Upregulation of ITGBL1 predicts poor prognosis and promotes chemoresistance in ovarian cancer

Abstract

BACKGROUND:

Ovarian cancer remains one of the most lethal malignancies in women and the unfavorable prognosis and frequent recurrence are mainly due to the chemoresistance. However, the main mechanism underlying chemoresistance is still elusive.

OBJECTIVE:

To determine the role and biological function of ITGBL1 in ovarian cancer chemoresistance.

METHODS:

Immunohistochemical staining was used to determine the expression of ITGBL1 in ovarian cancer tissues. The association between ITGBL1 expression and clinicopathological features and survival was determined. Functional analysis including cell viability, apoptosis assays were performed after chemo drugs treatment to confirm the role of ITGBL1 in chemoresistance. In vivo tumor growth assay was used to detect the chemosensitivity of tumor cells. Western blot was used to detect the expression of indicated proteins.

RESULTS:

We noticed that ITGBL1 expression was significantly upregulated in ovarian cancer tissues compared to that in adjacent non-cancer tissues and high expression of ITGBL1 was significantly associated with lymph node invasion and advanced FIGO stage. More importantly, high ITGBL1 was an independent prognostic factor of ovarian cancer. Further experiments demonstrated that ITGBL1 promoted tumor cell resistant to chemo drugs both in vitro and in vivo. Mechanically, we found that ITGBL1 could activate PI3K/Akt signaling and using PI3K/Akt inhibitor could abrogate ITGBL1 induced chemoresistance.

CONCLUSIONS:

Our findings indicate that upregulation of ITGBL1 has important clinical significance and drives chemoresistance in ovarian cancer. Detection and depletion of ITGBL1 might be the potential approaches for diagnosis and therapy for ovarian cancer patients.

Keywords

ITGBL1 ovarian cancer PI3K/Akt chemoresistance prognosis

1. Introduction

Ovarian cancer (OC) is one of the most lethal disease with high mortality in women worldwide [1]. The overall 5-year survival of ovarian cancer patients is about 46% [2, 3]. However, patients in Stage I/II have a much higher survival rate than them in Stage III/IV (74% vs 27%), due to the frequent recurrence and chemoresistance [4, 5]. Recently, studies on the understanding of mechanisms underlying ovarian cancer progression and chemoresistance have identified several genetic and epigenetic alternations, such as genomic breakage, CCNE1 amplifications and MICU1-induced glycolysis [6, 7, 8]. However, the molecular regulation associated with ovarian cancer remains elusive. Hence, it is urgently needed to identify the features of ovarian cancer and new biomarkers for predicting the prognosis and therapy effects.

Integrin subunit beta-like 1 (ITGBL1) was first cloned and reported in 1999 and characterized by the integrin-like cysteine-rich repeat, and was belong to the EGF-like family [9]. It was also known as TIED (Ten integrin EGF-like repeat domain-containing protein) [10]. However, ITGBL1 has no canonical transmembrane domains or RGD motif, indicating the distinct regulation pattern of ITGBL1 [11]. ITGBL1 could be secreted and has been demonstrated to regulate several critical biological functions such as cell adhesion, migration and invasion [12, 13]. ITGBL1 is overexpressed in bone metastatic subclone cells compared to the parental cell, which is coincident with the fact that ITGBL1 is frequently co-expressed with genes that associated with bone remodeling and metastasis [14]. ITGBL1 also has peak expression in mid-gestational placenta [15]. Recently, ITGBL1 was demonstrated to promote cell migration through FAK/SRC signaling In ovarian cancer [16]. However, the prognostic value of ITGBL1 and its role in ovarian cancer chemoresistance are largely unknown.

In the present study, we explored the expression and clinical significance of ITGBL1 in ovarian cancer. We found that high expression of ITGBL1 could be an independent prognostic factor of ovarian cancer. Moreover, we noticed that ITGBL1 confer tumor cell chemoresistance both in vitro and in vivo via upregulating PI3K/Akt signaling. These results provided new insight to the ovarian cancer treatments.

2. Materials and methods

2.1 Ethics statements and clinical samples

The study was approved by the Ethics Committees of The Affiliated Hospital of Inner Mongolia Medical University. 210 cases of ovarian cancer specimens and corresponding adjacent non-cancer tissues were surgically removed and restored in Department of Obstetrics and Gynecology, The Affiliated Hospital of Inner Mongolia Medical University from Match 2006 to June 2013. Written informed consent was obtained from all patients involved in this study and none of the patients had received chemotherapy, radiotherapy or related therapies before surgery. The histological and clinical stages were classified according to FIGO [17].

2.2 Immunohistochemical staining (IHC)

Immunohistochemical staining was performed as described previously [18]. Briefly, the slides were deparaffinized in xylene and rehydrated in gradient ethanol. After retrieving the antigen in citrate buffer, the slides were incubated with 0.3% H ${}_{2}$ O ${}_{2}$ to block the endogenous peroxidase and then incubated with primary anti-ITGBL1 antibody (ab251678, Abcam Inc, USA, 1:200 dilution) at 4 ${{}^{\circ}}$ C overnight. Rabbit IgG was used as negative control. After washing with PBS, the slides were incubated with HRP-labeled secondary antibody (GTvision III, Gene Tech, China, no dilution) for 30 min. finally, the staining was visualized with DAB and counterstained with hematoxylin. All the slides were observed and photographed by phase contrast microscope (Zeiss, Oberkochen, Germany).

