Suppression of OSCC malignancy by oral glands derived‐PIP identified by iTRAQ combined with 2D LC‐MS/MS

Abstract Oral squamous cell carcinoma (OSCC) is the most common malignancy in head and neck cancer and a global cause of cancer‐related death. Due to the poor survival rates associated with OSCC, there is a growing need to develop novel technologies and predictive biomarkers to improve disease diagnosis. The identification of new cellular targets in OSCC tumors will benefit such developments. In this study, isobaric tags for relative and absolute quantitation (iTRAQ)‐based proteomics analysis combined with 2‐dimensional liquid chromatography and tandem mass spectrometry (2D LC‐MS/MS) were used to identify differentially expressed proteins (DEPs) between tumor and normal tissues. Of the DEPs detected, the most significantly downregulated protein in OSCC tissue was prolactin‐inducible protein (PIP). Clonogenic and 3‐(4,5‐dimethyl‐2‐thiazolyl)‐2,5‐diphenyl‐2H‐tetrazolium bromide (MTT) experiments showed that the proliferation capacity of OSCC cells overexpressing PIP decreased due to cell cycle arrest at the G0/G1 checkpoint. Wound‐healing and transwell assay further showed that PIP overexpression also reduced the migration and invasion of OSCC cells. Immunohistochemistry (IHC) was used to analyze the expression in OSCC, indicating that PIP may be secreted by glandular cells and have an inhibitory effect on OSCC cells to produce. In western blot analysis, silencing studies confirmed that PIP mediates these effects through the AKT/mitogen‐activated protein kinase (MAPK) signaling axis in OSCC cells. Taken together, this study reveals PIP as a key mediator of OSCC cell growth, migration, and invasion through its effect on AKT/MAPK signaling.

prognosis of OSCC is improved during early disease stages, particularly in well-differentiated tumors that have not metastasized. However, the majority of patients with OSCC are diagnosed during the late stages of the disease. In this regard, new biomarkers are urgently required for early OSCC detection.
Isobaric tags for relative and absolute quantitation (iTRAQ) combined with 2-dimensional liquid chromatography and tandem mass spectrometry (2D LC-MS/MS) analysis have revealed promising candidate biomarkers for breast cancer, lung cancer, stomach cancer, colon cancer, and esophageal cancer. (Creaney, Dick, Leon, & Robinson, 2017;Sun et al., 2018;Xiao et al., 2015;Yang, Chen, Chan, Li, & Zhang, 2017;Zhang et al., 2017). This technique permits the identification of differentially expressed proteins (DEPs) between tumor and normal tissues . The traditional method for identifying DEPs is 2-dimensional gel electrophoresis, but this fails to reliably identify DEPs of high molecular weight, or those are greatly acidic or basic or dwell in the cell membrane (Jiang et al., 2017).
Using the iTRAQ approach, an 8-plex set of amine-reactive isobaric tags are used to derivatize peptides at the N-terminus and the lysine side chains for labeling all peptides (Chai et al., 2016).
iTRAQ frequently yields highly quantitative data and has the advantage of high throughput achieved by sample multiplexing, high confidence of identification, and highly sensitive quantification of protein expression (Narayana, Tomatis, Wang, Kvaskoff, & Meunier, 2015;Negroni et al., 2014).
In this study, iTRAQ labeling was coupled with 2D LC-MS/MS to detect the DEPs between OSCC and normal oral tissue (Xiao et al., 2015). Certain DEPs for binding were further selected and validated for their molecular function by Gene Ontology (GO) analyses. The value of ten  (Tomar, Sooch, Raj, Singh, & Yadav, 2013). We found that PIP could regulate the proliferation, migration, and invasion of OSCC through AKT/mitogen-activated protein kinase (MAPK) signaling. Using immunohistochemistry (IHC), we discovered that PIP is expressed in both gland and stromal cells, and it has been reported that PIP is related to secretory function. We therefore predict that PIP is secreted by acinar cells into the stroma to act on OSCC cells.

| Liquid chromatography and tandem mass spectrometry analyses
Samples were analyzed for 60 min using a Q-Exactive mass spectrometer (Thermo Fisher Scientific, CA). The detection mode was the positive ion at a parent ion scan range of 350-1800 m/z and a primary mass spectrometer resolution of 70,000. The level one maximum IT was 50 ms.
The mass-to-charge ratio of the peptides and polypeptide fragments were as follows: 10 MS spectra were acquired after each full scan. An HCD MS2 activation was used with an isolation window of 2 m/z. Secondary mass spectra were resolved at a rate was 17,500 (microscans = 1). The secondary maximum IT was 45 ms, and the normalized collision energy was 30 eV.

