Tumor Biol. (2014) 35:9777–9785 DOI 10.1007/s13277-014-2182-8
RESEARCH ARTICLE
Overexpression of HP1γ is associated with poor prognosis in non-small cell lung cancer cell through promoting cell survival Ji Zhou & Hui Bi & Ping Zhan & Cunjie Chang & Chunhua Xu & Xiaojing Huang & Like Yu & Xin Yao & Jun Yan
Received: 6 April 2014 / Accepted: 2 June 2014 / Published online: 1 July 2014 # International Society of Oncology and BioMarkers (ISOBM) 2014
Abstract Heterochromatin protein 1γ (HP1γ), which binds to di- or trimethylated lysine 9 on histone H3 (H3K9), plays an important role in chromatin packaging and gene transcriptional regulation. Recently, HP1γ has been implicated in cancer development. However, its clinical relevance and functional role in non-small cell lung cancer (NSCLC) remain elusive. In this study, we found that HP1γ expression was elevated in NSCLC samples at the messenger RNA (mRNA) level compared to adjacent normal lung tissues. In a cohort of 108 NSCLC patients, HP1γ overexpression is significantly associated with N stage (P=0.003), pathological tumor–node– metastasis (TNM) stage (P=0.013), smoking status (P= 0.009), and gender (P=0.042). Patients with a high level of HP1γ expression showed a poorer overall survival rate than those with low HP1γ expression (P=0.017). Multivariate analysis revealed that HP1γ expression is an independent prognostic marker. We also found knockdown of HP1γ in A549 and NCI-H1975 cells induced apoptosis accompanied with suppressed cell proliferation and colony formation. Ji Zhou, Hui Bi, and Ping Zhan have contributed equally to this study. J. Zhou : H. Bi : X. Yao (*) Department of Respiratory Medicine, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, Jiangsu, China e-mail: [email protected] P. Zhan : C. Xu : L. Yu The First Department of Respiratory Medicine, Nanjing Chest Hospital, Nanjing 210029, Jiangsu, China P. Zhan : C. Xu : L. Yu Clinical Center of Nanjing Respiratory Diseases and Imaging, Nanjing 210029, Jiangsu, China C. Chang : X. Huang : J. Yan (*) MOE Key Laboratory of Model Animal for Disease Study, Model Animal Research Center, Nanjing Biomedical Research Institute, Nanjing University, 12 Xuefu Road, Nanjing 210061, Jiangsu, China e-mail: [email protected]
Consistently, pro-apoptotic proteins, Bax and GADD45α, were upregulated in response to HP1γ depletion. Altogether, our data suggested that HP1γ plays an important role in promoting NSCLC and may represent a novel prognostic biomarker and therapeutic target for the disease. Keywords HP1γ . NSCLC . Prognosis . Proliferation . Apoptosis
Introduction Among all cancer types, lung cancer ranks the first in both incidence and mortality, with non-small lung cancer cell (NSCLC) accounting for about 85 % of all the cases [1]. In the USA, just over one in eight lung cancer patients will be living 5 years after their diagnosis [2], and the poor prognosis is attributable to lack of efficient diagnostic methods for early detection and lack of successful treatment. Hence, it is imperative to identify novel biomarkers and potential therapeutic target, in order to improve diagnostic accuracy, patient followup, and treatment selection. Epigenetic alterations are frequently detected in lung cancer and have been exploited as biomarkers [3–6]. Histone H3 methylation (H3K9) is associated with DNA methylation on the promoter region of tumor suppressor genes for silencing expression [7, 8]. Recent studies revealed that H3K9 methyltransferase G9a mediates lung cancer cell invasion through transactivating EpCAM, suggesting that the proteins affecting H3K9 methylation play important roles in lung cancer development [9]. Members of the heterochromatin protein 1 family (HP1α, β, and γ) bind to di- or trimethylated H3K9 on the promoter region and regulate gene expression [10, 11]. The misregulation of HP1 gene expression could cause many human diseases including cancer [11, 12]. Previous studies
9778
revealed that HP1γ was highly expressed in different cancers, including prostate and colon cancer [13, 14]. HP1γ overexpression has been identified as an independent prognostic marker and predicts biochemical recurrence after radical prostatectomy [13]. However, the clinical significance of HP1γ in NSCLC is not clear, and the molecular mechanisms in which HP1γ is employed to influence the development and relapse of NSCLC remain under investigation. In this study, we demonstrated that HP1γ expression is elevated in NSCLC. Further analysis in a cohort of 108 NSCLC patients indicated the positive association between HP1γ expression level and clinicopathological features and clinical outcome. By using NSCLC cell lines, we found that depleting HP1γ expression suppresses lung cancer cell proliferation and colony formation and upregulates pro-apoptosis gene expression levels.