All slides were determined by two independent pathologists. To evaluate the results of immunohistochemical staining, semiquantitative score system for the staining intensity and positive cell percentage was used as described previously [18]. Briefly, the intensity was scored as 1 (no staining), 2 (weak), 3 (moderate), 4 (strong) and the percentage was scored as 1 (0–15%), 2 (15–50%), 3 (50–75%), 4 ( $>$ 75%). The final score was calculated by multiplying the intensity score and percentage score. A cut-off of less than 8 was evaluated as low expression of ITGBL1 [19].

2.3 Reagents, cell lines and transfection

Cisplatin (Cis, #S1166) and paclitaxel (Taxol, #S11 50) were purchased from Selleck Inc (Houston, TX, USA). Ovarian cancer cell lines (OV90, SKOV3, ES2, CAOV3 and OVCAR4) were came from the American Type Culture Collection (ATCC) and maintained according to the protocol.

Small interfering RNAs (siRNAs) targeting ITGBL1 were purchased from Genepharma (Shanghai, China). The sequences were as followed: 5 ${}^{\prime}$ -GAGACTGCGA CAAACATGA-3 ${}^{\prime}$ , 5 ${}^{\prime}$ -GAGCTGTCTATGACCGATA-3 ${}^{\prime}$ . Lentivirus containing shRNAs targeting ITGBL1 were purchased from Genecopoeia (Guangzhou,China) according to the siRNA sequences. The ITGBL1 cDNA vector was purchased from Genecopoeia (Guangzhou, China). Cell transfection was performed using Lipofectamine 2000 according to the manufacture’s protocol. Cells were collected for the further experiments after 48 hours transfection. For lentivirus infection, ES2 cells were plated into 6-well plates and infected with lentivirus and polybrene (5 $\mu$ g/ml), and stably knocking down cells were selected with puromycin (2 $\mu$ g/ml) for 2 weeks.

Figure 1.

ITGBL1 is upregulated in ovarian cancer and associated with shorter survival and recurrence. (A) The representative images of IHC staining in 210 cases of paired ovarian cancer tissues and non-cancer tissues (left panel), and statistical analysis (right panel). Bar, 100 $\mu$ m. The mRNA (B) and protein (C) expression of ITGBL1 in ovarian cancer cell lines. (D) The K-M plotter showed the association between ITGBL1 and overall survival.

2.4 Quantitative real-time PCR (qRT-PCR)

Total RNA was extracted from primary tumour and adjacent tumour tissues or cells using Trizol reagent. The first strand cDNA was synthesized by PrimeScript RT-PCR kit (Takara, Dalian, China) according to the manufacturer’s recommendation. Quantitative real-time PCR was performed using 7500 Real-time PCR system (Applied Biosystem, Foster City, CA, USA) with SYBR Green method as described previously [20]. GAPDH gene was used as internal control. The primers used were as follows: ITGBL1, forward, 5 ${}^{\prime}$ -GGTTGGCATGGAGATAAATGTGA-3 ${}^{\prime}$ , reverse, 5 ${}^{\prime}$ -CGGATCAACATCGTGACAGGTA-3 ${}^{\prime}$ ; GA PDH, forward, 5 ${}^{\prime}$ -AACGTGTCAGTOGTGGACCTG-3 ${}^{\prime}$ , reverse, 5 ${}^{\prime}$ -AGTGGGTGTCGCTGTFGAAGT-3 ${}^{\prime}$ . The 2 ${}^{-\Delta\Delta{\rm Ct}}$ method was used to measure the ITGBL1 expression.

2.5 Western blot

Total proteins from fresh tissues were lysed in RIPA buffer containing proteinase inhibitors and phosphatase inhibitors cocktail according to the manufacturer’s recommendation. The protein concentration was determined by BCA methods. 80 $\mu$ g of total protein was separated in 12% SDS-PAGE and then transferred to polyvinylidene fluoride membranes. The membranes were blocked with 5% BSA and incubated with anti-ITGBL1 antibody at 4 ${{}^{\circ}}$ C overnight. The membranes were then incubated with secondary antibody (anti-mouse or anti-rabbit IgG, 1:3000 dilution, Santa Cruz Biotechnology, USA) and visualized by ECL methods. GAPDH was used as internal control. The antibodies used were as followed: ITGBL1 (AV42196, Sigma-Aldrich, St. Louis, MO, USA); GAPDH (Cat. 5174, 1:2000 dilution), p-AKT (Ser473, Cat. 4060, 1:1000 dilution), AKT (Cat. 2920, 1:1000 dilution), p-GSK3 $\beta$ (Ser9, Cat. 5558, 1:1000 dilution), GSK3 $\beta$ (Cat#9832, 1:1000 dilution), p-FOX O3a (Ser253) (Cat#13129, 1:1000 dilution) and FOXO 3a (Cat#12829, 1:1000 dilution) were came from Cell Signaling Technology (Cambridge, MA, USA).

Figure 2.

