467
Cancer Biomarkers 15 (2015) 467–475 DOI 10.3233/CBM-150474 IOS Press
Expression of BNIP3 and its correlations to hypoxia-induced autophagy and clinicopathological features in salivary adenoid cystic carcinoma Zhanwei Chen1, Haiwei Wu1 , Shengyun Huang1, Wengang Li, Shizhou Zhang, Peihui Zheng, Xiaoqing Zhou, Wenlei Liu and Dongsheng Zhang∗ Department of Oral and Maxillofacial Surgery, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, Shandong, China
Abstract. BACKGROUND: The expression of Bcl-2/adenovirus E1B 19 kDa-interacting protein3 (BNIP3) has been explored in many human malignancies, but not in adenoid cystic carcinoma (ACC). OBJECTIVE: This study investigated the clinical significance of expression of BNIP3 in ACC tissues and cells and elucidated its correlations to hypoxia-induced autophagy. METHODS: Immunohistochemical and immunofluorescence staining were used to explore BNIP3, HIF-1α and LC3 expression. RESULTS: BNIP3 was positively expressed in 41 cases (63.1%), and was significantly correlated with histological grade (P = 0.001). HIF-1α was positively expressed in 52 cases (80.0%) and was significantly correlated with TNM stage (P = 0.023) and histological grade (P = 0.024). LC3 was positively expressed in 37 cases (56.9%) and was significantly correlated with TNM stage (P = 0.019). The expression of BNIP3 was correlated with HIF-1α expression (P = 0.011). The overall survival in the negative BNIP3 expression group tended to be better than in the positive BNIP3 expression (P = 0.011). In vitro experiment, BNIP3 immunofluorescence staining was detected in cells treated with CoCl2 (for hypoxic condition). CONCLUSIONS: The data indicated that BNIP3 plays a vital role in the tumorigenesis of adenoid cystic carcinoma and could be a new target for gene therapy of adenoid cystic carcinoma. Keywords: Adenoid cystic carcinoma, hypoxia, BNIP3, HIF-1α, autophagy
1. Introduction Adenoid cystic carcinoma (ACC) is one of the most frequent malignant tumors of the salivary glands in the head and neck especially in the minor salivary glands. 1 These
authors contributed equally to this study. author: Dongsheng Zhang, Proffessor, Dean, Department of Oral and Maxillofacial Surgery, Shandong Provincial Hospital Affiliated to Shandong University, 324 Jingwu Road, Jinan 250021, China. Tel.: +86 531 68776950; Fax: +86 531 87035697; E-mail: [email protected]. ∗ Corresponding
Clinically, it is described as one of the most destroyed and unpredicted malignant tumors because its propensity of perineural invasion, multiple recurrences and a high incidence of distant metastases [1]. Currently, surgery is the most effective treatment, which is usually followed by adjuvant radiotherapy including implantation of Iodine-125 seeds [2]. Although the 5-year survival rate is relatively high, the 10-year and longer survival rate remains low [3]. Therefore, further understanding of the characteristic biological behavior and molecular genetics of ACC may provide new insights into the novel treatment of the disease.
c 2015 – IOS Press and the authors. All rights reserved ISSN 1574-0153/15/$35.00
468
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Hypoxia is a common feature of solid tumors which occurs when the proliferation of the tumor outstrips the speed of blood vessels formation. Clinically, hypoxia may be deleterious for the tumor control by increasing resistance to radiation, preventing drug delivery through lack of blood vessels, and adaptation to the microenvironment through genomic instability. The transcriptional factor hypoxia-inducible factor 1α (HIF1α), the key regulator of the hypoxic response, plays an essential role in the adaptive response of cells to reduced oxygen tension and undergoes conformational changes in response to varying oxygen concentrations. Under normoxic conditions HIF-1α is hydroxylated, followed by binding to Von Hippel Lindau (VHL) protein and proteosomal degradation; on the contrary, this hydroxylation does not occur under hypoxic conditions [4]. The presence of HIF-1α has been observed in the majority of the common human cancers, such as head and neck cancer, ovarian cancer and oesophageal cancer [5–7], which is associated with poor prognosis and resistance to therapy. Costa et al found that ordinary as well as high-grade transformation adenoid cystic carcinoma exhibited HIF-1a expression, whereas it was observed only occasionally in normal salivary tissue adjacent to the tumor [8]. Bcl-2/adenovirus E1B 19 kDa-interacting protein3 (BNIP3), a member of BH3-only subfamily of Bcl-2 superfamily, belongs to pro-apoptotic proteins. The expression of BNIP3 is regulated by hypoxia and other factors. BNIP3 protein is localized on the outside of the mitochondrial membrane, and once stimulated, it integrates into the mitochondrial membrane, leading to permeabilization of the transition pore and decreasing the mitochondrial membrane potential resulting in chromatin condensation, DNA-fragmentation and finally apoptosis [9,10]. Studies have shown that abnormal expression of BNIP3 is correlated with the growth progression and drug resistance of many tumors in recent years [11]. BNIP3 is closely related to tumor, and the expression of BNIP3 in different tumor tissue is not the same and represents different prognosis. Recent studies have shown that BNIP3 is also closely related to autophagy. BNIP3, which is upregulated by the transcription factor HIF-1α, plays a role in hypoxiainduced autophagic cell death. In our previous study, we found that autophagy plays an important role in tumor progression and positive LC3 expression showed a statistically significant correlation with absent lymph node involvement and TNM stage in ACC [12]. However, the role of BNIP3 in hypoxia-induced autophagy is still extensively unexplored in ACC. Furthermore,
the relationship between expression of BNIP3 and clinicopathologic features of tumors has not been studied in salivary adenoid cystic carcinoma tissue. In our study, we investigated the expression of BNIP3, HIF-1α and LC3 by immunohistochemical analysis in head and neck salivary adenoid cystic carcinoma and normal salivary tissue. We also correlated BNIP3, HIF-1α and LC3 expression with histopathogic prognostic factors and survival. Additionally, we observed the expression of BNIP3, HIF1α and LC3 in ACC cells under hypoxic condition.
2. Material and methods 2.1. Patients and specimens All sixty-five specimens of head and neck adenoid cystic carcinoma, primary surgically resected between December 2002 and March 2011, were derived from the archives of Shandong Provincial Hospital, affiliated to Shandong University of China. Two independent pathologists reviewed all of the original hematoxylin and eosinstained sections to confirm the pathological diagnosis and to select a representative section for immunohistochemical study. Two or three representative cores with 3 mm in diameter, were obtained from a representative area of each paraffin-embedded tumor specimens. The median age of the patients was 48.2 years (range, 18–75 years). Tumor size ranged from 0.6 to 4.5 cm (median size, 1.9 cm). According to the American Joint Committee on Cancer staging criteria, 27 patients (41.6%) had stage I and II disease, and 38 patients (58.4%) had stage III and IV disease. Three histologic grades were confirmed: Grade I, tumors with tubular and cribriform areas but without solid forms; Grade II, cribriform tumors only or mixed with less than 30% of solid areas; and Grade III tumors with a dominantly solid pattern [13]. All patients were followed up by reviewing their medical records and telephoning. 2.2. Immunohistochemical staining Serial sections of 4 μm of paraffin embedded material of each specimen were deparaffinized in turpentine oil and rehydrated in graded alcohol. Antigen retrieval (AR) was achieved by boiling in a steamer immersed in citrate buffer (pH 6.0) for BNIP3 and LC3, while trisethylenediamine tetraacetic acid (EDTA) (pH 9.0) was used for HIF-1α. Endogenous peroxidase ac-
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
469
Fig. 1. Negative BNIP3 expression in normal salivary gland (A1), and positive BNIP3 expression in solid (A2), tubular (A3) and cribriform (A4) pattern of adenoid cystic carcinoma (A1–A4, 200×); Negative HIF-1α expression in normal salivary gland (B1), and positive HIF-1α expression in solid (B2), tubular (B3) and cribriform (B4) pattern of adenoid cystic carcinoma (B1–B4, 200×); Positive LC3 expression in normal salivary gland (C1), and in solid (C2), tubular (C3) and cribriform (C4) pattern of adenoid cystic carcinoma (C1–C4, 200×). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
tivity was blocked by 3% hydrogen peroxide for 30 min in the incubator. Thereafter slides were incubated for overnight at 4◦ C with primary monoclonal antibodies for LC3A/B (1:200, Abcam, Cambridge, United Kingdom), BNIP3 (1:200, Santa Cruz, USA) and HIF1a (1:50, Boster, Wuhan, China). Then slides were incubated with secondary antibodies for 30 min, using diaminobenzidine-tetrahydrochloride (DAB) as the substrate to detect antigen-antibody binding, and nuclei were counterstained with hematoxylin. The positive controls were provided by Abcam, and negative controls were obtained by omitting primary antibody replaced by PBS.
