Knocking down glycoprotein nonmetastatic melanoma protein B suppresses the proliferation, migration, and invasion in bladder cancer cells

Abstract

Glycoprotein nonmetastatic melanoma protein B is a type 1 transmembrane protein that has been recently found to play a role in cancer cell proliferation, angiogenesis, and invasion. Due to its potential responsibility in cancer aggressiveness, the main objective of this work was to investigate its expression in bladder cancer and the biological functions in bladder cancer cells. Using immunohistochemistry, western blot, and reverse transcription polymerase chain reaction, we analyzed the expression of glycoprotein nonmetastatic melanoma protein B in bladder cancer tissues and bladder cancer cell lines. The effects of glycoprotein nonmetastatic melanoma protein B on proliferation, migration, and invasion were tested after knocking down the glycoprotein nonmetastatic melanoma protein B in bladder cancer cells with small interfering RNAs by CCK-8, Transwell, and Matrigel assays. Our results showed that glycoprotein nonmetastatic melanoma protein B protein was highly expressed in the bladder cancer tissues and cell lines. Downregulating glycoprotein nonmetastatic melanoma protein B could suppress the proliferation, migration, and invasion in bladder cancer cells. Glycoprotein nonmetastatic melanoma protein B expression was related to the poor differentiation and recurrence by immunohistochemistry analysis. The survival analysis also showed that glycoprotein nonmetastatic melanoma protein B was related to the patient prognosis. In conclusion, Glycoprotein nonmetastatic melanoma protein B protein was highly expressed in the bladder cancer, which was related to the poor prognosis in bladder cancer patients. Glycoprotein nonmetastatic melanoma protein B promoted the proliferation, migration, and invasion in bladder cancer cells.

Introduction

Bladder cancer is a major public health problem, comprising 90% of all primary bladder tumors, and is the fourth most prevalent type of cancer in men. The number of estimated new cases and estimated deaths for 2016 is 769,060 and 16,390, respectively,¹ in United States. The patients with early-stage bladder cancer can expect a 5-year survival rate of more than 90%.² However, the advanced-stage patients, especially the patients with distant metastasis, had a low 5-year survival rate. Therefore, early diagnosis is very important to patients, for early diagnosis helps to guide the medical strategy for the next step. To date, there are no effective methods to predict the prognosis in bladder cancer. Therefore, new tools should be developed for diagnosis and predicting prognosis for bladder cancer.

Glycoprotein nonmetastatic melanoma protein B (GPNMB) was initially found in metastatic melanoma cells, also named as osteoactivin (OA), which is a type 1 transmembrane glycoptotein.³ It is comprised of three domains: a long extracellular domain (ECD, ectodomain), a single transmembrane region, and a short cytoplasmic domain.⁴ In recent studies, GPNMB has been found to express in several cancer tissues, such as melanoma, uveal melanoma, colorectal cancer, breast cancer, glioma, and hepatocellular carcinoma.^3,5–11 The biological roles of GPNMB have been also linked to T-cell activation, fibroblast differentiation, and development of bone cells.¹² Besides, more studies showed that it was a promoter for cancer cell progression and metastasis.

Due to the expression and biological functions of GPNMB, in this study, we investigated GPNMB expression in bladder cancer tissues and its biological behaviors in bladder cancer cells. The results showed that GPNMB expression level was related to the prognosis factors of bladder cancer cells, and knockdown of GPNMB could suppress the bladder cancer cells’ proliferation, migration, and invasiveness.

Materials and methods

Clinical samples

The tissue microarray, including 60 different cases of bladder cancer and adjacent normal bladder tissues, was purchased from Shanghai Outdo Biotech Co., Ltd (Shanghai, China).

Cell culture and small interfering RNAs

Two bladder cancer cell lines (BIU87 and EJ) were gifts from Peking University People’s Hospital, and 5637 was purchased from American Type Culture Collection (ATCC). Each cell lines were cultured in RPMI 1640 (HyClone; HyClone Laboratories, Inc., Logan, UT, USA) medium with 10% fetal bovine serum (FBS) (HyClone) and incubated at 37°C in a humidified atmosphere containing 5% CO₂. The small interfering RNAs (siRNAs) were purchased from GenePharma Company (Shanghai, China). siRNA sequences are as follows: siRNA1: 5′-CCAUCUUGCUGUACAAAAAdTdT-3′ (sense) and 5′-UUUUUGUACAGCAAGAUGGdTdT-3′ (antisense); siRNA2: 5′-GCACGGGUUUCUAUAAACAdTdT-3′(sense) and 5′-UGUUUAUAGAAACCCGUGCdTdT-3′ (antisense).