Knocking down of ITGBL1 promotes chemodrugs induced apoptosis. A: The efficiency of siRNAs targeting ITGBL1 in CAOV3 and ES2 cells was examined by western blot. The expression value was calculated by ImageJ according to the density of each band. B: Cell apoptosis was examined by FACS after chemodrugs treatment. C: The viability of CAOV3 and ES2 knocking down of ITGBL1 cells and control cells after gradient Cis treatment. The IC50 value was calculated. D: The TUNEL assay of CAOV3 and ES2 knocking down of ITGBL1 cells and control cells after Cis treatment. ${}^{*}$ , $p<$ 0.05.

Table 1

Association between ITGBL1 expression and clinicopathologic parameters of ovarian cancer

Parameters		ITGBL1 expression		$\chi^{2}$	$P$
	Case number	Low	High
Overall	210	82	128
Age [34]
$\leqslant$ 50	103	37	66	0.592	0.442
$>$ 50	107	45	62
Histotype
Serous	117	50	67	1.620	0.655
Mucinous	46	16	30
Endometrioid	28	9	19
Other	19	7	12
Histologic grade
G1/G2	87	38	49	1.433	0.488
G3	86	30	56
Unknown	37	14	23
Tumor size (cm)
$\leqslant$ 5	33	16	17	1.032	0.310
$>$ 5	177	66	111
Lymph node invasion
None	135	65	70	12.105	0.001*
Yes	75	17	58
FIGO
I/II	73	39	34	8.814	0.003*
III/IV	137	43	94
CA125 (U/ml)
$\leqslant$ 600	87	39	48	1.691	0.193
$>$ 600	123	43	80

2.6 Cell viability assay

ITGBL1 overexpressing cells or knocking down cells with control cells were plated onto 96-well plates in the concentration of 1500 cells/well. Cells were treated with gradient concentration (1, 2, 4, 8, 16, 32, 64 $\mu$ g/ml) of Cis and cultured for 48 hours. Cell viability was determined by adding cell counting kit-8 (CCK-8, Boster Biotechnology, Wuhan, China) reagent according to the manufacturer’s protocol. The absorbance of each well was measured using microplate reader at 450 nm.

2.7 Cell apoptosis assay

Indicated cells were plated onto 6-well plates and then treated with 5 $\mu$ g/ml Cis or 100 nM Taxol and cultured for 24 hours. Apoptotic cells were determined by Annexin V-PI double staining and analyzed by flow cytometry in BD Calibur.

2.8 Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay

Cells were plated onto cover glass and treated with 5 $\mu$ g/ml Cis for 24 hours. TUNEL assay was performed using in situ cell death detection kit (Roche, Pleasanton, CA, USA) according to the manufacturer’s instruction. Nucleus was stained using DAPI. The number of TUNEL positive cells were counted in 5 randomly selected visions under microscope.

2.9 In vivo tumor growth assay

1 $\times$ 10 ${}^{6}$ ES2 cells with or without stably ITGBL1 knocking down were subcutaneously injected into the right flank of 6 nude mice (female, 6–8 weeks), respectively. After 2 weeks, we randomized the mice into four groups (shcon $+$ DMSO, shcon $+$ Cis, shITGBL1 $+$ DMSO, and shITGBL1 $+$ Cis), and each group had similar average tumor volume at about 100 mm ${}^{3}$ . Then we treated the mice for another 2 weeks and sacrificed them for analysis. Cis was intraperitoneal injected every three days (10 mg/kg). The tumor volume was calculated by the formula: L $\times$ W ${}^{2}$ /2. After 4 weeks, the mice were sacrificed and tumors were removed, formalin-fixed, paraffin-embedded for the immunohistochamical staining of Ki67.

2.10 Statistical analysis

Statistical analysis and graphical representations were performed using SPSS 21.0 (IBM, Chicago, IL, USA) and GraphPad Prism 5 (San Diego, CA, USA). ITGBL1 mRNA in tumour and adjacent non tumour tissues were compared using paired $t$ test. The associations between ITGBL1 expression and ovarian cancer clinicopathological features were measured by Chi-square test. The overall survival of patients was measured by Kaplan-Meier method with log-rank test. Cox proportional hazard regression model was employed to test the univariate and multivariate hazard ratios of the features. In vitro experiments were repeated at least three times and data were presented as mean $\pm$ SEM. Students’ $t$ test were used to compare the different group. A $p$ value less than 0.05 was considered statistically significant.

Table 2
Cox regression model for overall survival of ovarian cancer

Parameters	Univariate analysis				Multivariate analysis
	RR	95% CI	$P$		RR	95% CI	$P$
Age ( $\leqslant$ 50 vs. $>$ 50)	1.2	0.8–1.6	0.	427	0.9	0.6–1.3	0.	518
Histotype (serous vs. mucinous vs. endometrioid vs. other)	0.9	0.7–1.1	0.	272	1.0	0.8–1.2	0.	939
Histologic grade (G1/G2 vs. G3)	1.4	1.1–1.8	0.	011*	1.0	0.7–1.2	0.	711
Tumor size ( $\leqslant$ 5 vs. $>$ 5)	4.0	2.0–8.3	$<$ 0.	001*	3.4	1.6–7.2	0.	001*
Lymph node invasion (none vs. yes)	4.1	2.8–5.9	$<$ 0.	001*	2.8	1.9–4.2	$<$ 0.	001*
FIGO stage (I/II vs. III/IV)	4.3	2.6–7.1	$<$ 0.	001*	3.2	1.9–5.5	$<$ 0.	001*
CA125 (U/ml) ( $\leqslant$ 600 vs. $>$ 600)	1.5	1.0–2.2	0.	029*	1.3	0.9–1.9	0.	224
ITGBL1 expression (low vs. high)	2.6	1.7–4.0	$<$ 0.	001*	1.6	1.0–2.5	0.	036*

Figure 3.