0 (< 10% of tumor cells), 1 (10–49% of tumor cells), 2 ( 50% of tumor cells). The intensity score and proportion score were multiplied to give a total score. The staining pattern of the slides was defined as follows: 0–3 (negative expression) and 4–5 (positive expression). Tumor cells were considered immunohistochemical staining positive for HIF-1α when stained in 30% or more of the cells. All slides were evaluated independently by two observers without clinicopathologic information.
2.4. Statistical methods
2.3. Evaluation of immunohistochemical staining The expression of BNIP3 ascertained by immunohistochemical staining in the cytoplasm and a positive signal was considered if cytoplasm stained brown. Although HIF-1α and LC3 were expressed in both the nuclei and cytoplasm in the cell of adenoid cystic carcinoma, we only evaluated the cell nucleus expression of HIF-1α and cytoplasmic expression of LC3. Immunohistochemical staining for BNIP3 and LC3 was evaluated in compliance with intensity and proportion. The intensities were scored as 0 (no staining), 1 (weak staining), 2 (moderate staining), and 3 (strong staining). The percentage of staining area was classified as
Statistical calculations were performed using the SPSS software package (version 19.0; SPSS, Chicago, IL, USA). The association was evaluated according to the Pearson chi-square test or Fisher’s exact test between BNIP3, HIF-1α and LC3 expression and several clinicopathologic variables. The Kaplan-Meier method was used to determine the probability of survival, and the data were analyzed by applying the log-rank test. Overall survival (OS) was defined as from the date of surgery to the date of death due to cancer. P value of less than 0.05 was considered statistically signifcant and all tests of statistical significance were two-sided.
470
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Table 1 Correlation between clinicopathologic variables and expression of BNIP3, HIF-1α and LC3 expression in 65 cases of head and neck adenoid cystic carcinoma Variables
BNIP3 expression P value HIF-1α expression P value LC3 expression P value Negative n(%) Positive n(%) Negative n(%) Positive n(%) Negative n(%) Positive n(%)
Age < 50 years 50 years Gender Female Male TNM stage I, II III,IV Tumor size 3 cm > 3 cm Histological grade Grade I, II Grade III Distant metastasis Absent Present ∗ Significantly
11 (16.9) 13 (20.0)
16 (24.6) 25 (38.5)
0.591
6 (9.2) 7 (10.8)
21 (32.3) 31 (47.7)
0.706
13 (20.0) 15 (23.1)
14 (21.5) 23 (35.4)
0.486
9 (13.8) 15 (23.1)
17 (26.1) 24 (37.0)
0.753
4 (6.2) 9 (13.8)
22 (33.8) 30 (46.2)
0.448
11 (16.9) 17 (26.2)
15 (23.1) 22 (33.8)
0.919
10 (15.4) 14 (21.5)
17 (26.1) 24 (37.0)
0.987
9 (13.8) 4 (6.2)
18 (27.7) 34 (52.3)
0.023∗
7 (10.8) 21 (32.3)
20 (30.8) 17 (26.1)
0.019∗
16 (24.6) 8 (12.3)
36 (55.4) 5 (7.7)
0.056
8 (12.3) 5 (7.7)
44 (67.7) 8 (12.3)
0.114
21 (32.3) 7 (10.8)
31 (47.7) 6 (9.2)
0.381
20 (30.7) 4 (6.2)
17 (26.1) 24 (37.0)
0.001∗
11 (16.9) 2 (3.1)
26 (40.0) 26 (40.0)
0.024∗
15 (23.1) 13 (20.0)
22 (33.8) 15 (23.1)
0.635
20 (30.7) 4 (6.2)
34 (52.3) 7 (10.8)
1.000
9 (13.8) 4 (6.2)
45 (69.2) 7 (10.8)
0.209
23 (35.4) 5 (7.7)
31 (47.7) 6 (9.2)
1.000
different by the Pearson chi-square test or Fisher exact test.
2.5. Immunofluorescence staining ACC-M Cells grown on sterile glass coverslips in 6-cm cell culture dishes were allowed to attach by overnight incubation, and then cultured with DMSO (Control) or 200 μM CoCl2 (for hypoxic condition) for indicated time. At the end of treatments, cells were washed with PBS, fixed with 4% paraformaldehyde for 20 min, and permeabilized with 0.5% Triton X-100 for 10 min. Then the cells were incubated with primary antibody (BNIP3, HIF-1α and LC3) for 4◦ C overnight., followed by treatment with FITC conjugated goat antirabbit IgG (1:50, Beijing Cwbio Co., Ltd) or TRITC conjugated goat anti-mouse IgG (1:50, Beijing Cwbio Co., Ltd) for 1 h. Then nuclei were counterstained with DAPI for 7 min, followed by observation under a Nikon fluorescence microscope.
3. Results 3.1. BNIP3 expression and its association with clinicopathological variables BNIP3 immunohistochemical staining was detected mainly in the membrane and cytoplasm of tumor cells, but not in the normal head and neck salivary glands cells (Fig. 1 A1). BNIP3 expression of adenoid cystic carcinoma was observed in cribriform, tubular, and solid histological patterns, and was positive present in
41 cases (63.1%) (Fig. 1 A2–A4). The correlations between BNIP3 and clinical variables are listed in Table 1. BNIP3 expression was significantly correlated with histological grade (P = 0.001). BNIP3 expression was not significantly associated with age (P = 0.591), gender (P = 0.753), TNM stage (P = 0.987), tumor size (P = 0.056), and distant metastasis (P = 1.000) (Table 1). 3.2. HIF-1α expression and its association with clinicopathological variables HIF-1α immunohistochemical staining was detected mainly in the nucleus of tumor cells, but rarely in the normal head and neck salivary glands cells (Fig. 1 B1). HIF-1α expression of adenoid cystic carcinoma was also observed in cribriform, tubular, and solid histological patterns, and was positive present in 52 cases (80.0%) (Fig. 1 B2–B4). The correlations between BNIP3 and clinical variables are listed in Table 1. HIF-1α expression was significantly correlated with TNM stage (P = 0.023) and histological grade (P = 0.024). HIF-1α expression was not significantly associated with age (P = 0.706), gender (P = 0.448), tumor size (P = 0.114) and distant metastasis (P = 0.209) (Table 1). 3.3. LC3 expression and its association with clinicopathological variables Consist with our previous study [12], positive LC3 was showed in normal head and neck gland. LC3 ex-
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
471
Table 2 Correlation between BNIP3 and HIF-1α expression in 65 cases of head and neck adenoid cystic carcinoma BNIP3 expression Negative n(%)
0.011∗
HIF-1α expression Negative Positive ∗ Significantly
P value Positive n(%)
9 (13.8) 15 (23.1)
4 (6.2) 37 (56.9)
different by the Fisher exact test. Table 3 Correlation between expression BNIP3 and LC3 expression in 65 cases of head and neck adenoid cystic carcinoma BNIP3 expression
LC3 expression Negative Positive
P value
Negative n(%)
Positive n(%)
13 (20.0) 11 (16.9)
15 (23.1) 26 (40.0)
0.167
pression was also found in the plasma membrane, cytoplasm, and/or nucleus of tumor cells (Fig. 1 C1). The LC3 expression of adenoid cystic carcinoma was also observed in cribriform, tubular, and solid histological patterns, and was positive present in 37 cases (56.9%) (Fig. 1 C2–C4). The correlations between LC3 and clinical variables are listed in Table 1. LC3 expression was significantly correlated with TNM stage (P = 0.019). LC3 expression was not significantly associated with age (P = 0.486), gender (P = 0.919), tumor size (P = 0.381), histological grade (P = 0.635) and distant metastasis (P = 1.000).
