Histopathology 2015, 66, 955–965. DOI: 10.1111/his.12585
High expression of transcriptional factor HoxB9 predicts poor prognosis in patients with lung adenocarcinoma Jun Zhan,1,2 Peng Wang,1,2 Miaomiao Niu,1,2 Yunling Wang,1,2 Xiang Zhu,1,3 Yongqing Guo4 & Hongquan Zhang1,2 1
Key Laboratory of Carcinogenesis and Translational Research, Ministry of Education, and State Key Laboratory of Natural and Biomimetic Drugs, Peking University Health Science Center, Beijing, China, 2Department of Anatomy, Histology and Embryology, Laboratory of Molecular Cell Biology and Tumor Biology, Beijing, China, 3Department of Pathology, Peking University Health Science Center, Beijing, China, and 4Department of Thoracic Surgery, Sino-Japan Friendship Hospital, Beijing, China Date of submission 16 August 2014 Accepted for publication 12 October 2014 Published online Article Accepted 17 October 2014
Zhan J, Wang P, Niu M, Wang Y, Zhu X, Guo Y, Zhang H (2015) Histopathology 66, 955–965. DOI: 10.1111/his.12585
High expression of transcriptional factor HoxB9 predicts poor prognosis in patients with lung adenocarcinoma Aims: HoxB9, as a Hox family member, is known to play important roles in embryonic development. Recent studies showed that HoxB9 is engaged in cancer progression. However, the role of Hoxb9 in lung adenocarcinoma is unknown. The purpose of this study is to investigate the expression and prognostic value of HoxB9 in patients with lung adenocarcinoma. Methods and results: The localization and expression of HoxB9 in lung adenocarcinoma were examined by immunohistochemistry. The correlation between HoxB9 expression levels with patient survival was assessed by Kaplan–Meier analysis. The epithelial–mesenchymal transition (EMT) markers and migratory ability were evaluated in HoxB9 up- and down-regu-
lated H1299 lung adenocarcinoma cells. HoxB9 was found to be localized predominantly in the cell nuclei and expressed in 21.3% of lung adenocarcinomas. A significant increase in HoxB9 intensity in the high stage of lung adenocarcinoma was observed (P < 0.01). Increased expression of HoxB9 was related to T classification, more lymph node metastasis and a shorter patient overall survival (P < 0.05). However, the expression level of HoxB9 was not correlated with age and gender. Functionally, HoxB9 up-regulated EMT-related molecules and promoted cell migration in H1299 cells. Conclusion: High expression of HoxB9 is a prognostic marker for lung adenocarcinoma patients.
Keywords: HoxB9, lung adenocarcinoma, prognosis, tumour marker
Introduction Lung cancer is the leading cause of cancer-related death in men and women worldwide.1 Despite the recent development of cancer treatments based on TNM staging, there has been little improvement in
Address for correspondence: H Zhang, Laboratory of Molecular Cell Biology and Tumor Biology, Department of Anatomy, Histology and Embryology, Peking University Health Science Center #38 Xue Yuan Road, Beijing 100191, China. e-mail: [email protected] © 2014 John Wiley & Sons Ltd.
the prognosis of lung cancer patients. The survival rate of all stages in non-small cell lung cancer (NSCLC) is approximately 15–20%,2 and even the survival rate of stage I patients is only approximately 60–80%.3 Many efforts have been made in searching and validating novel molecular prognostic biomarkers using expression profiling-based approaches.4–8 In the past decades, lung adenocarcinoma (LA), as one histological subtype of NSCLC, has become the most common histological type among all lung cancers diagnosed.9 Despite years of effort the survival of LA remains dismal, and the 5-year survival rate is only
956
J Zhan et al.