Overexpression of ITGBL1 confer chemoresistance. A: The overexpression of ITGBL1 was examined in SKOV3 and OV90 cells. B: Cell apoptosis was examined by FACS after chemodrugs treatment. C: The viability of SKOV3 and OV90 overexpressing ITGBL1 cells and control cells after gradient Cis treatment. The IC50 value was calculated. D: The TUNEL assay of SKOV3 and OV90 overexpressing ITGBL1 cells and control cells after Cis treatment. E: Cell apoptosis was examined by FACS after chemodrugs treatment in the presence of gradient recombinant ITGBL1 (rITGBL1). F: The TUNEL assay of SKOV3 and OV90 treated with gradient recombinant ITGBL1 in the presence of Cis. ${}^{*}$ , $p<$ 0.05, ${}^{**}$ , $p<$ 0.01.

3. Results

3.1 Upregulation of ITGBL1 is associated with poor prognosis of ovarian cancer

To investigate the expression of ITGBL1, we performed IHC in 210 cases of ovarian cancer tissues and adjacent non-cancer tissues and found that ITGBL1 was predominant upregulated in cancer tissues (Fig. 1A). We also detected the mRNA and protein levels of ITGBL1 in ovarian cancer cell lines (Fig. 1B and C). Then, we determined the correlation between ITGBL1 expression and clinicopathological features in ovarian cancer. Among the 210 cases, the high expression of ITGBL1 was significantly correlated with the lymph node invasion and FIGO stages (Table 1). Whereas no significant association was observed between ITGBL1 expression and other features such as age, or histotype, or grade, or tumor size or CA125 levels.

To demonstrate the prognostic value of ITGBL1 in ovarian cancer, Kaplan-Meier plot with log-rank test in 210 cases was performed and we noticed that high expression of ITGBL1 indicated shorter overall survival (Fig. 1D). Furthermore, multivariate analysis indicated that ITGBL1, lymph node invasion and FIGO stage could be the independent prognostic factors of ovarian cancer (Table 2).

Figure 4.

Knocking down of ITGBL1 enhances Cisplatin treatment in vivo. A: Four different groups of tumors were shown (shcon, shcon $+$ DMSO; shRNAs, shITGBL1 $+$ DMSO; Cis, shcon $+$ Cis; shRNAs $+$ Cis, shITGBL1 $+$ Cis). The tumor volume (B) and weight (C) of each group. D: The Ki67 staining of tumor slides from each group. Bar, 100 $\mu$ m. ${}^{*}$ , $p<$ 0.05, ${}^{**}$ , $p<$ 0.01, ${}^{***}$ , $p<$ 0.001.

3.2 Downregulation of ITGBL1 enhances the efficacy of chemotherapy in ovarian cancer cells

We employed two siRNAs to knock down the expression of ITGBL1 in CAOV3 and ES2 (Fig. 2A). To detect the role of ITGBL1 in ovarian cancer chemoresistance, we examined the effects of two first-line drugs, Cisplatin (Cis) and paclitaxel (Taxol), on ITGBL1 knocking down cells. We found that ITGBL1 knocking down significantly increased chemodrugs induced apoptosis as evaluated by flow cytometry in CAOV3 and ES2 cells (Fig. 2B). Then, we treated cells with Cis in ITGBL1 knockdown and control cells with a gradient of concentrations to evaluate cell viability. As shown in Fig. 2C, compared with control cells, ITGBL1 knocking down significantly decreased the viability of ovarian cancer cells and increased the chemo-sensitivity (details were shown in Table 3). Furthermore, we performed TUNEL assay after Cis treatment in CAOV3 and ES2 ITGBL1 knocking down cells and control cells and noticed that ITGBL1 knocking down enhanced the percentage of TUNEL positive cells (Fig. 2D). Taken together, the results indicated that downregulation of ITGBL1 promoted chemodrugs induced apoptosis.

3.3 Overexpression of ITGBL1 confer cells cisplatin and taxol resistance

Next, we overexpressed ITGBL1 in SKOV3 and OV90 cells which had a relative low endogenous expression of ITGBL1 (Fig. 3A). Of note, we treated cells with Cis and Taxol and found that ITGB1L overexpression significantly decreased the apoptosis rate (Fig. 3B). The IC50 values of ITGBL1 overexpression cells were also higher than that of control cells (Fig. 3C and Table 3), demonstrated by the viability assay. In parallel, TUNEL assay indicated similar results as well (Fig. 3D).

Table 3
IC50 values of Cisplatin

Cell lines	IC50 ( $\mu$ M)
CAOV3-NC	33.0 $\pm$ 1.2
CAOV3-si1	17.7 $\pm$ 1.1
CAOV3-si2	13.2 $\pm$ 1.1
ES2-NC	16.3 $\pm$ 1.2
ES2-si1	9.1 $\pm$ 1.2
ES2-si2	9.6 $\pm$ 1.1
SKOV3-vector	6.9 $\pm$ 1.1
SKOV3-ITGBL1	14.8 $\pm$ 1.1
OV90-vector	12.8 $\pm$ 1.1
OV90-ITGBL1	39.0 $\pm$ 1.1

Figure 5.