Immunohistochemistry analysis

The primary antibody (GPNMB) was purchased from Abcam Company (Cambridge, UK). The clinical tissues were dewaxed and antigen retrieval was progressed. The tissues were incubated with the primary antibody overnight at 4°C and with the secondary antibody, which was purchased from Zhongshan Bridge (Beijing, China), for 20 min at room temperature. The Envision™ ABC Kit was used for staining. To evaluate the scores of immunohistochemistry, we used the following methods.

The percentage of stained cells per field was defined as follows: 0, negative; 1, 1%–25%; 2, 26%–50%; and 3, 51%–100% of the cells. The staining intensity was scored on a four-tiered scale as follows: 0, absence of signal; 1, low-intensity signal (light brown); 2, moderate-intensity signal (brown); and 3, high-intensity signal (dark brown). The frequency score and intensity score were multiplied to obtain the score for each field.

The frequency score and intensity score were multiplied to obtain the score for each field, and the final score for each case was the average score of the five fields. The score was described as negative (−) when the score was 0–1. The scores 2–3, 4–6, and 7–9 were described as +, ++, and +++, respectively. All evaluations were conducted using a LEICA DM4000 B/M microscope.

Western blot

Protein was extracted from the bladder cancer cells and was measured by Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, IL, USA). The membranes were incubated overnight at 4°C with the following primary antibodies: GPNMB, MMP2, MMP9, β-catenin, and GSK-3β (ABcom Company). The reference protein was glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The immunocomplex was detected using the ECL Plus kit (Amersham Biosciences (GE Healthcare), Buckinghamshire, UK), and the band density was analyzed with ImageJ software.

Reverse transcription polymerase chain reaction

Total RNA of bladder cancer cell lines 5637, BIU87, and EJ was extracted by TRIzol Reagent and then reverse transcribed using reverse transcription polymerase chain reaction (RT-PCR) RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific) according to the manufacturer’s instruction. Prime: GPNMB: forward: 5′-AAGTGAAAGATGTGTACGTGGTAACA-3′ and reverse: 5′-TCGGATGAATTTCGATCGTTCT-3′. GAPDH: forward: 5′-CAAGGTCATCCATGACAACTTTG-3′; reverse: 5′-GTCCACCACCCTGTTGCTGTAG-3′.

Cell proliferation assay

The 5637 and BIU87 cells were transfected with siRNA (50 µg/mL) in 96-well plates. Cell proliferation was analyzed using the CCK-8 Cell Proliferation Assay Kit (Invitrogen, Carlsbad, CA, USA) at optical density (OD) 450 nm.

Cell migration and invasion assay

Cellular migration and invasion were assessed using Transwell inserts (8-µm pore diameter) after being transfected with siRNA (50 µg/mL) in 24-well plates. The 5637 and BIU87 cells were counted as 2 × 10⁵ cells in the upper chamber with free FBS medium and the bottom chamber with 10% FBS medium for 30 h. The number of cells was counted in at least six randomized fields under a microscope. Results were obtained from at least three individual experiments.

Statistical analyses

Data were presented as values of mean ± standard error of the mean (SEM). Statistical analysis was carried out with chi-square test and Cox analysis. Values of p < 0.05 were considered as statistically significant different. All statistical evaluations were performed with SPSS 17.0 (SPSS, Chicago, IL, USA).

Results

GPNMB was overexpressed in the bladder cancer tissues and cell lines

In this study, the expression levels of GPNMB in three representative bladder cancer cell lines (5637, BIU87, and EJ) and normal bladder tissues were determined by western blot and RT-PCR. Immunohistochemistry showed that positive staining was frequently located in the cytoplasm of cancer cells in bladder cancer tissues (56/60, 93.3%). The normal bladder tissues were negatively stained (Figure 1).

Figure 1.

GPNMB was highly expressed in the bladder cancer tissues and cell lines. (a) The immunohistochemistry staining of bladder cancer tissue and normal tissues. (b) GPNMB was detectable in the 5637, BIU87, and EJ cell lines by western blot. (c) GPNMB was detectable in the 5637, BIU87, and EJ cell lines by RT-PCR.