approximately 10%.10 It has been reported that LA is associated with mutations in epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma (KRAS), or anaplastic lymphoma kinase (ALK) gene rearrangement.11,12 Recently, LA was also found to be related to leucine-rich repeat-containing G-protein-coupled receptor 5 (LGR5),13 Jumonji domain-containing 6 (JMJD6),14 microRNA (miR)21 and miR15515 expressions. In addition, octamer binding transcription factor 3/4 (OCT4), prominin-1 (CD133) and aldehyde dehydrogenase (ALDH)16–18 were found to be involved in the initial stage of LA, and can be used as predictors for early stage LA. However, more markers that can be used for monitoring LA progression and therapeutic outcome are required. The Hox represent a subset of clustered homeodomain genes encoding transcription factors that determine cellular identity during development. So far, 39 HOX family genes have been identified and classified into four clusters as A, B, C and D, which were located separately on four different chromosomes.19 Hox family genes play important roles in embryonic development, especially in the patterning of the anterior to posterior axis, from the hindbrain to the end of the spine.19 Recently, studies have demonstrated that Hox family genes were found expressed in cancers including leukaemia,20 lung,21,22 prostate,23,24 breast,25,26 colon27,28 and ovarian cancers.29,30 HoxB9 is also known as a direct transcriptional target of WNT/transcription factor 4 (TCF4) that is involved in cell proliferation, differentiation, embryonic segmentation and limbpatterning.31 HoxB9 can promote breast cancer progression by inducing an epithelial–mesenchymal transition (EMT), with increased angiogenesis and distant metastasis.32 Several studies have suggested that HoxB9 protein is upregulated in breast cancer and promotes tumour progression.33 Importantly, HoxB9 is a Wnt target gene that mediates chemotactic invasion and colony outgrowth. HoxB9 also enhances the competence of lung adenocarcinoma cells to colonize the bones and the brain.34 HoxB9 was also reported to be decreased in gastric cancer tissues compared with adjacent normal tissues and was related to poor overall survival in gastric carcinoma patients.34 HoxB9 was reported to be expressed in lung cancer cell lines.35 However, so far the possible involvement of HoxB9 has been unknown in lung cancer progression. In this study, we investigated the expression of HoxB9 in tumour tissues of a cohort containing 75 patients with LAs in comparison with the corresponding adjacent normal tissues. Furthermore, we analysed the relationship between HoxB9 expression and the clinical
data as well as the survival of the patients. Finally, we evaluated the HoxB9 regulation on some EMT-related molecules and cell migration in H1299 cells.
Materials and methods ETHICS AND PATIENT TUMOUR SECTIONS
Lung cancer patient tumour tissues employed in the current study was approved by the Ethics Committee of Sino-Japan Friendship Hospital, Beijing, China (permit no. ZRLW-5). The procedures for handling human materials were in accordance with the ethical standards of the Helsinki Declaration of 1975, which was revised in 1983. Primary lung tumour tissue specimens (n = 75) were derived from patients who underwent segmentectomies or lobectomies with clear surgical margins from the Department of Thoracic Surgery Sino-Japan Friendship Hospital. Adjacent normal lung tissue samples (n = 75) were obtained at least 3 cm from tumour tissues in the same panel group of patients and were used as controls. These patients were subject to surgery from July 2004 to April 2009 and the follow-up for these patients used for survival analyses were ended in December 2011. The clinical staging of these patients followed NCCN Guidelines version 1.2013 staging for non-small cell lung cancer (available at www.nccn.com). No patients involved in this investigation received chemotherapy prior to surgery. The clinicopathological characteristics of all patients are summarized in Table 1. IMMUNOHISTOCHEMISTRY
All tumours used for this investigation were reviewed by at least two pathologists to confirm the diagnosis. Archived formalin-fixed paraffin-embedded lung tumour tissue sections were treated as described previously.21–23 After deparaffinization in xylene, the sections were rehydrated in graded alcohols. Endogenous peroxidase was quenched with 3% hydrogen peroxide in methanol at room temperature (25°C). The sections were placed in a 95°C solution of 0.01 M sodium citrate buffer (pH 6.0) for antigen retrieval. The primary antibody used for detecting HoxB9 was a rabbit anti- HoxB9 polyclonal antibody (Epitomics, Burlingame, CA, USA). The primary antibody was applied overnight at 4°C in a 1:50 dilution. PV9000 two-step plus a poly-horseradish peroxidase (HRP) antimouse/rabbit immunoglobulin (Ig)G detection system (Zhongshan Jinqiao, Jiaxing, China) was then applied. Detection was accomplished with the Dako Envision System (Dako, Glostrup, Denmark), © 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
Role of HoxB9 in lung adenocarcinomas
Table 1. Characteristics of clinical data of lung adenocarcinoma (LA) patients Characteristic Age median (range), years 61 (37–84)
Value
100.00
30–40
2
2.67
41–49
5
6.67
51–59
28
37.33
61–69
30
40.00
71–79
8
10.67
81–89
2
2.67
75
100.00
Male
41
54.67
Female
34
45.33
75
100.00
T1
9
12.00
T2a
41
54.67
T2b
5
6.67
T3
14
18.67
T4
6
8.00
N0
35
46.67
N1
17
22.67
N2
18
24.00
N3
5
6.67
M0
70
93.33
M1
5
6.67
75
100.00
IA
8
10.67
IB
19
25.33
IIA
14
18.67
IIB
2
2.67
IIIA
18
24.00
IIIB
9
12.00
IV
5
6.67
TNM category
Table 1. (Continued) Characteristic
Value
%
Survival status median (range), months 40 (2–89)
75
100.00
Alive
32
42.67
Dead
23
30.67
Unknown
20
26.67
%
75
Gender
957
TNM, tumour–node–metastasis.