ITGBL1 promotes chemoresistance via activating PI3K/Akt signaling. (A) Detection of PI3K/Akt signaling activation and downstream targets after ITGBL1 knocking down or overexpression. (B) Knocking down of ITGBL1 suppressed p-AKT expression while rITGBL1 restored PI3K/Akt activation. (C) LY294002 inhibited rITGBL1 induced PI3K/Akt activation. (D, E) LY294002 abrogated rITGBL1 induced chemoresistance in SKOV3 (D) and OV90 (E) cells treated with Cis or Taxol. ${}^{*}$ , $p<$ 0.05 (compared to rITGBL1-/LY294002-group), ${}^{\#}$ , $p<$ 0.05 (compared to rITGBL1 $+$ /LY294002-group).

As ITGBL1 could be secreted, we performed apoptosis assay using recombinant ITGBL1 (rITGBL1) and chemodrugs. When treated with Cis or Taxol, the cell apoptosis rate was decreased along with the increased rITGBL1 concentration (Fig. 3E). The results of TUNEL assay were consistent with ITGBL1 overexpression (Fig. 3F). Taken together, the data indicated that ITGBL1 increased ovarian cancer cells chemoresistance.

3.4 Knocking down of ITGBL1 sensitives tumor cells against Cisplain in vivo

We further examined the tumor growth and chemo-resistance in vivo. ES2 cells with or without knocking down of ITGBL1 were s.c. injected into nude mice. After 2 weeks, we randomized separated mice and half of them in each group were treated with Cis. After 4 weeks, mice were sacrificed and we found that knocking down of ITGBL1 or treated with Cis only, suppressed tumor growth. But compared with control group, it was not statistically significant. However, combination of knocking down of ITGBL1 and Cis, the tumors were significantly suppressed (Fig. 4A and B). Also, the tumor weight had the same trend (Fig. 4C). Meanwhile, we performed IHC of Ki67, and indicated that knocking down of ITGB1L plus Cis treatment significantly abrogated tumor growth in vivo (Fig. 4D).

3.5 ITGBL1 induces chemoresistance via upregulating PI3K/Akt signaling

To explore the mechanism of ITGBL1 induced chemoresistance, we detected the PI3K/Akt signaling which had pivotal role on tumor proliferation, chemoresistance and metastasis. Downregulation of ITGBL1 significantly decreased the activation of PI3K/ Akt evidenced by the suppression of p-AKT and the downstream factors p-GSK3 $\beta$ and p-FOXO3a. However, overexpression of ITGBL1 had the opposite effects (Fig. 5A). Moreover, rITGBL1 reversed the suppression of PI3K/Akt signaling which was caused by ITGBL1 knocking down (Fig. 5B). Then, we treated cells with rITGBL1 and LY294002, which was a broad-spectrum inhibitor of PI3K, and found that the activation of PI3K/Akt signaling was induced by rITGBL1, while abrogated followed by LY294002 treatment (Fig. 5C).

To further examine whether PI3K/Akt signaling plays a role in chemoresistance induced by ITGBL1, we performed the apoptosis assay. Using LY294002, we noticed that the apoptosis was decreased after rITGBL1 treatment in the presence of Cis or Taxol, while restored by inhibition of PI3K/Akt signaling in SKOV3 and OV90 cells (Fig. 5D and E). Taken together, the data indicated that PI3K/Akt signaling was involved in ITGBL1 induced chemoresistance.

4. Discussion

Although ITGBL1 has been implied to be highly expressed in ovarian cancer and regulate cancer cell migration and adhesion, this is the first study to examine the role of ITGBL1 protein in ovarian cancer prognosis and chemoresistance [21]. Ovarian cancer is one of the most fatal gynecological malignancies and the chemoresistance is frequently occurred [22, 23]. A majority of patients relapses within 2–3 years, and becomes resistant to chemotherapy [24]. Therefore, our study suggested that the secreted protein, ITGBL1 might contribute to the malignancy, and was identified as an independent prognostic factor for ovarian cancer.

ITGBL1 has been reported to upregulated in a variety types of cancers, such as colorectal cancer, breast cancer and esophageal cancer [25, 26]. However, Gan et al. reported that ITGBL1 was epigenetic downregulated by miR-576-5p and inhibited Wnt signal in non-small cell lung cancer, suggesting its heterogeneous roles [27]. In addition, ITGBL1 was observed as predictor gene for cirrhosis risk in chronic hepatitis B patients [28]. In current study, we performed immunohistochemistry in 210 paired ovarian cancer samples and adjacent non-cancer tissues, together with detecting protein and mRNA expression of ITGBL1 in fresh tissues. We found that ITGBL1 expression was significantly increased in ovarian cancer. Higher expression of ITGBL1 was associated with advanced FIGO stage and lymph node invasion of patients. Importantly, ITGBL1 expression was identified as an independent prognostic factor for ovarian cancer. Given the fact that ITGBL1 could be secreted, further investigation will be focused on detecting ITGBL1 in serum as a biomarker for predicting ovarian cancer patients’ chemotherapy response, recurrence and prognosis.

