Golgi integral membrane protein 4 manipulates cellular proliferation, apoptosis, and cell cycle in human head and neck cancer.

The incidence of global head and neck cancer has increased markedly in the last 10 years, and its prognosis is poor, which seriously endangers people's life and health. At present, there are few studies on its pathogenesis. Golgi integral membrane protein 4 (GOLIM4) is a major member of the Golgi apparatus transporter complex, and its role in tumor is unclear. The present study found that GOLIM4 was the key target protein downstream of stromal interaction molecule 1 (STIM1), which can inhibit the proliferation of head and neck cancer cells FaDu (human pharyngeal squamous carcinoma cell) and Tca-8113 (human tongue squamous carcinoma cell) with knockdown of GOLIM4 by lentivirus. And the decreased expression of GOLIM4 induced cellular apoptosis. Further experiments revealed that FaDu cell cycle progression was changed after GOLIM4 silence, G1 phase arrest and the number of G2/M cells decreased significantly. It was also found that the cells in S-phase decreased markedly after GOLIM4 was knocked down compared with the control group by 5-bromo-2'-deoxyuridine (BrdU) incorporation experiment. In conclusion, we found that GOLIM4, as the target gene downstream of STIM1, inhibited the proliferation of head and neck cancer, promoted apoptosis, and regulated cell cycle progression, and GOLIM4 is a novel oncogene in head and neck cancer and might help in developing promising targetted therapies for head and neck cancer patients.

Introduction

The incidence of global head and neck cancer has increased markedly in the last 10 years, and over 90% of head and neck cancer are squamous cell carcinoma. The prognosis of head and neck cancer is not optimistic, due to its rapid growth and infiltration. Because of the anatomical structure and the importance of the preservation of organ function, the treatment of head and neck cancer is complicated, the operative or nonoperative treatment not only affects the patient’s survival, but also greatly affects the patient’s quality of life. Postoperative recurrence and metastasis remained the main cause of death in patients with head and neck cancer, such as oral tongue squamous cell carcinoma [1]. Therefore, it is urgent to study the pathogenesis and development of head and neck cancer.

The imbalance of Ca2+ homeostasis plays an important role in the development of tumor [2]. The dynamic equilibrium of cytoplasmic calcium concentration regulates several intracellular signaling pathways, which are related to the expression of many growth factors and cytokines. These affect many basic cellular functions, such as cell proliferation, migration, differentiation, apoptosis, and gene regulation [3]. And store-operated Ca2+ entry (SOCE) [4] plays a key role in maintaining Ca2+ homeostasis.

In our previous study, we found that the expression of calcium channel protein STIM1 (stromal interaction molecule 1), a member of SOCE complex in head and neck cancer was significantly higher than that in the adjacent tissues. Furthermore, the expression of STIM1 can be detected in FaDu, um-scc-17a, and HEP-2 head and neck cancer cell lines. Knockdown of STIM1 dramatically inhibited the proliferation rate and the cloning ability of FaDu cells and the proportion of apoptotic cells increased. Moreover, silencing STIM1 significantly restrained tumor growth in nude mice [5-7]. In order to further define the molecular mechanism of STIM1 in OSCC, we used gene microarray and bioinformatics analysis and found that Golgi integral membrane protein 4 (GOLIM4) was the target of STIM1. GOLIM4 is a component of the Golgi transport complex and plays an important role in the transport of Golgi proteins [8]. Although it has been reported that the Golgi apparatus can be involved in the biological process of tumor, the functions of GOLIM4 in the development of tumorigenesis are not clear.

In the present study, we found that the STIM1-GOLIM4 signal axis played a pivotal role in head and neck cancer. Consistent with STIM1, the expression of GOLIM4 protein in head and neck cancer cells was high, and low GOLIM4 significantly inhibited the proliferation of head and neck cancer cell lines. Moreover, the low expression of GOLIM4 induced apoptosis and regulated cell cycle process, resulting in cellular G1-phase block, and the number of S-phase cells significantly reduced. This suggested that STIM1 can influence the proliferation, apoptosis, and cell cycle progression of head and neck cancer by regulating the expression of GOLIM4.