followed by chromogen detection with diaminobenzidine (DAB). Haematoxylin was used for counterstaining. Negative controls were performed by omitting the use of primary antibody. SEMI-QUANTITATIVE SCORING
Semi-quantitative scoring of antibody staining on the slides was examined by one pathologist and rechecked by a second one, and scores were made without prior knowledge of clinical information. These scores were highly consistent with a correlation coefficient of 0.98. Nuclear HoxB9 staining on the tumour tissue section was evaluated using staining intensity and percentage of tumour cells stained (0 = not detected, 70% positive cells). We define ‘score 3’ as high expression and ‘scores 0, 1 and 2’ as low expression. CELL CULTURE AND GENE TRANSFECTION
Clinical stage
© 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
The lung adenocarcinoma cell line H1299 was purchased from the Cell Collection Center of Peking Union Medical School in China. Cells were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) with 10% foetal bovine serum (FBS) and 50 lg/ml gentamycin. HoxB9 full-length cDNA was cloned from a human placenta cDNA library, as described previously.36 SiRNAs and the scramble siRNA used in the experiments were described previously.36 WESTERN BLOTTING
Phosphate-buffered saline (PBS) 1% Tween-20, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate (TDS) lysis buffer with the presence of protease cocktail inhibitor (Roche Diagnostics GmbH, Mannheim, Germany) was
958
J Zhan et al.
used to extract total cell lysates. The primary antibodies anti-HoxB9 monoclonal antibody (mAb) (1:1000 dilution; Millipore, Billerica, CA, USA), anti-E-cadherin rabbit antibody (1:1000 dilution; Epitomics, Burlingame, CA, USA), anti-N-cadherin mAb (1:10000 dilution, clone M2; Sigma, St Louis, MO, USA) and anti-vimentin mAb (1:1000 dilution; Millipore) were incubated with the membranes separately under rotation. After thorough washing, membranes were incubated further with corresponding secondary antibodies recognizing either rabbit or mouse Ig (Jackson Laboratories, West Grove, PA, USA). Finally, the bands were visualized by enhanced chemoluminescence (Pierce, Rockford, IL, USA).
t-test was used for paired studies. Results were considered statistically significant at P < 0.05 using SPSS version 14.0 (SPSS Inc., Chicago, IL, USA).
Results HOXB9 WAS EXPRESSED IN NUCLEI OF TYPE II PNEUMOCYTES IN NORMAL LUNG TISSUE
We examined Hox9 expression in non-cancerous lung tissues and found that HoxB9 was scattered in normal lung tissue (Figure 1A). Under high magnification, we found that HoxB9 was expressed mainly in nuclei of some type II pneumocytes (Figure 1B, arrowed).
WOUND-HEALING ASSAY
Cells were seeded in six-well plates to 100% confluence. After starving for 10 h, a wound was induced by scratching the cell cultures with a 5-ll pipette tip. Following three rinses with PBS to remove the detached cells, cells were cultured in medium without serum. Images were captured immediately from each well, and again after 3 h and 6 h using a microscope (Nikon Corporation, Tokyo, Japan) in four random fields at 910 magnification. The width of the wound at these specific locations was visualized on each plate to quantify the rate of cell migration. STATISTICAL ANALYSIS
HoxB9 expression differences among various subgroups were determined using the Kruskal–Wallis rank sum test. Patient survival was calculated using the Kaplan–Meier analysis and comparisons were made using the log-rank test. The Cox univariate and multivariate analyses were used to examine the significance of other factors related to survival. Student’s
A
HOXB9 EXPRESSION WAS INCREASED IN LA
To detect the HoxB9 expression in LA, we examined a cohort of 75 patients with an age distribution of 37– 84 years. We observed that HoxB9 was expressed predominantly in nuclei of LA cells with concomitant cytoplasmic distribution. Nuclear HoxB9 staining on tumour tissue slides was evaluated using the following criteria: 0 = not detected (Figure 2A), 1 = weak (Figure 2B), 2 = moderate (Figure 2C) and 3 = strong (Figure 2D). We determined that 0–2 represents low HoxB9 expression and 3 represents high HoxB9 expression, detected in 59 (78.67%) and 16 (21.33%) samples, respectively. However, low and high expressions of HoxB9 were detected in 74 (98.67%) cases and one (1.33%) case of the paired adjacent non-cancerous tissue specimens, respectively. By comparing the two groups, HoxB9 expression in LA was found to be increased significantly (P < 0.01) (Figure 2E). These findings indicated that HoxB9 expression is enhanced in LA patients; the data suggest that HoxB9 may play a role in the progression of LAs.