[Figure omitted. See PDF]

Knocking down GPNMB reduced the proliferation of bladder cancer cells

To determine whether ectopic GPNMB expression was linked to proliferation and survival in bladder cancer tissues, we evaluated the proliferation ability using CCK-8 assays. The results showed that in both 5637 and BIU87 cells, the proliferation was suppressed significantly. In the third, fourth, and fifth days, the growth rate was higher in the SiNC group than the SiGPNMB group (Figure 2).

Figure 2.

Knocking down GPNMB reduced the proliferation of bladder cancer cells. Using CCK-8 assay, the proliferation of bladder cancer cells 5637 and BIU87 was investigated after being transfected with siRNAs for 5 days.

[Figure omitted. See PDF]

Downregulating GPNMB suppressed the migration and invasion in bladder cancer cells

To explore the involvement of GPNMB in cell motility, Transwell assays were carried out to quantitatively determine the effects of GPNMB on cell migration. After knocking down GPNMB, the SiNC group had more migrated cells than the SiGPNMB group. In the Matrigel assays, to explore the invasiveness, the SiNC group also had more migrated cells, suggesting that GPNMB promotes the bladder cancer migration and invasion (Figures 3 and 4).

Figure 3.

Knocking down GPNMB reduced the migration of bladder cancer cells. Knocking down the expression of GPNMB inhibited the migration ability of 5637 and BIU87 cells by Transwell assay, which is discussed above. Data represent the mean ± SEM of cells of at least six random fields under microscope, and the experiment was repeated three times.

[Figure omitted. See PDF]

Figure 4.

Knocking down GPNMB reduced the migration of bladder cancer cells. Knocking down the expression of GPNMB inhibited the migration ability of 5637 and BIU87 cells by Matrigel assay, which is discussed above. Data represent the mean ± SEM of cells of at least six random fields under microscope, and the experiment was repeated three times.

[Figure omitted. See PDF]

GPNMB was related to the poor differentiation and recurrence in bladder cancer

In addition, we analyzed the immunohistochemistry data of the bladder cancer tissues of GPNMB and found that a significant GPNMB expression increased poor differentiation and recurrence in tissues (p < 0.05). GPNMB was expressed in 91.2% high-grade patients (31/34) and 95.7% recurrence patients (22/23). GPNMB expression had no relationship with sex, age, and tumor size (Table 1).

Table 1.

Clinicopathologic variables and evaluation of GPNMB immunostaining in bladder cancer tissues.

	No. of patients	Scores for non-GPNMB signal				p value
		(−)	(+)	(++)	(+++)
Age (years)						0.8696
>60	35	2	11	12	10
⩽60	25	2	10	7	6
Sex						0.5128
Male	52	4	19	15	14
Female	8	0	2	4	2
Histological grade						0.0083
High grade	34	3	6	15	10
Low grade	26	1	15	4	6
Tumor size						0.9321
>3 cm	29	2	10	9	8
⩽3 cm	31	2	11	10	8
Recurrence						0.0205
Yes	33	1	8	13	11
No	27	3	13	6	5

Correlation between GPNMB expression and survival in bladder cancer patients

The correlation between GPNMB expression and 5.5-year overall survival of patients with bladder cancer patients was elevated according to the results of Kaplan–Meier analysis. The survival of patients with weak positive (negative and “+”) GPNMB staining was significantly longer than that of patients with strong positive (“++” and “+++”) GPNMB staining (p < 0.05) (Figure 5). Cox multivariate analysis also showed that GPNMB strong positive staining (hazard ratio = 3.842, p = 0.043) and recurrence (hazard ratio = 7.626, p = 0.013) were risk factors of poor prognosis for bladder cancer patients (Table 2).

Figure 5.

The correlation between GPNMB expression and 5.5-year overall survival of patients with bladder cancer patients.

[Figure omitted. See PDF]

Table 2.

Cox proportional hazards model analysis of prognostic factors in patients with bladder cancer.

Characteristics	HR	95% CI	p value
Age (⩽60/>60)	1.028	0.970–1.090	0.351
Gender (male/female)	0.282	0.079–1.009	0.052
Tumor size (>3 cm/⩽3 cm)	0.479	0.110–2.087	0.327
Histologic grade (high grade/low grade)	1.728	0.446–6.694	0.429
GPNMB staining (weak/strong)	3.842	1.043–14.148	0.043
Recurrence (yes/no)	7.626	1.531–37.983	0.013

CI: confidence interval; HR: hazard ratio.