Materials and methods

Cell lines and cell culture

The cell lines FaDu (human pharyngeal squamous carcinoma cell) and Tca-8113 (human tongue squamous carcinoma cell) were purchased from Cell Bank of the Chinese Academy of Sciences in Shanghai and cultured in DMEM medium (10-013-CVR, Corning) containing 10% FBS (VS500T, Ausbian), 100 U/ml penicillin and 0.1 mg/ml streptomycin at 37°C in a humidified atmosphere containing 5% CO2.

High-content screening assay

Cell viability was detected the by the Celigo™ cytometer (Nexcelom) using 96-well plates. FaDu and Tca-8113 cells were seeded at 3000 or 1500 cells per well with 100 μl culture medium. After 24 h, cells plates were read once a day using Celigo™ cytometer to accurately calculate the number of green fluorescent cells from each scan of a plate for five consecutive days. Each experiment was analyzed in triplicate and at least three independent experiments were performed.

Cell proliferation assay

Cell proliferation was detected by 5-bromo-2′-deoxyuridine (BrdU) incorporation assay, measuring DNA synthesis using the BrdU Cell Proliferation ELISA kit (11647229001, Roche). Cells were seeded in 96-well plates at least for three independent experiments. The experiment was performed according to the manufacturer’s protocol. The number of proliferating cells is represented by the level of BrdU incorporation which directly correlates to the absorbance values and the OD was quantitated at 450 nm with a reference wavelength of 630 nm.

Relative quantitative-PCR

Cell total RNA was extracted using TRIzol reagent (3101-100, Pufei Biotechnology) according to the manufacturer’s instructions. TRIzol-prepared RNA was reverse-transcribed into cDNA using M-MLV Reverse Transcriptase (M1705, Promega). Based on SYBR Green PCR Master Mix (DRR041B, TaKaRa), relative quantitative-PCR was subsequently performed on LightCycler 480 Ⅱ (Applied Roche) using cDNA as template. GAPDH was used as a control for normalization. The relative GOLIM4 expression was normalized to GAPDH, and data analysis was conducted using the comparative CT method. The PCR primers used were as follows: GOLIM4 forward primer: 5′-AAACTCTATGCTCCCACCC-3′, Reverse primer: 5′-GCTGCTCTTTCCACTCCC-3′, GAPDH forward primer: 5′-TGACTTCAACAGCGACACCCA-3′, Reverse primer: 5′-CACCCTGTTGCTGTAGCCAAA-3′.

Western blot analysis

Cell lysates were prepared with RIPA lysis buffer (P0013B, Beyotime) at 4°C. Protein concentrations were measured by BCA Protein Assay Kit (P0010S, Beyotime). Proteins were fractionated using SDS/PAGE and then electroblotted on to 0.45-μm PVDF membranes and then blocked in 5% nonfat milk dissolved in TBST for 1 h at room temperature. The membranes were incubated with primary antibodies overnight, washed in TBST for 10 min three times, and then incubated with the HRP-conjugated secondary antibody. Primary antibodies against GAPDH (sc-32233) were purchased from Santa Cruz Biotechnology. The antibodies against GOLIM4 was purchased from Abnova (PAB28477). The antibodies against MDM2 (ab38618), CDK6 (ab151247), and GADD45 (ab180768) were purchased from Abcam. Secondary antibody goat anti-rabbit IgG-HRP (sc-2005) were obtained from Santa Cruz Biotechnology. The blots were detected using Pierce™ ECL Western Blotting Substrate kit (M3121, Thermo Scientific).

RNAi

The lentivirus-mediated shRNA expressing vector was constructed and synthesized from Genechem Technology (Shanghai, China). The sequences of shRNA targetting genes were showed in Table 1. Cells were seeded at ~50% confluence and infected at a multiplicity of infection (MOI) of 50 and 5 μg/ml polybrene. Forty-eight hours after infection, the cells were cultured in DMEM containing puromycin (2 μg/ml) to obtain stably infected cells.