B
Figure 1. HoxB9 is expressed in normal lung tissue. Immunohistochemical (IHC) staining was performed to examine HoxB9 expression in normal lung tissue. A, Adjacent non-cancer tissues, B, enlarged normal lung tissue. HoxB9 is highly expressed in the nuclei of II alveoli epithelial in normal lung tissues (arrowed). © 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
Role of HoxB9 in lung adenocarcinomas
B
C
D
P = 0.0248
2.0 1.5 1.0 0.5
ar
m
ci
al
no
lu
m
ng
a
(n
(n
=7
=7
5)
5)
0.0
oc
or
en ad ng Lu
Figure 2. HoxB9 is expressed in lung adenocarcinoma (LA). Immunohistochemical (IHC) staining was performed to examine the expression of HoxB9 in LA and representative photographs are shown. A, Negative expression of HoxB9; B, weak expression of HoxB9; C, moderate expression of HoxB9; D, high expression of HoxB9; E, HoxB9 expression was higher than in the adjacent noncancer tissues.
N
E
Expression level of HOXB9
A
959
HOXB9 EXPRESSION WAS ASSOCIATED WITH
HOXB9 EXPRESSION WAS OF PROGNOSTIC VALUE
CLINICOPATHOLOGICAL VARIABLES IN LA
IN LA
Associations between HoxB9 expression level and clinicopathological variables are summarized in Table 2. The most obvious correlation was identified between HoxB9 expression and LA clinical stage, in which the expression level of HoxB9 was increased from stage I to stages II, III and IV (P < 0.0001). With regard to TNM stage, the expression level of HoxB9 was found related to both T (P < 0.05) and N (P < 0.01). Similarly, the higher tumour stage correlated with higher HoxB9 expression. However, HoxB9 expression was not found to be related significantly with patients’ gender (P = 0.111) or age (P = 0.311). These data indicate that HoxB9 expression was correlated with the tumour stage in LA.
To determine if HoxB9 expression was correlated with patient survival, we examined patients with available follow-up data of up to 92 months. We found that 23 (30.67%) patients died during follow-up, and 32 (42.67%) patients were alive and censored. However, for various reasons, 22 (26.67%) patients were not available for further contact. The post-operative median survival time of the follow-up patients was 40.1 months, butpatients with low or high HoxB9 expression had median survival times of 42.5 and 29.82 months, respectively. In a Kaplan–Meier analysis, we found that the survival of patients with high HoxB9 expression was shorter than those who expressed low HoxB9 (P = 0.0458) (Figure 3). Fur-
© 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
960
J Zhan et al.