Previous studies have shown that ITGBL1 was involved in epithelial-mesenchymal transition (EMT), migration and metastasis. By integrating Oncomine, GEO databases, Gene Set Enrichment Analysis of multiple independent datasets revealed that gastric cancer specimens with high expression of ITGBL1 had elevated expression of EMT genes [12]. Also, Sun et al. indicated that ITGBL1 promoted cell migration and adhesion through the Wnt/PCP and FAK/SRC signaling pathways [16]. Li et al. found that RUNX2 could transcriptional enhance ITGBL1 expression, subsequently activate TGF- $\beta$ signaling [11]. In addition, ITGBL1 mediated TGF- $\beta$ signaling was also a key regulator of HBV-related liver fibrogenesis [29]. Here, we found that the knocking down of ITGBL1 increased Cis and Taxol induced apoptosis and reduced cell viability as demonstrated by flow cytometry, IC50 value and TUNEL assay. On the contrary, not only ectopic expression of ITGBL1 in cells, but using recombinant ITGBL1 could confer chemoresistance to cancer cells. Also, in vivo xenograft transplantation experiment indicated that knocking down of ITGBL1 had a little effect on suppressing tumor growth, as well as Cis, suggesting tumor cells had the potential to resistant Cis. However, combination with Cis and knocking down of ITGBL1 could significantly abrogate tumor growth. Further study would focus on detecting the combination of inhibiting ITGBL1 secretion and chemotherapy. Mechanically, we noticed that ITGBL1 activated PI3K/Akt signaling, as well as the downstream GSK3 $\beta$ , FOXO3a [30]. Above all, PI3K/Akt signaling has critical role in cell migration, metastasis, apoptosis and stem-like phenotype [31, 32, 33]. These observations prompted us to explore that whether PI3K/Akt inhibitors could rescue ITGBL1-induced chemoresistance. Indeed, employing LY294002 increased cell apoptosis which was suppressed by ITGBL1 in the present of chemodrugs. The result gave insight into restoring cancer cell sensitivity by depletion of ITGBL1 or using PI3K/Akt inhibitor with chemodrugs.

In conclusion, our results for the first time, indicated that ITGBL1 played an essential role in chemoresistance in ovarian cancer. Upregulation of ITGBL1 activated PI3K/Akt signaling, thereby contributing to malignancy and poor prognosis. Our findings suggested that ITGBL1 was a potential biomarker for chemotherapy and provide new clinical target for ovarian cancer.

References

Bray

Ferlay

Soerjomataram

Siegel

R.L.

Torre

L.A.

and Jemal

, Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries, CA Cancer J Clin 68 (2018), 394–424.

Park

Blackburn

B.E.

Rowe

Snyder

Wan

Deshmukh

Newman

Fraser

Smith

Herget

Burt

Werner

Gaffney

D.K.

Lopez

A.M.

Mooney

and Hashibe

, Rural-metropolitan disparities in ovarian cancer survival: A statewide population-based study, Ann Epidemiol 28 (2018), 377–384.

Jacobs

I.J.

Menon

Ryan

Gentry-Maharaj

Burnell

Kalsi

J.K.

Amso

N.N.

Apostolidou

Benjamin

Cruickshank

Crump

D.N.

Davies

S.K.

Dawnay

Dobbs

Fletcher

Ford

Godfrey

Gunu

Habib

Hallett

Herod

Jenkins

Karpinskyj

Leeson

Lewis

S.J.

Liston

W.R.

Lopes

Mould

Murdoch

Oram

Rabideau

D.J.

Reynolds

Scott

Seif

M.W.

Sharma

Singh

Taylor

Warburton

Widschwendter

Williamson

Woolas

Fallowfield

McGuire

A.J.

Campbell

Parmar

and Skates

S.J.

, Ovarian cancer screening and mortality in the UK collaborative trial of ovarian cancer screening (UKCTOCS): A randomised controlled trial, Lancet 387 (2016), 945–956.

Mathieu

K.B.

Bedi

D.G.

Thrower

S.L.

Qayyum

and Bast

R.C.

, Jr., Screening for ovarian cancer: Imaging challenges and opportunities for improvement, Ultrasound Obstet Gynecol 51 (2018), 293–303.

Ledermann

J.A.

Raja

F.A.

Fotopoulou

Gonzalez-Martin

Colombo

Sessa

and Group

E.G.W.

, Newly diagnosed and relapsed epithelial ovarian carcinoma: ESMO clinical practice guidelines for diagnosis, treatment and follow-up, Annals of Oncology: Official Journal of the European Society for Medical Oncology/ESMO 29 (2018), iv259.

Patch

A.M.

Christie

E.L.

Etemadmoghadam

Garsed

D.W.

George

Fereday

Nones

Cowin

Alsop

Bailey

P.J.

Kassahn

K.S.

Newell

Quinn

M.C.

Kazakoff

Quek

Wilhelm-Benartzi

Curry

Leong

H.S.

, G. Australian Ovarian Cancer Study Hamilton

Mileshkin

Au-Yeung

Kennedy

Hung

Chiew

Y.E.