Table 1

The shRNA sequence for 20 genes used in Figure 1A

Gene symbol	shRNA targetting sequences
GOLIM4	GCACCAAGATGAAGCAGAA
DLGAP5	GGAAAUCACUGUCUCAGAA
PIF1	GGCCGAGCCUAGCACAGAA
PARP12	GGGAAGAACUGUAGGAAUA
EIF5A2	GGAGAUGUCAACUUCCAAA
GPX8	GCUAUUUCUUCUACAACUA
HIST1H2BD	GCUAUUCAGUGUAUGUGUA
BAG2	GGUUGAAGCUUUGAGAGAA
TPBG	GCACAGUCAAGUGCGUUAA
CSGALNACT2	GCAUAGGCUAUCAGAGCAA
PRELID2	GGAUUGCAAUCUGUCAGAA
TNFAIP2	AGAAGAAGAAGAAGUCCAA
KLHL5	GCUUCAAUUGGUUACCGAA
ARHGAP11A	GCAGUAGUACAGACUCUUA
HJURP	GGCUGAUAGAGAAGUACAA
DDX60L	GAUUCAGAGUGCAUAUAAA
PARP14	CAAAGAAGAUGUUAAAGAA
CHST11	GGUCAUCUUCUAUUUCCAA
CKAP2	CCUAAGGAUAGUAAUCAAA
OAS3	AGCGGCAGUUCGAGGUCAA

Cell apoptosis measurement

Cell apoptosis was evaluated by Annexin V Apoptosis Detection Kit APC (88-8007, BD Bioscience). After 4 days of lentivirus infection, FaDu and Tca-8113 cells were harvested and stained according to manufacturer’s protocol. Data acquisition and analysis were performed in a Guava easyCyte HT (Millipore) and data were analyzed and calculated using FlowJo software.

Cell cycle analysis

Cell cycle assay was determined by Propidium Iodide (P4170, Sigma) staining using a flow cytometer. In brief, cells infected with control lentivirus or GOILM4-shRNA lentivirus were seeded in six-well plates and cultured to 80% confluence, then harvested by trypsinization and fixed with 70% cold ethanol at 4°C for 6 h to overnight. Then, cells were incubated with 10 mg/ml RNase (EN0531, Fermentas) for 30 min at room temperature. DNA distribution was analyzed by FACS on a Guava easyCyte HT (Millipore) cytometry and data were analyzed and calculated using FlowJo software.

Statistical analysis

The statistical analyses were performed using SPSS 16.0 software (SPSS, Inc., Chicago, IL, U.S.A.). Each experiment was conducted as at least three independent repeats. Student’s t tests were used to analyze the differences between two groups. A probability value of less than 0.05 was considered significant.

Results

GOLIM4 is elevated in head and neck cancer

To identify the genes which regulated by STIM1 that affect the growth, apoptosis, and cell cycle of head and neck cancer cells, we silenced STIM1 in FaDu cells (human pharyngeal squamous carcinoma cell) and found 20 candidate genes significantly down-regulated. Then we used lentivirus to knockdown these 20 candidate genes in FaDu cells (Table 1), and tested the effect of candidate genes on cell proliferation. We found that knockdown of GOLIM4 and DLGAP5 could significantly inhibit proliferation of FaDu cells (Figure 1A). The fluorescence intensity of cells knockdown of GOLIM4 was observed under microscope, and it was found that the fluorescence intensity of cells knockdown of GOLIM4 decreased significantly compared with the negative control group (Figure 1B). And analysis the number of cells also found that knockdown of GOLIM4 significantly inhibited the growth of FaDu cells (Figure 1C). Furthermore, we compared the expressions of GOLIM4 in 44 normal tissues and 521 head and neck squamous cell carcinoma from TCGA (The Cancer Genome Atlas) database, and found that the expression of GOLIM4 was significantly higher in tumor tissues (Figure 1D). In addition, we also found a positive correlation between the expression of GOLIM4 and STIM1 in head and neck tumor tissues (Figure 1E).

Figure 1

GOLIM4 is decreased when knocked down of STIM1

(A) Cell proliferation was measured after knockdown of 20 candidate genes in FaDu cells. (B) The representative images of FaDu cells that infected with negative control lentivirus (shCtrl-EGFP) and shGOLIM4-EGFP lentivirus. Green fluorescence showed the viable cells. (C) The growth curves of the corresponding negative control group (shCtrl) and shGOLIM4 group in the FaDu cells as described in (A). (D) The expression of GOLIM4 in head and neck cancer tissues (n=521) and normal donors (n=44) from TCGA database. (E) The correlation analysis of GOLIM4 and STIM1 in head and neck cancer tissues from TCGA database. Pearson correlation coefficient was used, P<0.001, R = 0.43.