| Bioinformatics analysis
Blast2GO annotation on the target protein sets was roughly divided into blast, mapping, annotation, and annotation augmentation. First, the NCBI BLAST + target protein alignment tool and appropriate protein sequence databases were used to meet the E-value < = 1e−3 value. A total of 10 sequences were selected for subsequent analysis.
Secondly, the Blast2GO Command Line was used to extract GO items WANG ET AL.

| Quantitative real-time PCR
Total RNA was isolated from surgical specimens or OSCC cell lines using a RNeasy Kit (TaKaRa, Dalian, China). cDNA was synthesized according to the manufacturer's instructions. SYBR Premix Ex Taq (TaKaRa) was used for real-time polymerase chain reaction (RT-PCR) at a total reaction volume of 15 μl. The cycling conditions included a holding step at 95°C for 30 s, followed by 40 cycles in two steps: 5 s at 95°C and 30 s at 60°C. A dissociation step was added to verify that each primer pair produced only a single product at 15 s at 95°C, 30 s at 55°C, and 15 s at 95°C. β-Actin was used as an internal control.
Gene expression was quantified using the −ΔΔ 2 C t method.

| Cell culture
Cells were cultured in high-glucose Dulbecco's modified Eagle's Medium (DMEM) medium (HyClone, Logan, UT) supplemented with 10% fetal bovine serum (FBS; HyClone) at 37°C in a 5% CO 2 standard humidified incubator. Cells in the logarithmic phase were used in all experiments.

| Cell viability, cell cycle, and colony formation assay
For viability assays, cells were plated into 96-well plates at a density of 5 × 10 4 cells/ml for 24 hr and transfected. Cell viability was measured through MTT assays. For cell cycle analysis, cells were dissociated through trypsinization, fixed in 70% ethanol and stained with propidium iodide. Cell cycle distribution was analyzed by FlowJo 7.6.5 (Becton Dickinson, Franklin Lakes, NJ). For colony formation assays, 500 cells per well were seeded into six-well plates and cultured for 12 days. Culture medium was replaced every 3 days.
Viable colonies were scored through crystal violet staining. The average number of colonies was calculated and each experiment was repeated on three independent occasions.

| Wound-healing, cell migration, and invasion assays
For wound-healing assays, control, PIP-overexpressed, and PIP silenced OSCC cells were seeded into six-well plates and grown to 100% confluency in complete medium. Linear scratches were made into cell monolayers using a pipette tip and cells were washed three times with phosphate-buffered saline (PBS). Remaining cells were starved overnight in serum-free medium to exclude the effects of proliferation on cell migration. Cell was imaged at 12 and 24 hr on a digital camera (Nikon, Kobe, Japan) and the scratch area was quantified using ImageJ were randomly selected for counting on a light microscope (Nikon).
Each experiment was performed on a minimum of three occasions and the average number of migrated or invaded cells calculated. Sections were counterstained with hematoxylin and dehydrated.

| Protein-protein interaction network
The protein-protein interaction (PPI) network for PIP was retrieved from the Search tool for the retrieval of interacting proteins (STRING) database and reconstructed through the addition of recently identified PIP interacting proteins.

| Statistical analysis
All experiments were performed on a minimum of three occasions. Data are presented as the mean ± SD (standard deviation). Differences in the means of the test groups were compared through an analysis of variance (ANOVA) and t tests using the SPSS version 23.0 (Chicago, IL).
Statistical significance was determined at p < 0.05 and p < 0.01.

| Identification of DEPs between OSCC and normal samples by iTRAQ and LC-MS/MS
To identify potential OSCC biomarkers, we used iTRAQ combined with 2D LC-MS/MS to analyze OSCC samples. Using the ProteinPilot software (AB Sciex, CA), a total of 1251 DEPs were identified, including 893 upregulated and 358 downregulated proteins. DEPs were classified according to GO terms including biological process, molecular function, and cellular compartment. For molecular function, the majority of proteins were involved in binding (50.52%), catalytic activity (23.30%), structural molecule activity (7.27%), molecular function regulator (5.44%), transporter activity (4.14%), and signal transducer activity (2.00%; Figure   1a). Ten proteins related to binding were randomly selected from the downregulated proteins, of which PIP was the most significant (p < 0.05; Figure 1b). We simultaneously selected 29 paired tissues for further verification by qPCR and five of those were validated by western blot analysis (Figure 1c).