3.6. Expression of BNIP3 in ACC-M cells under hypoxic condition
3.4. Relationship between BNIP3 and HIF-1α expression, BNIP3 and LC3 expression
4. Discussion
As shown in Table 2, we found that BNIP3 expression was positively correlated with HIF-1α expression in head and neck adenoid cystic carcinoma (P = 0.011). By contrast, as shown in Table 3, BNIP3 expression was not significantly correlated with LC3 expression (P = 0.167). 3.5. Results of the overall survival analysis The mean follow-up period was 61.8 months (range, 18–115 months), and for 65 patients, 30 patients died of disease, 33 patients were alive on the day of the study and 2 patients were lost during the follow-up period. Based on the use of the Kaplan-Meier method and log-rank test, the overall survival in the negative BNIP3 expression group tended to be better than in the positive BNIP3 expression (P = 0.011) (Fig. 2); HIF1a expression showed no significant difference for the overall survival rate in patients with adenoid cystic carcinoma (P = 0.132) (Fig. 3).
In vitro study, BNIP3 immunofluorescence staining was detected in ACC-M cells treated with CoCl2 (for hypoxic condition) (Fig. 4). Additionally, HIF-1α and LC3 were also detected in ACC-M cells under hypoxia (Fig. 4). Furthermore, cytoplasmic puncta tethered with LC3 was detected in ACC-M cells, indicating the formation of autophagosome-like structure in ACC cells.
Our study evaluated that expression of BNIP3 and HIF-1α in salivary adenoid cystic carcinoma using immunohistochemical analysis and cell immunofluorescence. Furthermore, we demonstrated that expression of BNIP3 in salivary adenoid cystic carcinoma at the protein level in correlation with the hypoxic response and autophagy. We also correlated BNIP3, HIF-1α and LC3 expression with histopathogic prognostic factors and survival. BNIP3, a pro-apoptosis member of the Bcl-2 family, has been demonstrated in a wide variety of normal cells and in cells undergoing pathological hypoxic stress. Because the promoter of BNIP3 contains a hypoxia response element (HRE), BNIP3 appears to be a direct target for transcriptional activation by HIF1 [14]. The low expression of BNIP3 protein has been found in pancreatic cancer [15], colorectal and gastric cancer [16], and hematopoietic malignancies [17]. On the contrary, high levels of BNIP3 protein have been described in non-small cell lung cancer [18], basal cell carcinoma [19] and cholangiocarcinoma [20]. Re-
472
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Fig. 2. Cumulative Kaplan-Meier survival curve (BNIP3) showing that survival in the negative BNIP3 expression group (n=24) tended to be better than in the positive BNIP3 expression group (n=41) (P =0.011 , log-rank test). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
Fig. 3. Cumulative Kaplan-Meier survival curve (HIF-1α) showing that the survival in the negative HIF-1α expression group (n = 13) and the positive HIF-1α group (n = 52) was not significantly different (P = 0.132, log-rank test). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
473
Fig. 4. Positive HIF-1α expression in the nuclear of ACC-M cells; positive BNIP3 and LC3 expression in the cytoplasm of ACC-M cells. (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
cently it also has been shown that BNIP3 may have opposing functions at different stages of tumor progression in breast cancer [21]. Although expression of BNIP3 has been studied in a variety of tumors, little studies on BNIP3 expression in salivary adenoid cystic carcinoma have been published. To our knowledge, our study firstly evaluates the immunohistochemical expression of BNIP3 in head and neck adenoid cystic carcinoma compared to normal tissue. We demonstrated that 63.1% of patients expressed the BNIP3 protein in head and neck adenoid cystic carcinoma tissues while BNIP3 was not expressed in normal head and neck salivary gland tissues. Our study also demonstrated that positive expression of BNIP3 was significantly associated with the histological grade, and shorter overall survival in head and neck adenoid cystic carcinoma. These findings confirm that BNIP3 may play an important role in tumor progression, and BNIP3 could be a indicator of poor prognosis of head and neck adenoid cystic carcinoma. Hypoxia is a poor prognostic indicator for solid tumors due to increased resistance to radiation, restricted access to drug delivery through lack of blood vessels, and adaptation to the microenvironment through genomic instability [22]. One way in which cells respond to hypoxia is through hypoxia-inducible factor-1 (HIF1) which consists of α and β subunits. The stability of HIF-1 depends mainly on the α subunit [23]. Under normoxia, HIF-1α is degraded, whereas this process is inhibited leading to protein stabilization under hypoxia. Stabilized HIF-1α is transported into the nucleus where it heterodimerizes with HIF-1β to transactivate the expression of numerous hypoxiaresponse genes, including the pro-angiogenic vascular endothelial growth factor (VEGF) and glucose transporters [24]. In the literature, the correlation between levels of HIF-1α expression with tumor histological grade and biological behaviour is variable and seems to depend on the tumor type. In breast carcinoma, levels of HIF-1α have been reported to increase as the
degree of malignancy increases indicating its potential association with poor clinical outcome [25]. In contrast, in endometrial carcinoma, high HIF-1α was associated significantly with low-grade tumors [26]. In our study, we found that 80% of patients expressed the HIF-1α protein in head and neck adenoid cystic carcinoma tissues. This phenomenon also appeared in consist with Costa et al who found that ordinary as well as high-grade transformation adenoid cystic carcinoma exhibited HIF-1α expression [8]. Our study also demonstrated that positive expression of HIF-1α is significantly associated with the histological grade and TNM stage in head and neck adenoid cystic carcinoma. Based on the use of the Kaplan-Meier method and logrank test, HIF-1α expression showed no significant difference for the overall survival rate in patients with adenoid cystic carcinoma. These findings indicate that although hypoxia plays an important role in tumor progression, it could not be considered as a predictive indicator for prognosis in head and neck adenoid cystic carcinoma. Recent studies have shown that hypoxia can induce the autophagy through BNIP3. Hypoxia-induced BNIP3 plays an important role in hypoxia-induced autophagic cell death, and the BH-3 domain of BNIP3 is critical for this physiological process. In this progress, BH3 domains of BNIP3 were shown to compete with Beclin 1 for Bcl-2 binding, and Beclin 1 was released to induce autophagy [27]. Although hypoxia induces autophagy via BNIP3 as a pro-death mechanism, this progress has always been claimed to be a prosurvival mechanism promoting tumor progression by removing of potentially toxic damaged proteins and organelles [28]. Recently it has been shown that BNIP3 may have opposing functions at different stages of progression in breast cancer, which could be explained by its role in autophagy. In pre-invasive cancers, induction of autophagy through BNIP3 may provide with extra nutrients and promote further tumor progression. However at later stages, BNIP3 expression could pro-
474
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
mote prolonged autophagy in hypoxic cancer cells possibly through increased autophagosome-lysosome fusions, leading to autophagic cell death and better patient outcome [29]. Our results showed that BNIP3 expression was positively correlated with HIF-1a expression, while was not correlated with LC3 expression in head and neck adenoid cystic carcinoma. This suggested that BNIP3-induced autophagy may not play a dominant role in hypoxia-induced autophagy in ACC tissues. HIF-1α/BNIP3 pathway has been explored in several studies about hypoxia-induced autophagy, however, this pathway is not in charge of all hypoxiainduced autophagy. It’s reported that hypoxia-induced hypoxia is induced by HIF-1 dependent pathway and HIF-1 independent pathway, and HIF-1α/BNIP3 pathway is not involved in 5’ AMP activated protein kinase (AMPK) signaling pathway and unfolded protein response (UPR) pathway in the induction of autophagy under severe hypoxia [30,31]. In our in vitro experiment, up-regulation of HIF-1α and BNIP3 in ACC-M cells can obviously improve expression of autophagyrelated gene LC3, which indicates that BNIP3 was involved in hypoxia-induced autophagy in ACC. In conclusion, our study demonstrated expression of HIF-1α and BNIP3 in head and neck salivary adenoid cystic carcinoma and its relationship with clinicopathologic factors and patient survival. These results indicate that BNIP3 may play an important role in the tumorigenesis of salivary adenoid cystic carcinoma, and may be a new prognostic indicator of the outcome in patients with salivary adenoid cystic carcinoma. Further studies are needed to confirm our findings and to clarify the exact role of BNIP3 in the development of head and neck adenoid cystic carcinoma.