Knockdown of GOLIM4 inhibits head and neck cancer cell viability

In order to further clarify the effect of GOLIM4 on cell viability, we first examined the expression of GOLIM4 at RNA and protein levels with lentivirus infection, and found that the knockdown efficiency reached more than 60% (Figure 2A–D). Then, we use Celigo experiment to detect the effect of GOLIM4 on the two head and neck cancer cell lines FaDu cells and Tca-8113 cells (human tongue squamous carcinoma cell). According to the fluorescence intensity, the group that knockdown of GOLIM4 had lower active cells than the control group since day 4 (Figure 2E), and that the number of active cells decreased significantly (Figure 2F). It is proved that GOLIM4 can maintain cell proliferation activity, the low expression of GOLIM4 can inhibit the growth of head and neck cancer cells.

Figure 2

Knockdown of GOLIM4 significantly inhibits head and neck cancer cell viability

(A,B) The mRNA level (A) and protein level (B) of GOLIM4 after lentivirus infected in FaDu cells. **P<0.01, ***P<0.001, Student’s ttest. Three repeated experiments. (C,D) The mRNA level (C) and protein level (D) of GOLIM4 after lentivirus infected in Tca-8113 cells. **P<0.01, ***P<0.001, Student’s ttest. Three repeated experiments. (E,F) HCS assay detected cell number change for five consecutive days after knockdown of GOLIM4 in Tca-811 cells (E). The corresponding quantitative analysis as described in (F). **P<0.01, ***P<0.001, Student’s t test. Abbreviation: HCS, high-content screening.

GOLIM4 affects the cell cycle progression of head and neck cancer cells

As GOLIM4 regulates cell proliferation activity, we wanted to explore whether GOLIM4 affects the cell cycle of head and neck cancer cells. Therefore, we used the flow cytometry to detect the changes in the cell cycle of FaDu and Tca-8113 cells after knockdown of GOLIM4 (Figure 3A,C). We found that the percentage of cells in G1-phase was increased by GOLIM4 knockdown, whereas the cells in the G2/M phase were significantly reduced (Figure 3B,D). The results showed that low expression of GOLIM4 could cause G1-phase arrest, suggesting that the inhibitory effect of shGOLIM4 on cell growth may be caused by cell cycle block in G1 phase. Then we detected the expression of three key genes of p53 pathway MDM2, CDK6, and GADD45A when knockdown of GOLIM, and found that MDM2, CDK6 were decreased while GADD45A was increased (Supplementary Figure S1A). This is consistent with the effect of STIM1 on head and neck cancer cell cycle.

Figure 3

Knockdown of GOLIM4 results in G1 phase arrest in head and neck cancer cells

(A,B) Flow cytometry was used to detect cell cycle change after knockdown of GOLIM4 in FaDu cells (A). The number of cells in G1 phase, S phase, and G2/M phase was quantitatively analyzed after GOLIM4 knockdown in FaDu cells (B). (C,D) Flow cytometry was used to detect cell cycle change after knockdown of GOLIM4 in Tca-8113 cells (C). The number of cells in G1 phase, S phase, and G2/M phase was quantitatively analyzed after GOLIM4 knockdown in Tca-8113 cells (D). **P<0.01, Student’s t test.

Decreased GOLIM4 expression results in reduced S-phase cells

In addition, we further applied BrdU incorporation assay to test the effect of GOLIM4 expression on the number of S-phase cells. It was found that knockdown of GOLIM4 in FaDu and Tca-8113 cells could significantly reduce the number of S-phase cells (Figure 4A,B). This result implicates that GOLIM4 can not only lead to G1 phase arrest, but also significantly keeps cells from entering into S-phase, thus regulating the cell proliferation.

Figure 4

The low expression level of GOLIM4 in head and neck cancer cells results in a decrease in S-phase cells

(A,B) BrdU incorporation assay measured the number of S-phase cells in the FaDu and Tca-8113 cells after knockdown of GOLIM4. **P<0.01, ***P<0.001, Student’s t test.

Low expression of GOLIM4 induces apoptosis of head and neck cancer cells

Finally, we also used flow cytometry to detect the effect of GOLIM4 expression on cellular apoptosis (Figure 5A,C). Compared with the control group, the proportion of annexin V-positive cells in GOLIM4 knockdown group was increased significantly (Figure 5B,D), which indicated that the number of early-stage apoptotic cells elevated by GOLIM4 knockdown. Therefore, GOLIM4 can maintain cell viability by reducing apoptosis of head and neck cancer cells.