| PIP regulates cell proliferation and cell cycle progression
The results from clonogenic assays revealed that the PIP overexpression in OSCC cells inversely correlates with control cells growth. Consistent with this finding, PIP silencing promoted colony numbers growth (p < 0.05; Figure 2a). MTT assays were used to examine the proliferation of PIP overexpressed cells and siRNA-PIP cells versus control SCC15 and SCC25 cells. The cell proliferative capability of PIP overexpressed cells was significantly lower than control cells (p < 0.05; Figure 2b). Assessment of cell cycle status of PIP overexpressed cells revealed a G0/G1 phase arrest (p < 0.05; Figure 2c).

| Effects of PIP on the migration and invasion of OSCC cells
As shown in Figure 3a, PIP overexpression significantly inhibited the migration of SCC15 and SCC25 cells in wound-healing assays as evidenced by the narrowed healing borders. PIP silencing produced the opposite effect (p < 0.05). To confirm these findings, transwell assays were performed. Migration assays revealed that the number of PIP overexpressed SCC15 and SCC25 cells crossing the membrane were significantly reduced significantly compared with control cells (p < 0.01). However, PIP silencing in SCC15 and SCC25 cells significantly increased the number of cells crossing the membrane border (p < 0.05). Invasion assays produced similar findings (p < 0.05; Figure 3b). Taken together, these data suggest that the overexpression of PIP inhibits the migration and invasion abilities of SCC15 and SCC25 cells, whilst PIP silencing produces the opposite effect.

| IHC analysis of PIP expression in OSCC specimens
We divided the 45 cases into four groups according to tissue type.
From IHC staining, we observed PIP expression in both gland and stromal cells. The expression levels of PIP were high in adjacent nontumor and highly differentiated tissues, compared with moderately and poorly differentiated tissues (Figure 4).

| PIP influences cell signaling
PPI network analysis suggested that PIP was associated with components of the AKT/MAPK signaling axis. We therefore assessed the expression levels of AKT and ERK1/2 (Figure 5a). The expression levels of p-ERK1/2 and p-AKT were significantly higher in siRNA-PIP F I G U R E 3 The effects of PIP on OSCC migration and invasion in vitro. (a) The view of wound-healing migration assay; the average rate of SCC15 and SCC25 cells migration at 12 and 24 hr after control, PIP vector or siRNA-PIP transfection. All data were shown as mean ± SD. *p < 0.05, **p < 0.01. (b) The results of the transwell assay of SCC15 and SCC25 cells at 24 hr after control, PIP vector or siRNA-PIP transfection; relative ratios of migrated and invasive cells per field are shown. All data were shown as mean ± SD. Invasion assays followed an identical pattern (Figure 7b). tissue. We also found PIP expression in OSCC tissue is lower than normal tissue. Moreover, the network in a previous study (Debily et al., 2009) identified appears connected with an inhibition of proliferation coupled with an increase of apoptosis in breast cancer cell lines which aid us to explain our results of clonogenic and MTT assays.
PIP is a secretory glycoprotein found primarily in apocrine and mucosal tissues, including breast tissue and the salivary glands Parris et al., 2014;Priyadarsini et al., 2014). PIP binds to a diverse range of oral bacteria, signifying its protective function in the oral mucosa through inhibiting bacteria settlement and growth (Nistor, Bowden, Blanchard, & Myal, 2009).
PIP has also been shown to belong to a group of salivary proteins, whose profusion is reduced in the presence of oral bleeding (Fleissig et al., 2010;Haigh et al., 2010). PIP has also been reported to associate with secretory lesions and is related to atypical and benign precursors (Asirvatham, Falcone, & Kleer, 2016;Gangadharan et al., 2018;Gown, Fulton, & Kandalaft, 2016 Although these results are promising, this study has some limitations. We did not validate our findings in vivo which may affect the reliability of the data. According to our previous studies, a correlation between PIP and cell adhesion requires investigation. Simultaneously, experiments exploring the mechanism of PIP mediated regulation of OSCC cell growth requires further studies. We did however identify that PIP expression is significantly lower in both OSCC cells and tissue. These findings may aid clinical diagnosis should PIP expression be related to the degree of OSCC malignancy. PIP may also benefit the early diagnosis of OSCC, and therefore improve both treatment and patient prognosis.
In summary, iTRAQ analyses followed by high-throughput 2D LC-MS/MS were used to screen DEPs in OSCC and normal tissue in patient samples for the first time. We provide a comprehensive insight into PIP and its use as a novel biomarker of OSCC. We believe that these findings can improve clinical efficacy and reduce OSCC related mortality in future studies.