Acknowledgment This study is supported by Shandong Provincial Science and Technology Development Plan (No. 2006GG 2202034), the Science and Technology Foundation of Shandong Province (2006GG20002046) and Shandong Provincial Natural Science Foundation (ZR2012 HQ022).
Conflicts of interest The authors declare that there are no conflicts of interest in the research.
References [1]
Dodd RL, Slevin NJ, Salivary gland adenoid cystic carcinoma: A review of chemotherapy and molecular therapies, Oral Oncol 42 (2006), 759-769. [2] Zhang J1, Zhang JG, Song TL, Zhen L, Zhang Y, Zhang KH, Yang ZH, Yu GY, 125I seed implant brachytherapy-assisted surgery with preservation of the facial nerve for treatment of malignant parotid gland tumors, Int J Oral Maxillofac Surg 37 (2008), 515-520. [3] Spiro RH, Huvos AG, Strong EW, Adenoid cystic carcinoma of salivary origin: A clinicopathologic study of 242 cases, Am J Surg 128 (1974), 512-520. [4] Koop EA, van Laar T, van Wichen DF, de Weger RA, Wall Ev, van Diest PJ, Expression of BNIP3 in invasive breast cancer: correlations with the hypoxic response and clinicopathological features, BMC Cancer 9 (2009), 175. [5] Gruber G, Greiner RH, Hlushchuk R, et al, Hypoxia-inducible factor 1 alpha in high-risk breast cancer: an independent prognostic parameter? Breast Cancer Res 6 (2004), R191-R198. [6] Birner P, Schindl M, Obermair A, Breitenecker G,Oberhuber G, Expression of hypoxia-inducible factor 1 in epithelial ovarian tumors: Its impact on prognosis and on response to chemotherapy, Clin Cancer Res 7 (2001), 1661-1668. [7] Lu XG, Xing CG, Feng YZ, Chen J and Deng C, Clinical significance of immunohistochemical expression of hypoxiainducible factor-1 alpha as a prognostic marker in rectal adenocarcinoma, Clin Colorectal Cancer 5 (2006), 350-353. [8] Costa AF1, Tasso MG, Mariano FV, et al, Levels and patterns of expression of hypoxia-inducible factor-1α, vascular endothelial growth factor, glucose transporter-1 and CD105 in adenoid cystic carcinomas with high-grade transformation, Histopathology 60 (2012), 816-825. [9] Velde C Vande, Cizeau J, Dubik D, Alimonti J, Brown T, Israels S,Hakem R, Greenberg AH, BNIP3 and genetic control of necrosis-like cell death through the mitochondrial permeability transition pore, Mol Cell Biol 20 (2000), 5454-5468. [10] Imazu T, Shimizu S, Tagami S, Matsushima M, Nakamura Y, Miki T, Okuyama A, Tsujimoto Y, Bcl-2/E1B 19 kDainteracting protein 3-like protein (Bnip3L) interacts with bcl2/Bcl-xL and induces apoptosis by altering mitochondrial membrane permeability, Oncogene 18 (1999), 4523-4529. [11] Chinnadurai G1, Vijayalingam S, Gibson SB, BNIP3 subfamily BH3-only proteins: mitochondrial stress sensors in normal and pathological functions, Oncogene 1 (2008), S114-S127. [12] Jiang L, Huang S, Li W, Zhang D, Zhang S, Zhang W, Zheng P, Chen Z, Expression of autophagy and ER stress-related proteins in primary salivary adenoid cystic carcinoma, Pathol Res Pract 208 (2012), 635-641. [13] P.A.Szanto, M.A.Luna, M.E.Tortoledo, et al, Histologic grading of adenoid cystic carcinoma of the salivary glands, Cancer 54 (1984), 1062-1069. [14] Bruick RK, Expression of the gene encoding the proapoptotic Nip3 protein is induced by hypoxia, Proc Natl Acad Sci U S A 97 (2000), 9082-9087. [15] Okami J, Simeone DM, Logsdon CD, Silencing of the hypoxia-inducible cell death protein BNIP3 in pancreatic cancer, Cancer Res 64 (2004), 5338-5346. [16] Murai M, Toyota M, Suzuki H, Satoh A, Sasaki Y, Akino K, Ueno M,Takahashi F, Kusano M, Mita H, et al, Aberrant methylation and silencing of the BNIP3 gene in colorectal and gastric cancer, Clin Cancer Res 11 (2005), 1021-1027. [17] Murai M, Toyota M, Satoh A, Suzuki H, Akino K, Mita H, Sasaki Y, Ishida T, Shen L, Garcia-Manero G, Aberrant DNA
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy methylation associated with silencing BNIP3 gene expression in haematopoietic tumors, Br J Cancer 92 (2005), 1165-1172. [18] Giatromanolaki A, Koukourakis MI, Sowter HM, Sivridis E, Gibson S,Gatter KC, Harris AL, BNIP3 expression is linked with hypoxia-regulated protein expression and with poor prognosis in non-small cell lung cancer, Clin Cancer Res 10 (2004), 5566-5571. [19] Brinkhuizen T, Weijzen CA, Eben J, Thissen MR, van Marion AM, Lohman BG, Winnepenninckx VJ, Nelemans PJ, van Steensel MA, Immunohistochemical analysis of the mechanistic target of rapamycin and hypoxia signalling pathways in basal cell carcinoma and trichoepithelioma, Plos One (2014), doi: 10.1371/journal.pone.0106427. [20] Thongchot S, Yongvanit P, Loilome W, Seubwai W, Phunicom K, Tassaneeyakul W, Pairojkul C, Promkotra W, Techasen A, Namwat N, High expression of HIF-1α, BNIP3 and PI3KC3: hypoxia-induced autophagy predicts cholangiocarcinoma survival and metastasis, Asian Pac J Cancer Pre 15 (2014), 58735878. [21] Tan EY, Campo L, Han C, Turley H, Pezzella F, Gatter KC, Harris AL, Fox SB, BNIP3 as a progression marker in primary human breast cancer; opposing functions in in situ versus invasive cancer, Clin Cancer Res 13 (2007), 467-474. [22] Bacon AL, Harris AL, Hypoxia-inducible factors and hypoxic cell death in tumor physiology, Ann Med 36 (2004), 530-539. [23] Semenza GL, HIF-1: mediator of physiological and pathophysiological responses to hypoxia, J Appl Physiol 88 (2000), 1474-1480.
[24] [25] [26]
[27]
[28]
[29]
[30]
[31]
475
Harris AL, Hypoxia – a key regulatory factor in tumor growth. Nat Rev Cancer 2 (2002), 38-47. Gatenby RA, Gillies RJ, Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4 (2004), 891-899. Bos R, Zhong H, Hanrahan CF et al, Levels of hypoxiainducible factor-1alpha during breast carcinogenesis. J Natl Cancer Inst 93 (2001), 309-314. Zhang H, Bosch-Marce M, Shimoda LA, Tan YS, Baek JH, Wesley JB, et al, Mitochondrial autophagy is an HIF1-dependent adaptive metabolic response to hypoxia, J Biol Chem 283 (2008), 10892-10903. Seeber LM, Horree N, van der Groep P, van der Wall E, Verheijen RH, van Diest PJ, Necrosis related HIF-1alpha expression predicts prognosis in patients with endometrioid endometrial carcinoma, BMC Cancer 10 (2010), 307-317. Azad MB, Chen Y, Henson ES, Cizeau J, McMillan-Ward E, Israels SJ, Gibson SB, Hypoxia induces autophagic cell death in apoptosis-competent cells through a mechanism involving BNIP3, Autophagy 4 (2008), 195-204. Papandreou I, Lim AL, Laderoute K, Denko NC, Hypoxia signals autophagy in tumor cells via AMPK activity, independent of HIF-1, BNIP3, and BNIP3L, Cell Death Differ 15 (2008), 1572-1581. Healy SJ, Gorman AM, Mousavi-Shafaei P, Gupta S, Samali A, Targeting the endoplasmic reticulum-stress response as an anticancer strategy, Eur J Pharmacol 625 (2009), 234-246.