Harnett

Friedlander

Quinn

Pyman

Cordner

O’Brien

Leditschke

Young

Strachan

Waring

Azar

Mitchell

Traficante

Hendley

Thorne

Shackleton

Miller

D.K.

Arnau

G.M.

Tothill

R.W.

Holloway

T.P.

Semple

Harliwong

Nourse

Nourbakhsh

Manning

Idrisoglu

Bruxner

T.J.

Christ

A.N.

Poudel

Holmes

Anderson

Leonard

Lonie

Hall

Wood

Taylor

D.F.

Fink

J.L.

Waddell

Drapkin

Stronach

Gabra

Brown

Jewell

Nagaraj

S.H.

Markham

Wilson

P.J.

Ellul

McNally

Doyle

M.A.

Vedururu

Stewart

Lengyel

Pearson

J.V.

Waddell

deFazio

Grimmond

S.M.

and Bowtell

D.D.

, Whole-genome characterization of chemoresistant ovarian cancer, Nature 521 (2015), 489–494.

Chakraborty

P.K.

Mustafi

S.B.

Xiong

Dwivedi

S.K.D.

Nesin

Saha

Zhang

Dhanasekaran

Jayaraman

Mannel

Moore

McMeekin

Yang

Zuna

Ding

Tsiokas

Bhattacharya

and Mukherjee

, MICU1 drives glycolysis and chemoresistance in ovarian cancer, Nat Commun 8 (2017), 14634.

Zhang

Liu

Zhang

Payne

S.H.

Zhang

McDermott

J.E.

Zhou

J.Y.

Petyuk

V.A.

Chen

Ray

Sun

Yang

Chen

Wang

Shah

Cha

S.W.

Aiyetan

Woo

Tian

Gritsenko

M.A.

Clauss

T.R.

Choi

Monroe

M.E.

Thomas

Nie

Moore

R.J.

K.H.

Tabb

D.L.

Fenyo

Bafna

Wang

Rodriguez

Boja

E.S.

Hiltke

Rivers

R.C.

Sokoll

Zhu

Shih

I.M.

Cope

Pandey

Zhang

Snyder

M.P.

Levine

D.A.

Smith

R.D.

Chan

D.W.

Rodland

K.D.

and Investigators

, Integrated proteogenomic characterization of human high-grade serous ovarian cancer, Cell 166 (2016), 755–765.

Takagi

Beglova

Yalamanchili

Blacklow

S.C.

and Springer

T.A.

, Definition of EGF-like, closely interacting modules that bear activation epitopes in integrin beta subunits, Proc Natl Acad Sci USA 98 (2001), 11175–11180.

10.

Berg

R.W.

Leung

Gough

Morris

Yao

W.P.

Wang

S.X.

and Krissansen

G.W.

, Cloning and characterization of a novel beta integrin-related cDNA coding for the protein TIED (“ten beta integrin EGF-like repeat domains") that maps to chromosome band 13q33: A divergent stand-alone integrin stalk structure, Genomics 56 (1999), 169–178.

11.

X.Q.

D.M.

Kong

P.Z.

Sun

Liu

P.F.

Wang

Q.S.

and Feng

Y.M.

, ITGBL1 Is a Runx2 transcriptional target and promotes breast cancer bone metastasis by activating the TGFbeta signaling pathway, Cancer Res 75 (2015), 3302–3313.

12.

Zhuang

Jiang

Zhao

Cao

Zhang

Chen

, ITGBL1 predicts a poor prognosis and correlates EMT phenotype in gastric cancer, Journal of Cancer 8 (2017), 3764–3773.

13.

Matsuyama

Ishikawa

Takahashi

Yamada

Yasuno

Kawano

Uetake

and Goel

, Abstract 1971: ITGBL1 is a novel epithelial mesenchymal transition-associated prognostic biomarker in colorectal cancer, Cancer Research 77 (2017), 1971–1971.

14.

Garcia

Jackson

Bachelier

Clement-Lacroix

Baron

Clezardin

and Pujuguet

, A convenient clinically relevant model of human breast cancer bone metastasis, Clin Exp Metastasis 25 (2008), 33–42.

15.

Uuskula

Mannik

Rull

Minajeva

Koks

Vaas

Teesalu

Reimand

and Laan

, Mid-gestational gene expression profile in placenta and link to pregnancy complications, PLoS One 7 (2012), e49248

16.

Sun

Wang

Zhang

and Zhang

, Extracellular matrix protein ITGBL1 promotes ovarian cancer cell migration and adhesion through Wnt/PCP signaling and FAK/SRC pathway, Biomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie 81 (2016), 145–151.

17.

Prat

and Oncology

F.C.O.G.

, FIGO’s staging classification for cancer of the ovary, fallopian tube, and peritoneum: Abridged republication, J Gynecol Oncol 26 (2015), 87–89.

18.

Qian

Xia

and Feng

, Sox9 mediated transcriptional activation of FOXK2 is critical for colorectal cancer cells proliferation, Biochem Biophys Res Commun 483 (2017), 475–481.

19.

Luo

Hou

Chen

Wang

Yang

Liu

Chen

Xiong

Wang

Fan

and Xu

, High expression of NDRG3 associates with unfavorable overall survival in non-small cell lung cancer, Cancer Biomarkers: Section A of Disease Markers 21 (2018), 461–469.

20.