Figure 5

Knockdown of GOLIM4 induced apoptosis in head and neck cancer cells

(A) Flow cytometry detected the expression of annexin V after knockdown of GOLIM4 in FaDu cells. (B) Quantitation of cell apoptosis induced by knockdown of GOLIM4 in (A). *P<0.05, **P<0.01, ***P<0.001, Student’s t test. (C) Flow cytometry detected the expression of annexin V after knockdown of GOLIM4 in Tca-8113 cells. (D) Quantitation of cell apoptosis induced by knockdown of GOLIM4 in (B). **P<0.01, Student’s t test.

Discussion

Our research further revealed the molecular mechanism of STIM1 in head and neck cancer, which regulated cell proliferation, apoptosis, and cell cycle, and found GOLIM4 was the target gene downstream of STIM1. Results showed that knockdown of GOLIM4 could inhibit cell proliferation, promote apoptosis, and change cell cycle progression. These results revealed the mechanism of STIM1-GOLIM4 signaling pathway in the development of head and neck cancer, and expounded the regulation network of endoplasmic reticulum–Golgi interaction in the formation of head and neck cancer.

Some studies have found that the Golgi apparatus can participate in the process of tumor biology. For example, Golgi (Golgi protein, GOLPH2) participates in the development of hepatocellular carcinoma and is considered to be a potential indicator of liver cancer serum [9,10]. Golgi protein GOLM1 is related to the development of prostate cancer [11]. Phosphatidylinositol 4-phosphate can regulate the adhesion of tumor cells thus influence the metastasis of breast cancer [12], and it is reported that the polarization of the Golgi body also participates in the metastasis of breast cancer [13]. In addition, the Golgi apparatus also participate in colorectal cancer by controlling the sensitivity of mTOR to amino acids, and is closely related to the prognosis of colorectal cancer [14]. The difference of Golgi network also determines the secretory phenotype of different tumors [15].

GOLIM4, identified as a Golgi protein, is a 130-kDa membrane protein that is integrated into the cis/medial Golgi apparatus, which is a kind II Golgi protein. GOLIM4 is a Golgi protein that circulates between Golgi apparatus and endosomes, and its shuttle is pH sensitive. GOLIM4 is considered being a homologous with the membrane proteins of the Golgi apparatus of rat. Although located in the early Golgi apparatus, it has advanced Golgi apparatus modification function [16]. The ST-1 of Shiga toxin can directly combine with GOLIM4, so that the ST-1 is free from lysosome degradation, the increase in Mn can down-regulate the expression of GOLIM4, reducing the toxicity of Shiga toxin [17]. The increased, nontoxic dose of Mn protects the ST-1-induced cell death of Shigella by lowering the GOLIM4. Mn is combined with GOLIM4 in the Golgi matrix, causing the GOLIM4 to form an oligomer, and then the complex is transferred to lysosomal degradation [18]. These suggest that GOLIM4 may be associated with cell death, but the role of GOLIM4 in tumors has not been reported. The Golgi apparatus can also co-ordinate with the endoplasmic reticulum to regulate the biological behavior of the tumor [19]. Although the function of Golgi apparatus in the development of multiple tumors has been reported, the function of Golgi and ER complex proteins and their interactions with endoplasmic reticulum in head and neck cancer is unclear. Here, we showed that GOLIM4 regulated the proliferation, apoptosis, and cell cycle progression of head and neck cancer. We first clarified the functions of GOLIM4 in tumors and the roles of Golgi–endoplasmic reticulum interaction in head and neck cancer, but the interaction of STIM1 and the expression of GOLIM4 as well as the role of GOLIM4 in the formation of head and neck cancer needed further study.

In summary, we have revealed that STIM1-GOLIM4 influence the development of head and neck cancer through the regulation of cell proliferation, apoptosis, and cell cycle, and found that endoplasmic reticulum and Golgi apparatus interaction was important in the occurrence of head and neck cancer, suggesting that the endoplasmic reticulum and Golgi apparatus maybe the cut-in point targetted head and neck cancer cells. It provides a new potential target for clinical treatment of head and neck cancer.

GOLIM4 affects the expression of MDM2, CDK6 and GADD45A.