Cancer Biomarkers 15 (2015) 467–475 DOI 10.3233/CBM-150474 IOS Press
Expression of BNIP3 and its correlations to hypoxia-induced autophagy and clinicopathological features in salivary adenoid cystic carcinoma Zhanwei Chen1, Haiwei Wu1 , Shengyun Huang1, Wengang Li, Shizhou Zhang, Peihui Zheng, Xiaoqing Zhou, Wenlei Liu and Dongsheng Zhang∗ Department of Oral and Maxillofacial Surgery, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, Shandong, China
Abstract. BACKGROUND: The expression of Bcl-2/adenovirus E1B 19 kDa-interacting protein3 (BNIP3) has been explored in many human malignancies, but not in adenoid cystic carcinoma (ACC). OBJECTIVE: This study investigated the clinical significance of expression of BNIP3 in ACC tissues and cells and elucidated its correlations to hypoxia-induced autophagy. METHODS: Immunohistochemical and immunofluorescence staining were used to explore BNIP3, HIF-1α and LC3 expression. RESULTS: BNIP3 was positively expressed in 41 cases (63.1%), and was significantly correlated with histological grade (P = 0.001). HIF-1α was positively expressed in 52 cases (80.0%) and was significantly correlated with TNM stage (P = 0.023) and histological grade (P = 0.024). LC3 was positively expressed in 37 cases (56.9%) and was significantly correlated with TNM stage (P = 0.019). The expression of BNIP3 was correlated with HIF-1α expression (P = 0.011). The overall survival in the negative BNIP3 expression group tended to be better than in the positive BNIP3 expression (P = 0.011). In vitro experiment, BNIP3 immunofluorescence staining was detected in cells treated with CoCl2 (for hypoxic condition). CONCLUSIONS: The data indicated that BNIP3 plays a vital role in the tumorigenesis of adenoid cystic carcinoma and could be a new target for gene therapy of adenoid cystic carcinoma. Keywords: Adenoid cystic carcinoma, hypoxia, BNIP3, HIF-1α, autophagy
1. Introduction Adenoid cystic carcinoma (ACC) is one of the most frequent malignant tumors of the salivary glands in the head and neck especially in the minor salivary glands. 1 These
authors contributed equally to this study. author: Dongsheng Zhang, Proffessor, Dean, Department of Oral and Maxillofacial Surgery, Shandong Provincial Hospital Affiliated to Shandong University, 324 Jingwu Road, Jinan 250021, China. Tel.: +86 531 68776950; Fax: +86 531 87035697; E-mail: [email protected]. ∗ Corresponding
Clinically, it is described as one of the most destroyed and unpredicted malignant tumors because its propensity of perineural invasion, multiple recurrences and a high incidence of distant metastases [1]. Currently, surgery is the most effective treatment, which is usually followed by adjuvant radiotherapy including implantation of Iodine-125 seeds [2]. Although the 5-year survival rate is relatively high, the 10-year and longer survival rate remains low [3]. Therefore, further understanding of the characteristic biological behavior and molecular genetics of ACC may provide new insights into the novel treatment of the disease.
c 2015 – IOS Press and the authors. All rights reserved ISSN 1574-0153/15/$35.00
468
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Hypoxia is a common feature of solid tumors which occurs when the proliferation of the tumor outstrips the speed of blood vessels formation. Clinically, hypoxia may be deleterious for the tumor control by increasing resistance to radiation, preventing drug delivery through lack of blood vessels, and adaptation to the microenvironment through genomic instability. The transcriptional factor hypoxia-inducible factor 1α (HIF1α), the key regulator of the hypoxic response, plays an essential role in the adaptive response of cells to reduced oxygen tension and undergoes conformational changes in response to varying oxygen concentrations. Under normoxic conditions HIF-1α is hydroxylated, followed by binding to Von Hippel Lindau (VHL) protein and proteosomal degradation; on the contrary, this hydroxylation does not occur under hypoxic conditions [4]. The presence of HIF-1α has been observed in the majority of the common human cancers, such as head and neck cancer, ovarian cancer and oesophageal cancer [5–7], which is associated with poor prognosis and resistance to therapy. Costa et al found that ordinary as well as high-grade transformation adenoid cystic carcinoma exhibited HIF-1a expression, whereas it was observed only occasionally in normal salivary tissue adjacent to the tumor [8]. Bcl-2/adenovirus E1B 19 kDa-interacting protein3 (BNIP3), a member of BH3-only subfamily of Bcl-2 superfamily, belongs to pro-apoptotic proteins. The expression of BNIP3 is regulated by hypoxia and other factors. BNIP3 protein is localized on the outside of the mitochondrial membrane, and once stimulated, it integrates into the mitochondrial membrane, leading to permeabilization of the transition pore and decreasing the mitochondrial membrane potential resulting in chromatin condensation, DNA-fragmentation and finally apoptosis [9,10]. Studies have shown that abnormal expression of BNIP3 is correlated with the growth progression and drug resistance of many tumors in recent years [11]. BNIP3 is closely related to tumor, and the expression of BNIP3 in different tumor tissue is not the same and represents different prognosis. Recent studies have shown that BNIP3 is also closely related to autophagy. BNIP3, which is upregulated by the transcription factor HIF-1α, plays a role in hypoxiainduced autophagic cell death. In our previous study, we found that autophagy plays an important role in tumor progression and positive LC3 expression showed a statistically significant correlation with absent lymph node involvement and TNM stage in ACC [12]. However, the role of BNIP3 in hypoxia-induced autophagy is still extensively unexplored in ACC. Furthermore,
the relationship between expression of BNIP3 and clinicopathologic features of tumors has not been studied in salivary adenoid cystic carcinoma tissue. In our study, we investigated the expression of BNIP3, HIF-1α and LC3 by immunohistochemical analysis in head and neck salivary adenoid cystic carcinoma and normal salivary tissue. We also correlated BNIP3, HIF-1α and LC3 expression with histopathogic prognostic factors and survival. Additionally, we observed the expression of BNIP3, HIF1α and LC3 in ACC cells under hypoxic condition.
2. Material and methods 2.1. Patients and specimens All sixty-five specimens of head and neck adenoid cystic carcinoma, primary surgically resected between December 2002 and March 2011, were derived from the archives of Shandong Provincial Hospital, affiliated to Shandong University of China. Two independent pathologists reviewed all of the original hematoxylin and eosinstained sections to confirm the pathological diagnosis and to select a representative section for immunohistochemical study. Two or three representative cores with 3 mm in diameter, were obtained from a representative area of each paraffin-embedded tumor specimens. The median age of the patients was 48.2 years (range, 18–75 years). Tumor size ranged from 0.6 to 4.5 cm (median size, 1.9 cm). According to the American Joint Committee on Cancer staging criteria, 27 patients (41.6%) had stage I and II disease, and 38 patients (58.4%) had stage III and IV disease. Three histologic grades were confirmed: Grade I, tumors with tubular and cribriform areas but without solid forms; Grade II, cribriform tumors only or mixed with less than 30% of solid areas; and Grade III tumors with a dominantly solid pattern [13]. All patients were followed up by reviewing their medical records and telephoning. 2.2. Immunohistochemical staining Serial sections of 4 μm of paraffin embedded material of each specimen were deparaffinized in turpentine oil and rehydrated in graded alcohol. Antigen retrieval (AR) was achieved by boiling in a steamer immersed in citrate buffer (pH 6.0) for BNIP3 and LC3, while trisethylenediamine tetraacetic acid (EDTA) (pH 9.0) was used for HIF-1α. Endogenous peroxidase ac-
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
469
Fig. 1. Negative BNIP3 expression in normal salivary gland (A1), and positive BNIP3 expression in solid (A2), tubular (A3) and cribriform (A4) pattern of adenoid cystic carcinoma (A1–A4, 200×); Negative HIF-1α expression in normal salivary gland (B1), and positive HIF-1α expression in solid (B2), tubular (B3) and cribriform (B4) pattern of adenoid cystic carcinoma (B1–B4, 200×); Positive LC3 expression in normal salivary gland (C1), and in solid (C2), tubular (C3) and cribriform (C4) pattern of adenoid cystic carcinoma (C1–C4, 200×). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
tivity was blocked by 3% hydrogen peroxide for 30 min in the incubator. Thereafter slides were incubated for overnight at 4◦ C with primary monoclonal antibodies for LC3A/B (1:200, Abcam, Cambridge, United Kingdom), BNIP3 (1:200, Santa Cruz, USA) and HIF1a (1:50, Boster, Wuhan, China). Then slides were incubated with secondary antibodies for 30 min, using diaminobenzidine-tetrahydrochloride (DAB) as the substrate to detect antigen-antibody binding, and nuclei were counterstained with hematoxylin. The positive controls were provided by Abcam, and negative controls were obtained by omitting primary antibody replaced by PBS.