Materials and methods Patients and specimen collection One hundred and eight lung cancer patients who received surgery at the Department of Thoracic Surgery of Nanjing Chest Hospital during 2007 to 2009 of May were included in this study. Lung specimens from cancer tissues and normal tissues (with >5 cm distance from the tumor edge) were immediately snap frozen in liquid nitrogen and then stored at −80 °C until use. None of the patients received any chemotherapy or radiotherapy before surgery. Clinicopathological data such as age, gender, smoking condition, size of tumor, histological differentiation, lymph nodal status, and pathological stage were obtained at the same time. The postoperative pathological staging was determined according to the 7th edition of the tumor–node–metastasis (TNM) classification. Histological type was determined according to the classification by the World Health Organization. The Research Ethics Committee of Nanjing Chest Hospital approved this protocol and informed written consent was obtained from all the participants. Immunohistochemical staining Resected specimens were fixed in 10 % formalin and paraffinembedded blocks were prepared. Five-micrometer-thick sections were cut from the specimens and placed on slides coated with poly-L-lysine. Sections were routinely deparaffinized with xylene and rehydrated in decreasing concentrations of alcohol. The sections were boiled in 10 mM citric acid buffer (pH 6.0) in a steaming cooker for 25 min for antigen retrieval. Endogenous peroxidase activity was blocked for 15 min of incubation with 3 % hydrogen peroxide. Following two washes in phosphate-buffered saline (PBS), sections were
Tumor Biol. (2014) 35:9777–9785
incubated in 10 % normal goat serum for 15 min. The sections were then incubated overnight at 4 °C with the rabbit antihuman HP1γ polyclonal antibody (1:300, Cell Signaling Technologies, Inc.). Then, the sections were incubated in biotinylated goat anti-rabbit IgG antibody for 15 min at room temperature. The streptavidin-biotinylated peroxidase solution was applied for 15 min at room temperature. Subsequently, the sections were washed with PBS and DAB colorization was applied until the color developed. Staining was monitored under a bright-field microscope and the reaction was stopped by washing with distilled water. The sections were then counterstained with hematoxylin. The slides were independently evaluated by two pathologists of Nanjing Chest Hospital, in a double-blinded manner, as described previously [15]. If discrepancies existed between these two reviewers, a consensus judgment was reached through discussion. The proportion of staining tumor cells in each selected field was determined by counting individual tumor cells at four randomly selected high magnification fields. The immunoreactivities were graded as (−), (+), (++), and (+++) according to the percentage of positive tumor cells identified: (−) represents 0 or 50 % tumor cells present. The immunoreactivities were graded according to the percentage of positive tumor cells identified. High expression of HP1γ was arbitrarily defined, if ≥40 % of the tumor cells showed granular staining at nuclear. RNA collection, reverse transcription, and quantitative RT-PCR RNA was extracted using TRIzol® (Cat no. 15596018, Invitrogen, Carlsbad, CA, USA) as per instruction and then reverse-transcribed by random primers using TaKaRa reverse transcription kit (Dalian, China). Real-time PCR (qRT-PCR) was carried out, using SYBR Green in an ABI 7500 StepOne Plus Real Time PCR instrument (Applied Biosystems, USA). The relative expression level of each target gene was normalized to β-actin in one sample. Primers used in our study were as follows: HP1γ: forward primer: 5′-AACCAAGAGGATTT GCCAGA-3′ and reverse primer: 5′-CTGGACAAGAATGC CAAGTT-3′; gadd45a: forward primer: 5′-GAGAGCAGAA GACCGAAAGGA-3′ and reverse primer: 5′-CACAACAC CACGTTATCGGG-3′; and β-actin: forward primer: 5′CATGTACGTTGCTATCCAGGC-3′ and reverse primer: 5′CTCCTTAATGTCACGCACGA-3′. Cell line and small interfering RNA transfection A549 and NCI-H1975 were purchased from ATCC (Manassas, VA, USA) and cultured in the DMEM containing 10 % FBS (Life Technologies) at 37 °C with 5 % CO2. Small
Tumor Biol. (2014) 35:9777–9785
9779
interfering RNA (siRNA) transfection was conducted using Lipofectamine 2000 (Cat no. 11668019, Life Technologies) as per the manufacturer’s instruction. siRNA sequences are as follows: HP1γ siRNA: 5′-GAGGCAGAGCCUGAAG AAUtt-3′; siNC: 5′-UUCUCCGAACGUGUCACGUtt-3′. Briefly, 2.5 μl Lipofectamine 2000 and 100 pM siRNA were mixed in 500 μl OPTI-MEM (Cat no. 31985088, Invitrogen) for 20 min, followed by the addition of the mixture into medium, and the final concentration of siRNA was 50 μM.