Table 2. Correlation of HoxB9 expression with clinicopathological features in lung adenocarcinoma (LA) Characteristic
Low-expression group
Normal lung tissue
74
98.67%
1
1.33%
LA
59
78.67%
16
21.33%
Gender Male
32
42.67%
9
12.00%
27
36.00%
7
9.33%
28
37.33%
7
9.33%
31
41.33%
9
12.00%
6
8.00%
3
4.00%
T2a
36
48.00%
5
6.67%
T2b
4
5.33%
1
1.33%
T3
8
10.67%
6
8.00%
T4
7
9.33%
1
1.33%
N0
30
40.00%
5
6.67%
N1
11
14.67%
3
4.00%
N2
12
16.00%
6
8.00%
N3
3
4.00%
2
2.67%
M0
57
76.00%
13
17.33%
M1
2
2.67%
3
4.00%
6
8.00%
2
2.67%
IB
18
24.00%
1
1.33%
IIA
13
17.33%
1
1.33%
IIB
2
2.67%
0
0.00%
IIIA
16
21.33%
6
8.00%
IIIB
6
8.00%
3
4.00%
IV
2
2.67%
3
4.00%
44
80.00%
11
20.00%
Alive
28
37.33%
4
5.33%
Dead
16
21.33%
7
9.33%
Female Age (years)
High expression of transcriptional factor HoxB9 predicts poor prognosis in patients with lung adenocarcinoma Jun Zhan,1,2 Peng Wang,1,2 Miaomiao Niu,1,2 Yunling Wang,1,2 Xiang Zhu,1,3 Yongqing Guo4 & Hongquan Zhang1,2 1
Key Laboratory of Carcinogenesis and Translational Research, Ministry of Education, and State Key Laboratory of Natural and Biomimetic Drugs, Peking University Health Science Center, Beijing, China, 2Department of Anatomy, Histology and Embryology, Laboratory of Molecular Cell Biology and Tumor Biology, Beijing, China, 3Department of Pathology, Peking University Health Science Center, Beijing, China, and 4Department of Thoracic Surgery, Sino-Japan Friendship Hospital, Beijing, China Date of submission 16 August 2014 Accepted for publication 12 October 2014 Published online Article Accepted 17 October 2014
Zhan J, Wang P, Niu M, Wang Y, Zhu X, Guo Y, Zhang H (2015) Histopathology 66, 955–965. DOI: 10.1111/his.12585
High expression of transcriptional factor HoxB9 predicts poor prognosis in patients with lung adenocarcinoma Aims: HoxB9, as a Hox family member, is known to play important roles in embryonic development. Recent studies showed that HoxB9 is engaged in cancer progression. However, the role of Hoxb9 in lung adenocarcinoma is unknown. The purpose of this study is to investigate the expression and prognostic value of HoxB9 in patients with lung adenocarcinoma. Methods and results: The localization and expression of HoxB9 in lung adenocarcinoma were examined by immunohistochemistry. The correlation between HoxB9 expression levels with patient survival was assessed by Kaplan–Meier analysis. The epithelial–mesenchymal transition (EMT) markers and migratory ability were evaluated in HoxB9 up- and down-regu-
lated H1299 lung adenocarcinoma cells. HoxB9 was found to be localized predominantly in the cell nuclei and expressed in 21.3% of lung adenocarcinomas. A significant increase in HoxB9 intensity in the high stage of lung adenocarcinoma was observed (P < 0.01). Increased expression of HoxB9 was related to T classification, more lymph node metastasis and a shorter patient overall survival (P < 0.05). However, the expression level of HoxB9 was not correlated with age and gender. Functionally, HoxB9 up-regulated EMT-related molecules and promoted cell migration in H1299 cells. Conclusion: High expression of HoxB9 is a prognostic marker for lung adenocarcinoma patients.
Keywords: HoxB9, lung adenocarcinoma, prognosis, tumour marker
Introduction Lung cancer is the leading cause of cancer-related death in men and women worldwide.1 Despite the recent development of cancer treatments based on TNM staging, there has been little improvement in
Address for correspondence: H Zhang, Laboratory of Molecular Cell Biology and Tumor Biology, Department of Anatomy, Histology and Embryology, Peking University Health Science Center #38 Xue Yuan Road, Beijing 100191, China. e-mail: [email protected] © 2014 John Wiley & Sons Ltd.
the prognosis of lung cancer patients. The survival rate of all stages in non-small cell lung cancer (NSCLC) is approximately 15–20%,2 and even the survival rate of stage I patients is only approximately 60–80%.3 Many efforts have been made in searching and validating novel molecular prognostic biomarkers using expression profiling-based approaches.4–8 In the past decades, lung adenocarcinoma (LA), as one histological subtype of NSCLC, has become the most common histological type among all lung cancers diagnosed.9 Despite years of effort the survival of LA remains dismal, and the 5-year survival rate is only
956
J Zhan et al.