0 (< 10% of tumor cells), 1 (10–49% of tumor cells), 2 ( 50% of tumor cells). The intensity score and proportion score were multiplied to give a total score. The staining pattern of the slides was defined as follows: 0–3 (negative expression) and 4–5 (positive expression). Tumor cells were considered immunohistochemical staining positive for HIF-1α when stained in 30% or more of the cells. All slides were evaluated independently by two observers without clinicopathologic information.
2.4. Statistical methods
2.3. Evaluation of immunohistochemical staining The expression of BNIP3 ascertained by immunohistochemical staining in the cytoplasm and a positive signal was considered if cytoplasm stained brown. Although HIF-1α and LC3 were expressed in both the nuclei and cytoplasm in the cell of adenoid cystic carcinoma, we only evaluated the cell nucleus expression of HIF-1α and cytoplasmic expression of LC3. Immunohistochemical staining for BNIP3 and LC3 was evaluated in compliance with intensity and proportion. The intensities were scored as 0 (no staining), 1 (weak staining), 2 (moderate staining), and 3 (strong staining). The percentage of staining area was classified as
Statistical calculations were performed using the SPSS software package (version 19.0; SPSS, Chicago, IL, USA). The association was evaluated according to the Pearson chi-square test or Fisher’s exact test between BNIP3, HIF-1α and LC3 expression and several clinicopathologic variables. The Kaplan-Meier method was used to determine the probability of survival, and the data were analyzed by applying the log-rank test. Overall survival (OS) was defined as from the date of surgery to the date of death due to cancer. P value of less than 0.05 was considered statistically signifcant and all tests of statistical significance were two-sided.
470
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Table 1 Correlation between clinicopathologic variables and expression of BNIP3, HIF-1α and LC3 expression in 65 cases of head and neck adenoid cystic carcinoma Variables
BNIP3 expression P value HIF-1α expression P value LC3 expression P value Negative n(%) Positive n(%) Negative n(%) Positive n(%) Negative n(%) Positive n(%)
Age < 50 years 50 years Gender Female Male TNM stage I, II III,IV Tumor size 3 cm > 3 cm Histological grade Grade I, II Grade III Distant metastasis Absent Present ∗ Significantly
11 (16.9) 13 (20.0)
16 (24.6) 25 (38.5)
0.591
6 (9.2) 7 (10.8)
21 (32.3) 31 (47.7)
0.706
13 (20.0) 15 (23.1)
14 (21.5) 23 (35.4)
0.486
9 (13.8) 15 (23.1)
17 (26.1) 24 (37.0)
0.753
4 (6.2) 9 (13.8)
22 (33.8) 30 (46.2)
0.448
11 (16.9) 17 (26.2)
15 (23.1) 22 (33.8)
0.919
10 (15.4) 14 (21.5)
17 (26.1) 24 (37.0)
0.987
9 (13.8) 4 (6.2)
18 (27.7) 34 (52.3)
0.023∗
7 (10.8) 21 (32.3)
20 (30.8) 17 (26.1)
0.019∗
16 (24.6) 8 (12.3)
36 (55.4) 5 (7.7)
0.056
8 (12.3) 5 (7.7)
44 (67.7) 8 (12.3)
0.114
21 (32.3) 7 (10.8)
31 (47.7) 6 (9.2)
0.381
20 (30.7) 4 (6.2)
17 (26.1) 24 (37.0)
0.001∗
11 (16.9) 2 (3.1)
26 (40.0) 26 (40.0)
0.024∗
15 (23.1) 13 (20.0)
22 (33.8) 15 (23.1)
0.635
20 (30.7) 4 (6.2)
34 (52.3) 7 (10.8)
1.000
9 (13.8) 4 (6.2)
45 (69.2) 7 (10.8)
0.209
23 (35.4) 5 (7.7)
31 (47.7) 6 (9.2)
1.000
different by the Pearson chi-square test or Fisher exact test.
2.5. Immunofluorescence staining ACC-M Cells grown on sterile glass coverslips in 6-cm cell culture dishes were allowed to attach by overnight incubation, and then cultured with DMSO (Control) or 200 μM CoCl2 (for hypoxic condition) for indicated time. At the end of treatments, cells were washed with PBS, fixed with 4% paraformaldehyde for 20 min, and permeabilized with 0.5% Triton X-100 for 10 min. Then the cells were incubated with primary antibody (BNIP3, HIF-1α and LC3) for 4◦ C overnight., followed by treatment with FITC conjugated goat antirabbit IgG (1:50, Beijing Cwbio Co., Ltd) or TRITC conjugated goat anti-mouse IgG (1:50, Beijing Cwbio Co., Ltd) for 1 h. Then nuclei were counterstained with DAPI for 7 min, followed by observation under a Nikon fluorescence microscope.
3. Results 3.1. BNIP3 expression and its association with clinicopathological variables BNIP3 immunohistochemical staining was detected mainly in the membrane and cytoplasm of tumor cells, but not in the normal head and neck salivary glands cells (Fig. 1 A1). BNIP3 expression of adenoid cystic carcinoma was observed in cribriform, tubular, and solid histological patterns, and was positive present in
41 cases (63.1%) (Fig. 1 A2–A4). The correlations between BNIP3 and clinical variables are listed in Table 1. BNIP3 expression was significantly correlated with histological grade (P = 0.001). BNIP3 expression was not significantly associated with age (P = 0.591), gender (P = 0.753), TNM stage (P = 0.987), tumor size (P = 0.056), and distant metastasis (P = 1.000) (Table 1). 3.2. HIF-1α expression and its association with clinicopathological variables HIF-1α immunohistochemical staining was detected mainly in the nucleus of tumor cells, but rarely in the normal head and neck salivary glands cells (Fig. 1 B1). HIF-1α expression of adenoid cystic carcinoma was also observed in cribriform, tubular, and solid histological patterns, and was positive present in 52 cases (80.0%) (Fig. 1 B2–B4). The correlations between BNIP3 and clinical variables are listed in Table 1. HIF-1α expression was significantly correlated with TNM stage (P = 0.023) and histological grade (P = 0.024). HIF-1α expression was not significantly associated with age (P = 0.706), gender (P = 0.448), tumor size (P = 0.114) and distant metastasis (P = 0.209) (Table 1). 3.3. LC3 expression and its association with clinicopathological variables Consist with our previous study [12], positive LC3 was showed in normal head and neck gland. LC3 ex-
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
471
Table 2 Correlation between BNIP3 and HIF-1α expression in 65 cases of head and neck adenoid cystic carcinoma BNIP3 expression Negative n(%)
0.011∗
HIF-1α expression Negative Positive ∗ Significantly
P value Positive n(%)
9 (13.8) 15 (23.1)
4 (6.2) 37 (56.9)
different by the Fisher exact test. Table 3 Correlation between expression BNIP3 and LC3 expression in 65 cases of head and neck adenoid cystic carcinoma BNIP3 expression
LC3 expression Negative Positive
P value
Negative n(%)
Positive n(%)
13 (20.0) 11 (16.9)
15 (23.1) 26 (40.0)
0.167
pression was also found in the plasma membrane, cytoplasm, and/or nucleus of tumor cells (Fig. 1 C1). The LC3 expression of adenoid cystic carcinoma was also observed in cribriform, tubular, and solid histological patterns, and was positive present in 37 cases (56.9%) (Fig. 1 C2–C4). The correlations between LC3 and clinical variables are listed in Table 1. LC3 expression was significantly correlated with TNM stage (P = 0.019). LC3 expression was not significantly associated with age (P = 0.486), gender (P = 0.919), tumor size (P = 0.381), histological grade (P = 0.635) and distant metastasis (P = 1.000).
