Sage Journals: Discover world-class research

Abstract

Objective

To determine the expression of the gene programmed cell death 5 (PDCD5) and its protein PDCD5 in laryngeal squamous cell carcinoma and to analyse possible correlations with clinicopathological parameters.

Methods

PDCD5 mRNA expression was assessed using reverse transcription–polymerase chain reaction and expression of PDCD5 protein was studied using Western blot analysis and immunohistochemistry in laryngeal squamous cell carcinoma and morphologically normal para-carcinoma tissue.

Results

A total of 41 laryngeal squamous cell carcinoma and 29 normal para-carcinoma tissue specimens were examined. Expression of both PDCD5 mRNA and PDCD5 protein was significantly reduced in laryngeal squamous cell carcinoma compared with normal tissue. Expression was correlated with clinical stage and histological grade, but was not associated with age, sex, location of primary tumour or the presence of lymph node metastases.

Conclusion

The expression of PDCD5 and its protein were shown to be reduced in laryngeal squamous cell carcinoma. The functional importance of PDCD5 as a regulating agent in cell apoptosis suggests that it may play a key role in tumour pathogenesis and development.

Keywords

Programmed cell death 5 PCDC5 TFAR19 expression laryngeal squamous cell carcinoma clinical significance

Introduction

Laryngeal squamous cell carcinoma is one of most frequent cancers in elderly men and its incidence has increased markedly during recent years.¹ The pathogenesis and development of laryngeal squamous cell carcinoma involves multiple factors, multiple genes and multiple stages,² and the exact mechanisms are unknown. The gene programmed cell death 5 (PDCD5), also known as TF-1 cell apoptosis-related gene 19 (TFAR19), was first identified in the leukaemic cell clone TF-1 at the Human Disease Gene Center of Peking University and was shown to participate in the process of cell apoptosis.³ Subsequent studies have confirmed the functional importance of PDCD5 in regulating cell apoptosis and its downregulation in a number of cancers, suggesting its clinical potential in the early diagnosis and treatment of malignancies.^4–7 The present study aimed to determine the expression of PDCD5 and its protein PDCD5 in laryngeal squamous cell carcinoma and to analyse possible correlations with clinicopathological parameters.

Patients and methods

Patients

Patients undergoing first-line surgical treatment for laryngeal squamous cell carcinoma in the Department of Otorhinolaryngology, Qilu Hospital, Shandong University, Shandong, China, between January 2010 and October 2012 who had not received preoperative radiotherapy, chemotherapy or biological therapy (defined as immunotherapy, biotherapy or biological response modifier treatment) were recruited to the study. Imaging studies, histological grading using the American Joint Committee on Cancer cancer staging system⁸ and clinical staging using the TNM criteria⁹ were performed before surgery.

The study was approved by the ethics committees of the Qilu Hospital of Shandong University, Shandong, China, and the Jinling Hospital of Nanjing University Medical School, Jiangsu, China. Written informed consent was obtained from all study participants.

Specimens

Specimens of tumour tissue and morphologically normal tissue adjacent to the tumour were harvested during open surgery: tumour tissue and normal tissue were identified visually and tumour tissue was harvested directly from the centre of the tumour. Normal tissue samples were obtained ≥5 mm from the margin of the tumour. Each fresh tissue specimen was divided into two: one half was fixed with ice-cold 10% paraformaldehyde for immunohistochemical analysis and other half was stored at −80℃ for reverse transcription–polymerase chain reaction (RT–PCR) and Western blot analysis.