Rao

Huang

Zhou

and Lin

, An improvement of the 2(-Δ⁢Δ⁢CT) method for quantitative real-time polymerase chain reaction data analysis, Biostat Bioinforma Biomath 3 (2013), 71–85.

21.

Lisowska

K.M.

Olbryt

Student

Kujawa

K.A.

Cortez

A.J.

Simek

Dansonka-Mieszkowska

Rzepecka

I.K.

Tudrej

and Kupryjanczyk

, Unsupervised analysis reveals two molecular subgroups of serous ovarian cancer with distinct gene expression profiles and survival, J Cancer Res Clin Oncol 142 (2016), 1239–1252.

22.

Jung

J.G.

Shih

I.M.

Park

J.T.

Gerry

Kim

T.H.

Ayhan

Handschuh

Davidson

Fader

A.N.

Selleri

and Wang

T.L.

, Ovarian cancer chemoresistance relies on the stem cell reprogramming factor PBX1, Cancer Res 76 (2016), 6351–6361.

23.

Ween

M.P.

Armstrong

M.A.

Oehler

M.K.

and Ricciardelli

, The role of ABC transporters in ovarian cancer progression and chemoresistance, Crit Rev Oncol Hematol 96 (2015), 220–256.

24.

Agarwal

and Kaye

S.B.

, Ovarian cancer: Strategies for overcoming resistance to chemotherapy, Nat Rev Cancer 3 (2003), 502–516.

25.

Qiu

Feng

J.R.

Qiu

Liu

Xie

Zhang

Y.P.

Liu

and Zhao

, ITGBL1 promotes migration, invasion and predicts a poor prognosis in colorectal cancer, Biomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie 104 (2018), 172–180.

26.

Miller

C.T.

Lin

Casper

A.M.

Lim

Thomas

D.G.

Orringer

M.B.

Chang

A.C.

Chambers

A.F.

Giordano

T.J.

Glover

T.W.

and Beer

D.G.

, Genomic amplification of MET with boundaries within fragile site FRA7G and upregulation of MET pathways in esophageal adenocarcinoma, Oncogene 25 (2006), 409–418.

27.

Gan

Liu

Tong

and Zhou

, Epigenetic downregulated ITGBL1 promotes non-small cell lung cancer cell invasion through Wnt/PCP signaling, Tumour biology: The journal of the International Society for Oncodevelopmental Biology and Medicine 37 (2016), 1663–1669.

28.

M.Y.

Xiao

C.Y.

and Lu

L.G.

, A 6 gene signature identifies the risk of developing cirrhosis in patients with chronic hepatitis B, Front Biosci (Landmark Ed) 21 (2016), 479–486.

29.

Wang

Gong

Zhang

Chen

Zhang

Han

Zhang

Chen

Zhang

Yuan

Huang

and Zhang

, Characterization of gene expression profiles in HBV-related liver fibrosis patients and identification of ITGBL1 as a key regulator of fibrogenesis, Scientific Reports 7 (2017), 43446

30.

Dey

Bharti

Dhanarajan

Das

Dey

K.K.

Kumar

B.N.

Sen

and Mandal

, Marine lipopeptide Iturin A inhibits Akt mediated GSK3beta and FoxO3a signaling and triggers apoptosis in breast cancer, Scientific Reports 5 (2015), 10316

31.

Mayer

I.A.

and Arteaga

C.L.

, The PI3K/AKT pathway as a target for cancer treatment, Annu Rev Med, 67 (2016), 11–28.

32.

Massacesi

Di Tomaso

Urban

Germa

Quadt

Trandafir

Aimone

Fretault

Dharan

Tavorath

and Hirawat

, PI3K inhibitors as new cancer therapeutics: Implications for clinical trial design, OncoTargets and Therapy 9 (2016), 203–210.

33.

Almozyan

Colak

Mansour

Alaiya

Al-Harazi

Qattan

Al-Mohanna

Al-Alwan

and Ghebeh

, PD-L1 promotes OCT4 and Nanog expression in breast cancer stem cells by sustaining PI3K/AKT pathway activation, International journal of cancer, Journal International du Cancer 141 (2017), 1402–1412.

34.

Tian

Lange

A.R.

Yearsley

Robertson

F.M.

and Barsky

S.H.

, The genesis and unique properties of the lymphovascular tumor embolus are because of calpain-regulated proteolysis of E-cadherin, Oncogene 32 (2013), 1702–1713.

Upregulation of ITGBL1 predicts poor prognosis and promotes chemoresistance in ovarian cancer

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Ethics statements and clinical samples

2.2 Immunohistochemical staining (IHC)

2.3 Reagents, cell lines and transfection

2.5 Western blot

2.7 Cell apoptosis assay

2.8 Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay

2.9 In vivo tumor growth assay

2.10 Statistical analysis

Table 2 Cox regression model for overall survival of ovarian cancer

3.1 Upregulation of ITGBL1 is associated with poor prognosis of ovarian cancer

3.3 Overexpression of ITGBL1 confer cells cisplatin and taxol resistance

Table 3 IC50 values of Cisplatin

3.5 ITGBL1 induces chemoresistance via upregulating PI3K/Akt signaling

4. Discussion

References

Table 2
Cox regression model for overall survival of ovarian cancer

Table 3
IC50 values of Cisplatin