Kit (Invitrogen, Cat no. V13245). Briefly, an appropriate number of cells was seeded to ensure that more than 1×106 cells can be harvested after siRNA treatment, followed by PBS washing and harvesting by trypsinization. Then, the cells were washed with PBS twice and resuspended in 100 μl of binding buffer provided in the kit and incubated with Annexin V/PI for 15 min in the dark at room temperature. Finally, the cells were detected on FACS Calibur® Instrument (Becton Dickinson) and analyzed using the FlowJo 7.6 software.
MTT assay
Statistical analysis
A time course assay for proliferation was performed after gene knockdown. Cells were plated in 96-well plates at a density of 1,000/well. siRNAs were transfected into the cells 24 h after seeding and cell proliferation was detected by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay every 24 h after siRNA transfection. Ten microliters of 0.5 mg/ml MTT (Sigma, M5655) was added into each well for 4 h, then the cell medium was removed, and 100 μl of dimethyl sulfoxide was added to dissolve the formazan at 37 °C for 10 min. Optical density at 490 nm was measured by a multiwell spectrophotometer (BioTek).
The difference in the level of expression of HP1γ messenger RNA (mRNA) and protein was calculated using paired t test. Associations between HP1γ protein expression and the clinicopathological characteristics were analyzed using Mann– Whitney U test. Overall survival (OS) was calculated from the day of surgery to the date of the last follow-up or the date of death. Survival at the last follow-up date was censored. The postoperative survival curves were calculated using the Kaplan–Meier method and differences in the survival rates were analyzed using the log-rank test. All statistical procedures were performed using SPSS (version 16.0, SPSS Inc, Chicago). P0.05).
Table 1 HP1γ protein expression and clinicopathological factors in NSCLC patients Characteristic
Number of patients HP1γ protein (%) expression Low High P value
All patients Gender Male Female Age 3 Smoking status No smoking Smoking Histology type Squamous cell carcinoma Adenocarcinoma
108
71
37
85 (78.7) 23 (21.3)
60 11
25 12
0.042*
0.505 66 (61.1) 42 (38.9)
45 26
21 16
30 (27.8) 78 (72.2)
18 53
12 25
0.438
0.009* 40 (37.0) 68 (63.0)
20 51
20 17 0.311
47 (43.5) 53 (49.1)
Other 8 (7.4) Lymph node metastasis (pN) N0 65 (60.2) N1 +N2 +N3 43 (39.8) pTNM stages I 51 (47.3) II 40 (37.0) III 16 (14.8) IV 1 (0.9) *P65/3 cm/≤3 cm) N stage (N1+2+3/N0)
0.585 (0.330–1.039) 0.689 (0.351–1.353) 2.452 (0.414–4.537) 0.735 (0.414–1.303) 1.734 (1.383–2.665)
0.068 0.279 0.743 0.291 0.023*
Stage (I and II vs III and IV) HP1γ expression (high/low)
1.932 (1.312–2.675) 2.125 (1.644–2.760)
0.005* 0.008*
HR hazard ratio, 95% CI 95 % confidence interval. *P
RESEARCH ARTICLE
Overexpression of HP1γ is associated with poor prognosis in non-small cell lung cancer cell through promoting cell survival Ji Zhou & Hui Bi & Ping Zhan & Cunjie Chang & Chunhua Xu & Xiaojing Huang & Like Yu & Xin Yao & Jun Yan
Received: 6 April 2014 / Accepted: 2 June 2014 / Published online: 1 July 2014 # International Society of Oncology and BioMarkers (ISOBM) 2014
Abstract Heterochromatin protein 1γ (HP1γ), which binds to di- or trimethylated lysine 9 on histone H3 (H3K9), plays an important role in chromatin packaging and gene transcriptional regulation. Recently, HP1γ has been implicated in cancer development. However, its clinical relevance and functional role in non-small cell lung cancer (NSCLC) remain elusive. In this study, we found that HP1γ expression was elevated in NSCLC samples at the messenger RNA (mRNA) level compared to adjacent normal lung tissues. In a cohort of 108 NSCLC patients, HP1γ overexpression is significantly associated with N stage (P=0.003), pathological tumor–node– metastasis (TNM) stage (P=0.013), smoking status (P= 0.009), and gender (P=0.042). Patients with a high level of HP1γ expression showed a poorer overall survival rate than those with low HP1γ expression (P=0.017). Multivariate analysis revealed that HP1γ expression is an independent prognostic marker. We also found knockdown of HP1γ in A549 and NCI-H1975 cells induced apoptosis accompanied with suppressed cell proliferation and colony formation. Ji Zhou, Hui Bi, and Ping Zhan have contributed equally to this study. J. Zhou : H. Bi : X. Yao (*) Department of Respiratory Medicine, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, Jiangsu, China e-mail: [email protected] P. Zhan : C. Xu : L. Yu The First Department of Respiratory Medicine, Nanjing Chest Hospital, Nanjing 210029, Jiangsu, China P. Zhan : C. Xu : L. Yu Clinical Center of Nanjing Respiratory Diseases and Imaging, Nanjing 210029, Jiangsu, China C. Chang : X. Huang : J. Yan (*) MOE Key Laboratory of Model Animal for Disease Study, Model Animal Research Center, Nanjing Biomedical Research Institute, Nanjing University, 12 Xuefu Road, Nanjing 210061, Jiangsu, China e-mail: [email protected]
Consistently, pro-apoptotic proteins, Bax and GADD45α, were upregulated in response to HP1γ depletion. Altogether, our data suggested that HP1γ plays an important role in promoting NSCLC and may represent a novel prognostic biomarker and therapeutic target for the disease. Keywords HP1γ . NSCLC . Prognosis . Proliferation . Apoptosis
Introduction Among all cancer types, lung cancer ranks the first in both incidence and mortality, with non-small lung cancer cell (NSCLC) accounting for about 85 % of all the cases [1]. In the USA, just over one in eight lung cancer patients will be living 5 years after their diagnosis [2], and the poor prognosis is attributable to lack of efficient diagnostic methods for early detection and lack of successful treatment. Hence, it is imperative to identify novel biomarkers and potential therapeutic target, in order to improve diagnostic accuracy, patient followup, and treatment selection. Epigenetic alterations are frequently detected in lung cancer and have been exploited as biomarkers [3–6]. Histone H3 methylation (H3K9) is associated with DNA methylation on the promoter region of tumor suppressor genes for silencing expression [7, 8]. Recent studies revealed that H3K9 methyltransferase G9a mediates lung cancer cell invasion through transactivating EpCAM, suggesting that the proteins affecting H3K9 methylation play important roles in lung cancer development [9]. Members of the heterochromatin protein 1 family (HP1α, β, and γ) bind to di- or trimethylated H3K9 on the promoter region and regulate gene expression [10, 11]. The misregulation of HP1 gene expression could cause many human diseases including cancer [11, 12]. Previous studies
9778
revealed that HP1γ was highly expressed in different cancers, including prostate and colon cancer [13, 14]. HP1γ overexpression has been identified as an independent prognostic marker and predicts biochemical recurrence after radical prostatectomy [13]. However, the clinical significance of HP1γ in NSCLC is not clear, and the molecular mechanisms in which HP1γ is employed to influence the development and relapse of NSCLC remain under investigation. In this study, we demonstrated that HP1γ expression is elevated in NSCLC. Further analysis in a cohort of 108 NSCLC patients indicated the positive association between HP1γ expression level and clinicopathological features and clinical outcome. By using NSCLC cell lines, we found that depleting HP1γ expression suppresses lung cancer cell proliferation and colony formation and upregulates pro-apoptosis gene expression levels.