RT–PCR

Expression of PDCD5 mRNA was measured using RT–PCR. Total RNA was extracted from thawed samples weighing approximately 1 g using TRIzol® reagent (Invitrogen, Shanghai, China) according to the manufacturer’s instructions. Reverse transcription was performed on 1 µg RNA using Moloney murine leukaemia virus reverse transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The resulting cDNA was then amplified by PCR using GoTaq® PCR Core Systems (Promega) according to the manufacturer’s instructions, with primers for human PDCD5 (forward primer 5′-CAACAGGAAGCAAAGCAC-3′, reverse primer 5′-GATCTTAACTTCTGCCTAGAC-3′; 296 bp) and the internal control gene β-actin (forward primer 5′-ATCATGTTTGAGACCTTCAACA-3′, reverse primer 5′- CATCTCTTGCTCGAAGTCCA-3′; 318 bp) (Invitrogen, Shanghai, China). The PCR protocol consisted of preliminary denaturation at 95℃ for 4 min, followed by 35 cycles of denaturation at 95℃ for 30 s, annealing at 56℃ for 30 s and elongation at 72℃ for 30 s, with a final elongation step at 72℃ for 5 min. The PCR products were electrophoresed on 1.5% agarose gel and stained with ethidium bromide. Expression levels were calculated after adjustment for the expression levels of the housekeeping gene β-actin using Kodak 1D Image Analysis software (Eastman Kodak, Rochester, NY, USA) and given as the ratio of PDCD5 to β-actin. Positive specimens were defined as those with a ratio of PDCD5 to β-actin >0.25.

Western blot analysis

Expression of PDCD5 protein was measured using Western blotting. Total protein was extracted from thawed tissue homogenates using TRIzol® reagent (Invitrogen) according to the manufacturer’s instructions. After quantification by the Bradford method,¹⁰ equal amounts of protein (30 µg) were separated using a sodium dodecyl sulphate–polyacrylamide gel electrophoresis kit with a 15% acrylamide gel (Pierce, Rockford, IL, USA) and transferred to nitrocellulose membranes using the Trans-Blot® system (Bio-Rad, Hercules, CA, USA) (2 h at 450 mA in 25 mM Tris-buffered saline, pH 6.8, 192 mM glycine and 20% methanol). The membranes were then blocked with 50 g/l skimmed milk powder overnight at 4℃ and incubated at 37℃ for 1 h with rabbit polyclonal antibody to human PDCD5 (Abcam, Cambridge, MA, USA) at a dilution of 1:150. After washing with Tris-buffered saline and Tween 20 (TBST) buffer (pH 7.6), the membranes were incubated with horseradish peroxidase-labelled goat anti-rabbit secondary antibody (Santa Cruz Biotechnology, Dallas, TX, USA) at a dilution of 1:500 with gentle agitation for 1 h at room temperature. The membranes were then rinsed thoroughly with TBST buffer and the bands were developed using 3,3′-diaminobenzidine (DAB) for 10 min. The bands were detected using the AlphaImager® HP system (ProteinSimple, Santa Clara, CA, USA); positive expression of PDCD5 was defined as the appearance of a clear band. Expression was quantified by densitometry using Quantity One® analysis software (Bio-Rad). The PDCD5 protein level was normalized to that of β-actin, which was used as an internal control.

Immunohistochemistry

Expression of PDCD5 protein was also analysed by immunohistochemistry using a streptavidin–biotin–peroxidase complex kit (MaiXin Bio, Fuzhou, China) according to the manufacturer’s instructions. Tissues were embedded in paraffin and cut into 4 µm sections. The sections were deparaffinized and incubated with rabbit anti-human PDCD5 antibody (Abcam) at a dilution of 1:600 overnight at 4℃ and then washed three times in phosphate-buffered saline (pH 7.4). The specimens were then incubated with biotinized goat anti-rabbit immunoglobulin G secondary antibody (Abcam) at 37℃ for 30 min. After a further three washes in phosphate-buffered saline, antibody binding was visualized using a DAB kit (Jingmei, Shenzhen, China) according to the manufacturer’s instructions and counterstained with haematoxylin. Negative control sections were incubated with secondary antibody alone.

The protein PDCD5 was identified as brown staining in the cytoplasm under light microscopy and areas with the most and well-distributed stained cells under low power were selected. The number of stained cells was counted in at least five randomly chosen high-power (×400 magnification) fields for each specimen. PDCD5 expression was quantified according to the percentage of cells stained and the intensity of staining. A score of 0 was assigned to specimens with no stained cells, <5% stained cells had a score of 1, 5–25% stained cells had a score of 2, 26–50% stained cells had a score of 3, and >50% stained cells had a score of 4. The intensity of staining was scored as 0 for no staining, 1 for weak staining, 2 for moderate staining and 3 for strong staining.¹¹ The product of these two scores was termed the staining score and was used to quantify PDCD5 expression. Each section was analysed by two experienced technicians who were blinded to the study group and the final result was obtained by consensus in cases of discrepancies.¹² Specimens with a staining score ≥1 were defined as being positive for PDCD5 protein expression.