approximately 10%.10 It has been reported that LA is associated with mutations in epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma (KRAS), or anaplastic lymphoma kinase (ALK) gene rearrangement.11,12 Recently, LA was also found to be related to leucine-rich repeat-containing G-protein-coupled receptor 5 (LGR5),13 Jumonji domain-containing 6 (JMJD6),14 microRNA (miR)21 and miR15515 expressions. In addition, octamer binding transcription factor 3/4 (OCT4), prominin-1 (CD133) and aldehyde dehydrogenase (ALDH)16–18 were found to be involved in the initial stage of LA, and can be used as predictors for early stage LA. However, more markers that can be used for monitoring LA progression and therapeutic outcome are required. The Hox represent a subset of clustered homeodomain genes encoding transcription factors that determine cellular identity during development. So far, 39 HOX family genes have been identified and classified into four clusters as A, B, C and D, which were located separately on four different chromosomes.19 Hox family genes play important roles in embryonic development, especially in the patterning of the anterior to posterior axis, from the hindbrain to the end of the spine.19 Recently, studies have demonstrated that Hox family genes were found expressed in cancers including leukaemia,20 lung,21,22 prostate,23,24 breast,25,26 colon27,28 and ovarian cancers.29,30 HoxB9 is also known as a direct transcriptional target of WNT/transcription factor 4 (TCF4) that is involved in cell proliferation, differentiation, embryonic segmentation and limbpatterning.31 HoxB9 can promote breast cancer progression by inducing an epithelial–mesenchymal transition (EMT), with increased angiogenesis and distant metastasis.32 Several studies have suggested that HoxB9 protein is upregulated in breast cancer and promotes tumour progression.33 Importantly, HoxB9 is a Wnt target gene that mediates chemotactic invasion and colony outgrowth. HoxB9 also enhances the competence of lung adenocarcinoma cells to colonize the bones and the brain.34 HoxB9 was also reported to be decreased in gastric cancer tissues compared with adjacent normal tissues and was related to poor overall survival in gastric carcinoma patients.34 HoxB9 was reported to be expressed in lung cancer cell lines.35 However, so far the possible involvement of HoxB9 has been unknown in lung cancer progression. In this study, we investigated the expression of HoxB9 in tumour tissues of a cohort containing 75 patients with LAs in comparison with the corresponding adjacent normal tissues. Furthermore, we analysed the relationship between HoxB9 expression and the clinical
data as well as the survival of the patients. Finally, we evaluated the HoxB9 regulation on some EMT-related molecules and cell migration in H1299 cells.
Materials and methods ETHICS AND PATIENT TUMOUR SECTIONS
Lung cancer patient tumour tissues employed in the current study was approved by the Ethics Committee of Sino-Japan Friendship Hospital, Beijing, China (permit no. ZRLW-5). The procedures for handling human materials were in accordance with the ethical standards of the Helsinki Declaration of 1975, which was revised in 1983. Primary lung tumour tissue specimens (n = 75) were derived from patients who underwent segmentectomies or lobectomies with clear surgical margins from the Department of Thoracic Surgery Sino-Japan Friendship Hospital. Adjacent normal lung tissue samples (n = 75) were obtained at least 3 cm from tumour tissues in the same panel group of patients and were used as controls. These patients were subject to surgery from July 2004 to April 2009 and the follow-up for these patients used for survival analyses were ended in December 2011. The clinical staging of these patients followed NCCN Guidelines version 1.2013 staging for non-small cell lung cancer (available at www.nccn.com). No patients involved in this investigation received chemotherapy prior to surgery. The clinicopathological characteristics of all patients are summarized in Table 1. IMMUNOHISTOCHEMISTRY
All tumours used for this investigation were reviewed by at least two pathologists to confirm the diagnosis. Archived formalin-fixed paraffin-embedded lung tumour tissue sections were treated as described previously.21–23 After deparaffinization in xylene, the sections were rehydrated in graded alcohols. Endogenous peroxidase was quenched with 3% hydrogen peroxide in methanol at room temperature (25°C). The sections were placed in a 95°C solution of 0.01 M sodium citrate buffer (pH 6.0) for antigen retrieval. The primary antibody used for detecting HoxB9 was a rabbit anti- HoxB9 polyclonal antibody (Epitomics, Burlingame, CA, USA). The primary antibody was applied overnight at 4°C in a 1:50 dilution. PV9000 two-step plus a poly-horseradish peroxidase (HRP) antimouse/rabbit immunoglobulin (Ig)G detection system (Zhongshan Jinqiao, Jiaxing, China) was then applied. Detection was accomplished with the Dako Envision System (Dako, Glostrup, Denmark), © 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
Role of HoxB9 in lung adenocarcinomas
Table 1. Characteristics of clinical data of lung adenocarcinoma (LA) patients Characteristic Age median (range), years 61 (37–84)
Value
100.00
30–40
2
2.67
41–49
5
6.67
51–59
28
37.33
61–69
30
40.00
71–79
8
10.67
81–89
2
2.67
75
100.00
Male
41
54.67
Female
34
45.33
75
100.00
T1
9
12.00
T2a
41
54.67
T2b
5
6.67
T3
14
18.67
T4
6
8.00
N0
35
46.67
N1
17
22.67
N2
18
24.00
N3
5
6.67
M0
70
93.33
M1
5
6.67
75
100.00
IA
8
10.67
IB
19
25.33
IIA
14
18.67
IIB
2
2.67
IIIA
18
24.00
IIIB
9
12.00
IV
5
6.67
TNM category
Table 1. (Continued) Characteristic
Value
%
Survival status median (range), months 40 (2–89)
75
100.00
Alive
32
42.67
Dead
23
30.67
Unknown
20
26.67
%
75
Gender
957
TNM, tumour–node–metastasis.