Materials and methods Patients and specimen collection One hundred and eight lung cancer patients who received surgery at the Department of Thoracic Surgery of Nanjing Chest Hospital during 2007 to 2009 of May were included in this study. Lung specimens from cancer tissues and normal tissues (with >5 cm distance from the tumor edge) were immediately snap frozen in liquid nitrogen and then stored at −80 °C until use. None of the patients received any chemotherapy or radiotherapy before surgery. Clinicopathological data such as age, gender, smoking condition, size of tumor, histological differentiation, lymph nodal status, and pathological stage were obtained at the same time. The postoperative pathological staging was determined according to the 7th edition of the tumor–node–metastasis (TNM) classification. Histological type was determined according to the classification by the World Health Organization. The Research Ethics Committee of Nanjing Chest Hospital approved this protocol and informed written consent was obtained from all the participants. Immunohistochemical staining Resected specimens were fixed in 10 % formalin and paraffinembedded blocks were prepared. Five-micrometer-thick sections were cut from the specimens and placed on slides coated with poly-L-lysine. Sections were routinely deparaffinized with xylene and rehydrated in decreasing concentrations of alcohol. The sections were boiled in 10 mM citric acid buffer (pH 6.0) in a steaming cooker for 25 min for antigen retrieval. Endogenous peroxidase activity was blocked for 15 min of incubation with 3 % hydrogen peroxide. Following two washes in phosphate-buffered saline (PBS), sections were
Tumor Biol. (2014) 35:9777–9785
incubated in 10 % normal goat serum for 15 min. The sections were then incubated overnight at 4 °C with the rabbit antihuman HP1γ polyclonal antibody (1:300, Cell Signaling Technologies, Inc.). Then, the sections were incubated in biotinylated goat anti-rabbit IgG antibody for 15 min at room temperature. The streptavidin-biotinylated peroxidase solution was applied for 15 min at room temperature. Subsequently, the sections were washed with PBS and DAB colorization was applied until the color developed. Staining was monitored under a bright-field microscope and the reaction was stopped by washing with distilled water. The sections were then counterstained with hematoxylin. The slides were independently evaluated by two pathologists of Nanjing Chest Hospital, in a double-blinded manner, as described previously [15]. If discrepancies existed between these two reviewers, a consensus judgment was reached through discussion. The proportion of staining tumor cells in each selected field was determined by counting individual tumor cells at four randomly selected high magnification fields. The immunoreactivities were graded as (−), (+), (++), and (+++) according to the percentage of positive tumor cells identified: (−) represents 0 or 50 % tumor cells present. The immunoreactivities were graded according to the percentage of positive tumor cells identified. High expression of HP1γ was arbitrarily defined, if ≥40 % of the tumor cells showed granular staining at nuclear. RNA collection, reverse transcription, and quantitative RT-PCR RNA was extracted using TRIzol® (Cat no. 15596018, Invitrogen, Carlsbad, CA, USA) as per instruction and then reverse-transcribed by random primers using TaKaRa reverse transcription kit (Dalian, China). Real-time PCR (qRT-PCR) was carried out, using SYBR Green in an ABI 7500 StepOne Plus Real Time PCR instrument (Applied Biosystems, USA). The relative expression level of each target gene was normalized to β-actin in one sample. Primers used in our study were as follows: HP1γ: forward primer: 5′-AACCAAGAGGATTT GCCAGA-3′ and reverse primer: 5′-CTGGACAAGAATGC CAAGTT-3′; gadd45a: forward primer: 5′-GAGAGCAGAA GACCGAAAGGA-3′ and reverse primer: 5′-CACAACAC CACGTTATCGGG-3′; and β-actin: forward primer: 5′CATGTACGTTGCTATCCAGGC-3′ and reverse primer: 5′CTCCTTAATGTCACGCACGA-3′. Cell line and small interfering RNA transfection A549 and NCI-H1975 were purchased from ATCC (Manassas, VA, USA) and cultured in the DMEM containing 10 % FBS (Life Technologies) at 37 °C with 5 % CO2. Small
Tumor Biol. (2014) 35:9777–9785
9779
interfering RNA (siRNA) transfection was conducted using Lipofectamine 2000 (Cat no. 11668019, Life Technologies) as per the manufacturer’s instruction. siRNA sequences are as follows: HP1γ siRNA: 5′-GAGGCAGAGCCUGAAG AAUtt-3′; siNC: 5′-UUCUCCGAACGUGUCACGUtt-3′. Briefly, 2.5 μl Lipofectamine 2000 and 100 pM siRNA were mixed in 500 μl OPTI-MEM (Cat no. 31985088, Invitrogen) for 20 min, followed by the addition of the mixture into medium, and the final concentration of siRNA was 50 μM.
