, our study revealed a novel tumor-promoting mechanism involving LINC00669/SOCS1/STAT1 regulatory network in NPC (Fig. 8d). Briefly, LINC00669 is aberrantly increased in the cytoplasm of NPC cells, where it binds to and prevents SOCS1 from imposing ubiquitination modification on the JAK/STAT signaling pathway component transcription factor STAT1. As a result, total and phosphorylated STAT1 are accumulated by escaping from proteasomal degradation. Phosphorylation of STAT1 facilitates its nuclear translocation and initiation of the transcriptional program for the enhanced proliferation and invasion.