3.6. Expression of BNIP3 in ACC-M cells under hypoxic condition
3.4. Relationship between BNIP3 and HIF-1α expression, BNIP3 and LC3 expression
4. Discussion
As shown in Table 2, we found that BNIP3 expression was positively correlated with HIF-1α expression in head and neck adenoid cystic carcinoma (P = 0.011). By contrast, as shown in Table 3, BNIP3 expression was not significantly correlated with LC3 expression (P = 0.167). 3.5. Results of the overall survival analysis The mean follow-up period was 61.8 months (range, 18–115 months), and for 65 patients, 30 patients died of disease, 33 patients were alive on the day of the study and 2 patients were lost during the follow-up period. Based on the use of the Kaplan-Meier method and log-rank test, the overall survival in the negative BNIP3 expression group tended to be better than in the positive BNIP3 expression (P = 0.011) (Fig. 2); HIF1a expression showed no significant difference for the overall survival rate in patients with adenoid cystic carcinoma (P = 0.132) (Fig. 3).
In vitro study, BNIP3 immunofluorescence staining was detected in ACC-M cells treated with CoCl2 (for hypoxic condition) (Fig. 4). Additionally, HIF-1α and LC3 were also detected in ACC-M cells under hypoxia (Fig. 4). Furthermore, cytoplasmic puncta tethered with LC3 was detected in ACC-M cells, indicating the formation of autophagosome-like structure in ACC cells.
Our study evaluated that expression of BNIP3 and HIF-1α in salivary adenoid cystic carcinoma using immunohistochemical analysis and cell immunofluorescence. Furthermore, we demonstrated that expression of BNIP3 in salivary adenoid cystic carcinoma at the protein level in correlation with the hypoxic response and autophagy. We also correlated BNIP3, HIF-1α and LC3 expression with histopathogic prognostic factors and survival. BNIP3, a pro-apoptosis member of the Bcl-2 family, has been demonstrated in a wide variety of normal cells and in cells undergoing pathological hypoxic stress. Because the promoter of BNIP3 contains a hypoxia response element (HRE), BNIP3 appears to be a direct target for transcriptional activation by HIF1 [14]. The low expression of BNIP3 protein has been found in pancreatic cancer [15], colorectal and gastric cancer [16], and hematopoietic malignancies [17]. On the contrary, high levels of BNIP3 protein have been described in non-small cell lung cancer [18], basal cell carcinoma [19] and cholangiocarcinoma [20]. Re-
472
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
Fig. 2. Cumulative Kaplan-Meier survival curve (BNIP3) showing that survival in the negative BNIP3 expression group (n=24) tended to be better than in the positive BNIP3 expression group (n=41) (P =0.011 , log-rank test). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
Fig. 3. Cumulative Kaplan-Meier survival curve (HIF-1α) showing that the survival in the negative HIF-1α expression group (n = 13) and the positive HIF-1α group (n = 52) was not significantly different (P = 0.132, log-rank test). (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
473
Fig. 4. Positive HIF-1α expression in the nuclear of ACC-M cells; positive BNIP3 and LC3 expression in the cytoplasm of ACC-M cells. (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/CBM-150474)
cently it also has been shown that BNIP3 may have opposing functions at different stages of tumor progression in breast cancer [21]. Although expression of BNIP3 has been studied in a variety of tumors, little studies on BNIP3 expression in salivary adenoid cystic carcinoma have been published. To our knowledge, our study firstly evaluates the immunohistochemical expression of BNIP3 in head and neck adenoid cystic carcinoma compared to normal tissue. We demonstrated that 63.1% of patients expressed the BNIP3 protein in head and neck adenoid cystic carcinoma tissues while BNIP3 was not expressed in normal head and neck salivary gland tissues. Our study also demonstrated that positive expression of BNIP3 was significantly associated with the histological grade, and shorter overall survival in head and neck adenoid cystic carcinoma. These findings confirm that BNIP3 may play an important role in tumor progression, and BNIP3 could be a indicator of poor prognosis of head and neck adenoid cystic carcinoma. Hypoxia is a poor prognostic indicator for solid tumors due to increased resistance to radiation, restricted access to drug delivery through lack of blood vessels, and adaptation to the microenvironment through genomic instability [22]. One way in which cells respond to hypoxia is through hypoxia-inducible factor-1 (HIF1) which consists of α and β subunits. The stability of HIF-1 depends mainly on the α subunit [23]. Under normoxia, HIF-1α is degraded, whereas this process is inhibited leading to protein stabilization under hypoxia. Stabilized HIF-1α is transported into the nucleus where it heterodimerizes with HIF-1β to transactivate the expression of numerous hypoxiaresponse genes, including the pro-angiogenic vascular endothelial growth factor (VEGF) and glucose transporters [24]. In the literature, the correlation between levels of HIF-1α expression with tumor histological grade and biological behaviour is variable and seems to depend on the tumor type. In breast carcinoma, levels of HIF-1α have been reported to increase as the
degree of malignancy increases indicating its potential association with poor clinical outcome [25]. In contrast, in endometrial carcinoma, high HIF-1α was associated significantly with low-grade tumors [26]. In our study, we found that 80% of patients expressed the HIF-1α protein in head and neck adenoid cystic carcinoma tissues. This phenomenon also appeared in consist with Costa et al who found that ordinary as well as high-grade transformation adenoid cystic carcinoma exhibited HIF-1α expression [8]. Our study also demonstrated that positive expression of HIF-1α is significantly associated with the histological grade and TNM stage in head and neck adenoid cystic carcinoma. Based on the use of the Kaplan-Meier method and logrank test, HIF-1α expression showed no significant difference for the overall survival rate in patients with adenoid cystic carcinoma. These findings indicate that although hypoxia plays an important role in tumor progression, it could not be considered as a predictive indicator for prognosis in head and neck adenoid cystic carcinoma. Recent studies have shown that hypoxia can induce the autophagy through BNIP3. Hypoxia-induced BNIP3 plays an important role in hypoxia-induced autophagic cell death, and the BH-3 domain of BNIP3 is critical for this physiological process. In this progress, BH3 domains of BNIP3 were shown to compete with Beclin 1 for Bcl-2 binding, and Beclin 1 was released to induce autophagy [27]. Although hypoxia induces autophagy via BNIP3 as a pro-death mechanism, this progress has always been claimed to be a prosurvival mechanism promoting tumor progression by removing of potentially toxic damaged proteins and organelles [28]. Recently it has been shown that BNIP3 may have opposing functions at different stages of progression in breast cancer, which could be explained by its role in autophagy. In pre-invasive cancers, induction of autophagy through BNIP3 may provide with extra nutrients and promote further tumor progression. However at later stages, BNIP3 expression could pro-
474
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy
mote prolonged autophagy in hypoxic cancer cells possibly through increased autophagosome-lysosome fusions, leading to autophagic cell death and better patient outcome [29]. Our results showed that BNIP3 expression was positively correlated with HIF-1a expression, while was not correlated with LC3 expression in head and neck adenoid cystic carcinoma. This suggested that BNIP3-induced autophagy may not play a dominant role in hypoxia-induced autophagy in ACC tissues. HIF-1α/BNIP3 pathway has been explored in several studies about hypoxia-induced autophagy, however, this pathway is not in charge of all hypoxiainduced autophagy. It’s reported that hypoxia-induced hypoxia is induced by HIF-1 dependent pathway and HIF-1 independent pathway, and HIF-1α/BNIP3 pathway is not involved in 5’ AMP activated protein kinase (AMPK) signaling pathway and unfolded protein response (UPR) pathway in the induction of autophagy under severe hypoxia [30,31]. In our in vitro experiment, up-regulation of HIF-1α and BNIP3 in ACC-M cells can obviously improve expression of autophagyrelated gene LC3, which indicates that BNIP3 was involved in hypoxia-induced autophagy in ACC. In conclusion, our study demonstrated expression of HIF-1α and BNIP3 in head and neck salivary adenoid cystic carcinoma and its relationship with clinicopathologic factors and patient survival. These results indicate that BNIP3 may play an important role in the tumorigenesis of salivary adenoid cystic carcinoma, and may be a new prognostic indicator of the outcome in patients with salivary adenoid cystic carcinoma. Further studies are needed to confirm our findings and to clarify the exact role of BNIP3 in the development of head and neck adenoid cystic carcinoma.