followed by chromogen detection with diaminobenzidine (DAB). Haematoxylin was used for counterstaining. Negative controls were performed by omitting the use of primary antibody. SEMI-QUANTITATIVE SCORING
Semi-quantitative scoring of antibody staining on the slides was examined by one pathologist and rechecked by a second one, and scores were made without prior knowledge of clinical information. These scores were highly consistent with a correlation coefficient of 0.98. Nuclear HoxB9 staining on the tumour tissue section was evaluated using staining intensity and percentage of tumour cells stained (0 = not detected, 70% positive cells). We define ‘score 3’ as high expression and ‘scores 0, 1 and 2’ as low expression. CELL CULTURE AND GENE TRANSFECTION
Clinical stage
© 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
The lung adenocarcinoma cell line H1299 was purchased from the Cell Collection Center of Peking Union Medical School in China. Cells were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) with 10% foetal bovine serum (FBS) and 50 lg/ml gentamycin. HoxB9 full-length cDNA was cloned from a human placenta cDNA library, as described previously.36 SiRNAs and the scramble siRNA used in the experiments were described previously.36 WESTERN BLOTTING
Phosphate-buffered saline (PBS) 1% Tween-20, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate (TDS) lysis buffer with the presence of protease cocktail inhibitor (Roche Diagnostics GmbH, Mannheim, Germany) was
958
J Zhan et al.
used to extract total cell lysates. The primary antibodies anti-HoxB9 monoclonal antibody (mAb) (1:1000 dilution; Millipore, Billerica, CA, USA), anti-E-cadherin rabbit antibody (1:1000 dilution; Epitomics, Burlingame, CA, USA), anti-N-cadherin mAb (1:10000 dilution, clone M2; Sigma, St Louis, MO, USA) and anti-vimentin mAb (1:1000 dilution; Millipore) were incubated with the membranes separately under rotation. After thorough washing, membranes were incubated further with corresponding secondary antibodies recognizing either rabbit or mouse Ig (Jackson Laboratories, West Grove, PA, USA). Finally, the bands were visualized by enhanced chemoluminescence (Pierce, Rockford, IL, USA).
t-test was used for paired studies. Results were considered statistically significant at P < 0.05 using SPSS version 14.0 (SPSS Inc., Chicago, IL, USA).
Results HOXB9 WAS EXPRESSED IN NUCLEI OF TYPE II PNEUMOCYTES IN NORMAL LUNG TISSUE
We examined Hox9 expression in non-cancerous lung tissues and found that HoxB9 was scattered in normal lung tissue (Figure 1A). Under high magnification, we found that HoxB9 was expressed mainly in nuclei of some type II pneumocytes (Figure 1B, arrowed).
WOUND-HEALING ASSAY
Cells were seeded in six-well plates to 100% confluence. After starving for 10 h, a wound was induced by scratching the cell cultures with a 5-ll pipette tip. Following three rinses with PBS to remove the detached cells, cells were cultured in medium without serum. Images were captured immediately from each well, and again after 3 h and 6 h using a microscope (Nikon Corporation, Tokyo, Japan) in four random fields at 910 magnification. The width of the wound at these specific locations was visualized on each plate to quantify the rate of cell migration. STATISTICAL ANALYSIS
HoxB9 expression differences among various subgroups were determined using the Kruskal–Wallis rank sum test. Patient survival was calculated using the Kaplan–Meier analysis and comparisons were made using the log-rank test. The Cox univariate and multivariate analyses were used to examine the significance of other factors related to survival. Student’s
A
HOXB9 EXPRESSION WAS INCREASED IN LA
To detect the HoxB9 expression in LA, we examined a cohort of 75 patients with an age distribution of 37– 84 years. We observed that HoxB9 was expressed predominantly in nuclei of LA cells with concomitant cytoplasmic distribution. Nuclear HoxB9 staining on tumour tissue slides was evaluated using the following criteria: 0 = not detected (Figure 2A), 1 = weak (Figure 2B), 2 = moderate (Figure 2C) and 3 = strong (Figure 2D). We determined that 0–2 represents low HoxB9 expression and 3 represents high HoxB9 expression, detected in 59 (78.67%) and 16 (21.33%) samples, respectively. However, low and high expressions of HoxB9 were detected in 74 (98.67%) cases and one (1.33%) case of the paired adjacent non-cancerous tissue specimens, respectively. By comparing the two groups, HoxB9 expression in LA was found to be increased significantly (P < 0.01) (Figure 2E). These findings indicated that HoxB9 expression is enhanced in LA patients; the data suggest that HoxB9 may play a role in the progression of LAs.