Knocking down GPNMB downregulated the expression of matrix metalloproteinases, β-catenin, and GSK-3β

Since GPNMB may promote the bladder cancer cells to proliferate and migrate, we investigated the mechanism of how GPNMB could affect the biological behaviors in bladder cancer cells. After knocking down GPNMB, we found that the MMP2, MMP9, β-catenin, and GSK-3β expression levels were downregulated, suggesting that GPNMB may affect the Wnt pathways to affect the cell migration (Figure 6).

Figure 6.

Knocking down GPNMB downregulated the expression of matrix metalloproteinases, β-catenin, and GSK-3β. After being transfected with siRNA, the expression of matrix metalloproteinases, β-catenin, and GSK-3β was examined by western blot.

[Figure omitted. See PDF]

Discussion

In this study, we found that GPNMB was overexpressed in bladder cancer tissues and cell lines 5637, BIU87, and EJ. By knocking down the expression of GPNMB, the bladder cancer cells’ proliferation, migration, and invasion ability were suppressed. Moreover, GPNMB expression was related to the poor differentiation and recurrence in bladder cancer.

GPNMB is a widely expressed protein in the cancer tissues. In the urinary system, GPNMB was found to express in the prostate cancer and renal carcinoma.¹³ GPNMB is expressed in the cancer tissues and gets secreted in the cancer microenvironment by some unknown mechanism. In this study, GPNMB expression was detected in the bladder cancer tissues cells by immunohistochemistry, western blot, and PCR, but we did not investigate whether GPNMB could be released in the cell culture yet. Therefore, this part may be carried out in the future.

Besides, we explored the relationship between GPNMB and biological behaviors. The results showed that cell proliferation, migration, and invasion of bladder cancer cells 5637 and BIU87 were inhibited after GPNMB knockdown. In other studies, such as lung cancer and pancreatic cancer, downregulating GPNMB expression could suppress proliferation, migration, and invasion abilities in these cancer cells.^8,14 Elevated expressions of GPNMB/OA have been shown to promote invasion and metastasis of prostate cancer, hepatocellular carcinoma, glioma, and breast cancer. Besides, downregulating GPNMB also induced the apoptosis in pancreatic cancer cells. The previous reports could further verify the biological functions of GPNMB, which promote the cancer cells development.

More importantly, we analyzed that GPNMB expression was related to the poor differentiation and recurrence in bladder cancer, which was important to the 5-year survival for a patient, suggesting that GPNMB might be the new tool to assess the prognosis in bladder cancer diagnosis. The survival results also suggested that GPNMB expression might predict the patient prognosis and guide the next treatment.

To date, there is no exact mechanisms for GPNMB affecting the biological behaviors in bladder cancer, but some reports might help us to a further study. Vucic et al. found that GPNMB expression might have effects with matrix metalloproteinases (MMPs),¹⁵ which were vital for the migration in the cancer cells. Fiorentini et al. found that MMP2 and MMP9 were regulated by GPNMB protein. Besides, Rose et al. proved that GPNMB expression was an independent prognosis indicator of breast cancer and was related to MMP3 expression. Some studies also showed that extracellular signal–regulated protein kinase (ERK) pathway might be related to GPNMB. In this study, knocking down GPNMB downregulated the expression of MMP2 and MMP9, suggesting that GPNMB might affect cell migration into this pathway. Then, we investigated the β-catenin and GSK-3β expression after GPNMB knockdown; the results suggested that GPNMB affected cell migration within Wnt pathway.

Thus, GPNMB was overexpressed in the bladder cancer tissues and bladder cancer cell lines, which promotes the development of bladder cancer cells. GPNMB expression might be the new target for predicting prognosis and therapy.

Take home messages

• GPNMB was highly expressed in the bladder cancer tissues and cell lines.

• GPNMB could promote the biological behavior in bladder cancer cells, such as proliferation, migration, and invasiveness.

• GPNMB expression was correlated with the poor prognosis factors in bladder cancer.

The authors thank the Shenzhen People’s Hospital, Pathology Department, for assisting to evaluate the immunohistochemistry staining.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Y.Z. and Y.Y. guided all technical procedures and contributed to experimental design. C.Q. and X.Z. worked for the experimental conduct. Q.W. and C.Z. collected the data and did the data analysis. J.Y. contributed to article preparation. Y.Z. and C.Q. contributed equally to this work.

Funding

This work was supported by grants from the Shenzhen Commission of Science and Innovation program (no. JCYJ20150403101028172).