Acknowledgment This study is supported by Shandong Provincial Science and Technology Development Plan (No. 2006GG 2202034), the Science and Technology Foundation of Shandong Province (2006GG20002046) and Shandong Provincial Natural Science Foundation (ZR2012 HQ022).
Conflicts of interest The authors declare that there are no conflicts of interest in the research.
References [1]
Dodd RL, Slevin NJ, Salivary gland adenoid cystic carcinoma: A review of chemotherapy and molecular therapies, Oral Oncol 42 (2006), 759-769. [2] Zhang J1, Zhang JG, Song TL, Zhen L, Zhang Y, Zhang KH, Yang ZH, Yu GY, 125I seed implant brachytherapy-assisted surgery with preservation of the facial nerve for treatment of malignant parotid gland tumors, Int J Oral Maxillofac Surg 37 (2008), 515-520. [3] Spiro RH, Huvos AG, Strong EW, Adenoid cystic carcinoma of salivary origin: A clinicopathologic study of 242 cases, Am J Surg 128 (1974), 512-520. [4] Koop EA, van Laar T, van Wichen DF, de Weger RA, Wall Ev, van Diest PJ, Expression of BNIP3 in invasive breast cancer: correlations with the hypoxic response and clinicopathological features, BMC Cancer 9 (2009), 175. [5] Gruber G, Greiner RH, Hlushchuk R, et al, Hypoxia-inducible factor 1 alpha in high-risk breast cancer: an independent prognostic parameter? Breast Cancer Res 6 (2004), R191-R198. [6] Birner P, Schindl M, Obermair A, Breitenecker G,Oberhuber G, Expression of hypoxia-inducible factor 1 in epithelial ovarian tumors: Its impact on prognosis and on response to chemotherapy, Clin Cancer Res 7 (2001), 1661-1668. [7] Lu XG, Xing CG, Feng YZ, Chen J and Deng C, Clinical significance of immunohistochemical expression of hypoxiainducible factor-1 alpha as a prognostic marker in rectal adenocarcinoma, Clin Colorectal Cancer 5 (2006), 350-353. [8] Costa AF1, Tasso MG, Mariano FV, et al, Levels and patterns of expression of hypoxia-inducible factor-1α, vascular endothelial growth factor, glucose transporter-1 and CD105 in adenoid cystic carcinomas with high-grade transformation, Histopathology 60 (2012), 816-825. [9] Velde C Vande, Cizeau J, Dubik D, Alimonti J, Brown T, Israels S,Hakem R, Greenberg AH, BNIP3 and genetic control of necrosis-like cell death through the mitochondrial permeability transition pore, Mol Cell Biol 20 (2000), 5454-5468. [10] Imazu T, Shimizu S, Tagami S, Matsushima M, Nakamura Y, Miki T, Okuyama A, Tsujimoto Y, Bcl-2/E1B 19 kDainteracting protein 3-like protein (Bnip3L) interacts with bcl2/Bcl-xL and induces apoptosis by altering mitochondrial membrane permeability, Oncogene 18 (1999), 4523-4529. [11] Chinnadurai G1, Vijayalingam S, Gibson SB, BNIP3 subfamily BH3-only proteins: mitochondrial stress sensors in normal and pathological functions, Oncogene 1 (2008), S114-S127. [12] Jiang L, Huang S, Li W, Zhang D, Zhang S, Zhang W, Zheng P, Chen Z, Expression of autophagy and ER stress-related proteins in primary salivary adenoid cystic carcinoma, Pathol Res Pract 208 (2012), 635-641. [13] P.A.Szanto, M.A.Luna, M.E.Tortoledo, et al, Histologic grading of adenoid cystic carcinoma of the salivary glands, Cancer 54 (1984), 1062-1069. [14] Bruick RK, Expression of the gene encoding the proapoptotic Nip3 protein is induced by hypoxia, Proc Natl Acad Sci U S A 97 (2000), 9082-9087. [15] Okami J, Simeone DM, Logsdon CD, Silencing of the hypoxia-inducible cell death protein BNIP3 in pancreatic cancer, Cancer Res 64 (2004), 5338-5346. [16] Murai M, Toyota M, Suzuki H, Satoh A, Sasaki Y, Akino K, Ueno M,Takahashi F, Kusano M, Mita H, et al, Aberrant methylation and silencing of the BNIP3 gene in colorectal and gastric cancer, Clin Cancer Res 11 (2005), 1021-1027. [17] Murai M, Toyota M, Satoh A, Suzuki H, Akino K, Mita H, Sasaki Y, Ishida T, Shen L, Garcia-Manero G, Aberrant DNA
Z. Chen et al. / Expression of BNIP3 and its correlations to hypoxia-induced autophagy methylation associated with silencing BNIP3 gene expression in haematopoietic tumors, Br J Cancer 92 (2005), 1165-1172. [18] Giatromanolaki A, Koukourakis MI, Sowter HM, Sivridis E, Gibson S,Gatter KC, Harris AL, BNIP3 expression is linked with hypoxia-regulated protein expression and with poor prognosis in non-small cell lung cancer, Clin Cancer Res 10 (2004), 5566-5571. [19] Brinkhuizen T, Weijzen CA, Eben J, Thissen MR, van Marion AM, Lohman BG, Winnepenninckx VJ, Nelemans PJ, van Steensel MA, Immunohistochemical analysis of the mechanistic target of rapamycin and hypoxia signalling pathways in basal cell carcinoma and trichoepithelioma, Plos One (2014), doi: 10.1371/journal.pone.0106427. [20] Thongchot S, Yongvanit P, Loilome W, Seubwai W, Phunicom K, Tassaneeyakul W, Pairojkul C, Promkotra W, Techasen A, Namwat N, High expression of HIF-1α, BNIP3 and PI3KC3: hypoxia-induced autophagy predicts cholangiocarcinoma survival and metastasis, Asian Pac J Cancer Pre 15 (2014), 58735878. [21] Tan EY, Campo L, Han C, Turley H, Pezzella F, Gatter KC, Harris AL, Fox SB, BNIP3 as a progression marker in primary human breast cancer; opposing functions in in situ versus invasive cancer, Clin Cancer Res 13 (2007), 467-474. [22] Bacon AL, Harris AL, Hypoxia-inducible factors and hypoxic cell death in tumor physiology, Ann Med 36 (2004), 530-539. [23] Semenza GL, HIF-1: mediator of physiological and pathophysiological responses to hypoxia, J Appl Physiol 88 (2000), 1474-1480.
[24] [25] [26]
[27]
[28]
[29]
[30]
[31]
475
Harris AL, Hypoxia – a key regulatory factor in tumor growth. Nat Rev Cancer 2 (2002), 38-47. Gatenby RA, Gillies RJ, Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4 (2004), 891-899. Bos R, Zhong H, Hanrahan CF et al, Levels of hypoxiainducible factor-1alpha during breast carcinogenesis. J Natl Cancer Inst 93 (2001), 309-314. Zhang H, Bosch-Marce M, Shimoda LA, Tan YS, Baek JH, Wesley JB, et al, Mitochondrial autophagy is an HIF1-dependent adaptive metabolic response to hypoxia, J Biol Chem 283 (2008), 10892-10903. Seeber LM, Horree N, van der Groep P, van der Wall E, Verheijen RH, van Diest PJ, Necrosis related HIF-1alpha expression predicts prognosis in patients with endometrioid endometrial carcinoma, BMC Cancer 10 (2010), 307-317. Azad MB, Chen Y, Henson ES, Cizeau J, McMillan-Ward E, Israels SJ, Gibson SB, Hypoxia induces autophagic cell death in apoptosis-competent cells through a mechanism involving BNIP3, Autophagy 4 (2008), 195-204. Papandreou I, Lim AL, Laderoute K, Denko NC, Hypoxia signals autophagy in tumor cells via AMPK activity, independent of HIF-1, BNIP3, and BNIP3L, Cell Death Differ 15 (2008), 1572-1581. Healy SJ, Gorman AM, Mousavi-Shafaei P, Gupta S, Samali A, Targeting the endoplasmic reticulum-stress response as an anticancer strategy, Eur J Pharmacol 625 (2009), 234-246.