B
Figure 1. HoxB9 is expressed in normal lung tissue. Immunohistochemical (IHC) staining was performed to examine HoxB9 expression in normal lung tissue. A, Adjacent non-cancer tissues, B, enlarged normal lung tissue. HoxB9 is highly expressed in the nuclei of II alveoli epithelial in normal lung tissues (arrowed). © 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
Role of HoxB9 in lung adenocarcinomas
B
C
D
P = 0.0248
2.0 1.5 1.0 0.5
ar
m
ci
al
no
lu
m
ng
a
(n
(n
=7
=7
5)
5)
0.0
oc
or
en ad ng Lu
Figure 2. HoxB9 is expressed in lung adenocarcinoma (LA). Immunohistochemical (IHC) staining was performed to examine the expression of HoxB9 in LA and representative photographs are shown. A, Negative expression of HoxB9; B, weak expression of HoxB9; C, moderate expression of HoxB9; D, high expression of HoxB9; E, HoxB9 expression was higher than in the adjacent noncancer tissues.
N
E
Expression level of HOXB9
A
959
HOXB9 EXPRESSION WAS ASSOCIATED WITH
HOXB9 EXPRESSION WAS OF PROGNOSTIC VALUE
CLINICOPATHOLOGICAL VARIABLES IN LA
IN LA
Associations between HoxB9 expression level and clinicopathological variables are summarized in Table 2. The most obvious correlation was identified between HoxB9 expression and LA clinical stage, in which the expression level of HoxB9 was increased from stage I to stages II, III and IV (P < 0.0001). With regard to TNM stage, the expression level of HoxB9 was found related to both T (P < 0.05) and N (P < 0.01). Similarly, the higher tumour stage correlated with higher HoxB9 expression. However, HoxB9 expression was not found to be related significantly with patients’ gender (P = 0.111) or age (P = 0.311). These data indicate that HoxB9 expression was correlated with the tumour stage in LA.
To determine if HoxB9 expression was correlated with patient survival, we examined patients with available follow-up data of up to 92 months. We found that 23 (30.67%) patients died during follow-up, and 32 (42.67%) patients were alive and censored. However, for various reasons, 22 (26.67%) patients were not available for further contact. The post-operative median survival time of the follow-up patients was 40.1 months, butpatients with low or high HoxB9 expression had median survival times of 42.5 and 29.82 months, respectively. In a Kaplan–Meier analysis, we found that the survival of patients with high HoxB9 expression was shorter than those who expressed low HoxB9 (P = 0.0458) (Figure 3). Fur-
© 2014 John Wiley & Sons Ltd, Histopathology, 66, 955–965.
960
J Zhan et al.
Table 2. Correlation of HoxB9 expression with clinicopathological features in lung adenocarcinoma (LA) Characteristic
Low-expression group
Normal lung tissue
74
98.67%
1
1.33%
LA
59
78.67%
16
21.33%
Gender Male
32
42.67%
9
12.00%
27
36.00%
7
9.33%
28
37.33%
7
9.33%
31
41.33%
9
12.00%
6
8.00%
3
4.00%
T2a
36
48.00%
5
6.67%
T2b
4
5.33%
1
1.33%
T3
8
10.67%
6
8.00%
T4
7
9.33%
1
1.33%
N0
30
40.00%
5
6.67%
N1
11
14.67%
3
4.00%
N2
12
16.00%
6
8.00%
N3
3
4.00%
2
2.67%
M0
57
76.00%
13
17.33%
M1
2
2.67%
3
4.00%
6
8.00%
2
2.67%
IB
18
24.00%
1
1.33%
IIA
13
17.33%
1
1.33%
IIB
2
2.67%
0
0.00%
IIIA
16
21.33%
6
8.00%
IIIB
6
8.00%
3
4.00%
IV
2
2.67%
3
4.00%
44
80.00%
11
20.00%
Alive
28
37.33%
4
5.33%
Dead
16
21.33%
7
9.33%
Female Age (years)