Kit (Invitrogen, Cat no. V13245). Briefly, an appropriate number of cells was seeded to ensure that more than 1×106 cells can be harvested after siRNA treatment, followed by PBS washing and harvesting by trypsinization. Then, the cells were washed with PBS twice and resuspended in 100 μl of binding buffer provided in the kit and incubated with Annexin V/PI for 15 min in the dark at room temperature. Finally, the cells were detected on FACS Calibur® Instrument (Becton Dickinson) and analyzed using the FlowJo 7.6 software.
MTT assay
Statistical analysis
A time course assay for proliferation was performed after gene knockdown. Cells were plated in 96-well plates at a density of 1,000/well. siRNAs were transfected into the cells 24 h after seeding and cell proliferation was detected by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay every 24 h after siRNA transfection. Ten microliters of 0.5 mg/ml MTT (Sigma, M5655) was added into each well for 4 h, then the cell medium was removed, and 100 μl of dimethyl sulfoxide was added to dissolve the formazan at 37 °C for 10 min. Optical density at 490 nm was measured by a multiwell spectrophotometer (BioTek).
The difference in the level of expression of HP1γ messenger RNA (mRNA) and protein was calculated using paired t test. Associations between HP1γ protein expression and the clinicopathological characteristics were analyzed using Mann– Whitney U test. Overall survival (OS) was calculated from the day of surgery to the date of the last follow-up or the date of death. Survival at the last follow-up date was censored. The postoperative survival curves were calculated using the Kaplan–Meier method and differences in the survival rates were analyzed using the log-rank test. All statistical procedures were performed using SPSS (version 16.0, SPSS Inc, Chicago). P0.05).
Table 1 HP1γ protein expression and clinicopathological factors in NSCLC patients Characteristic
Number of patients HP1γ protein (%) expression Low High P value
All patients Gender Male Female Age 3 Smoking status No smoking Smoking Histology type Squamous cell carcinoma Adenocarcinoma
108
71
37
85 (78.7) 23 (21.3)
60 11
25 12
0.042*
0.505 66 (61.1) 42 (38.9)
45 26
21 16
30 (27.8) 78 (72.2)
18 53
12 25
0.438
0.009* 40 (37.0) 68 (63.0)
20 51
20 17 0.311
47 (43.5) 53 (49.1)
Other 8 (7.4) Lymph node metastasis (pN) N0 65 (60.2) N1 +N2 +N3 43 (39.8) pTNM stages I 51 (47.3) II 40 (37.0) III 16 (14.8) IV 1 (0.9) *P65/3 cm/≤3 cm) N stage (N1+2+3/N0)
0.585 (0.330–1.039) 0.689 (0.351–1.353) 2.452 (0.414–4.537) 0.735 (0.414–1.303) 1.734 (1.383–2.665)
0.068 0.279 0.743 0.291 0.023*
Stage (I and II vs III and IV) HP1γ expression (high/low)
1.932 (1.312–2.675) 2.125 (1.644–2.760)
0.005* 0.008*
HR hazard ratio, 95% CI 95 % confidence interval. *P