Statistical analyses

Results were expressed as the mean ± SD. Intergroup differences were analysed using one-way analysis of variance. Non-parametric statistics were used for non-normally distributed results. The percentages of positive specimens in the two groups were compared using the Mann–Whitney U-test. Semiquantitative expression of PDCD5 mRNA and PDCD5 protein was compared in matched specimens from the two groups using the paired Student’s t-test. Correlation analysis was performed using the Pearson test or χ² test. A P-value < 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS software version 11.0 (SPSS Inc., Chicago, IL, USA).

Results

A total of 41 laryngeal squamous cell carcinoma specimens were collected from 41 patients (mean ± SD 56.35 ± 12.31 years); in addition, 29 normal para-carcinoma tissue specimens were collected from 29 of the 41 patients who consented to this.

Expression of PDCD5

Examples of expression of PDCD5 mRNA as detected by RT–PCR are shown in Figure 1. Expression was seen in a significantly lower proportion of carcinoma tissue specimens than in normal tissue specimens (P = 0.0038) (Table 1). Semiquantitative analysis of expression in the 29 paired samples produced similar results, with expression of PDCD5 mRNA being significantly higher in normal tissue than in carcinoma tissue (P = 0.013) (Table 2).

Figure 1.

Expression of programmed cell death 5 gene (PDCD5) mRNA in representative normal para-carcinoma tissue and laryngeal squamous cell carcinoma as detected by reverse transcription–polymerase chain reaction; data were normalized to β-actin.

Table 1.

Proportions of normal para-carcinoma and laryngeal squamous cell carcinoma specimens positive for expression of programmed cell death 5 (PDCD5) mRNA and PDCD5 protein.

Tissue	n	PDCD5 mRNA		PDCD5 protein
		PDCD5 mRNA		Western blot		Immunohistochemistry
		Positive specimens	Statistical significance^a	Positive specimens	Statistical significance^a	Positive specimens	Statistical significance^a
Normal	29	24 (82.8)	P = 0.0038	29 (100)	P = 0.0012	23 (79.3)	P = 0.0097
Carcinoma	41	20 (48.8)	P = 0.0038	18 (43.9)	P = 0.0012	20 (48.8)	P = 0.0097

Data presented as n (%) of specimens.

Using Mann–Whitney U-test.

Table 2.

Semiquantitative expression of programmed cell death 5 (PDCD5) mRNA and PDCD5 protein in 29 matched normal para-carcinoma and laryngeal squamous cell carcinoma specimens.

Tissue	PDCD5 mRNA		PDCD5 protein
	PDCD5 mRNA		Western blot		Immunohistochemistry
	Expression ratio^a	Statistical significance^b	Expression ratio^a	Statistical significance^b	Staining score	Statistical significance^b
Normal	0.72 ± 0.39	P = 0.013	0.57 ± 0.24	P = 0.027	6.21 ± 3.72	P = 0.0079
Carcinoma	0.81 ± 0.47	P = 0.013	0.42 ± 0.19	P = 0.027	4.03 ± 1.67	P = 0.0079

Data presented as mean ± SD.

Relative to β-actin.

Using paired Student’s t-test.

Expression of PDCD5 protein

A high proportion of normal tissue specimens were shown to express PDCD5 protein on Western blotting, whereas most tumour tissue specimens had lost the ability to express PDCD5 or showed only low expression (P = 0.0012) (Table 1; Figure 2). Semiquantitative analysis of PDCD5 protein expression in the 29 paired samples produced similar results, with expression being significantly higher in normal tissue than in carcinoma tissue (P = 0.027) (Table 2).

Figure 2.

Expression of programmed cell death 5 (PDCD5) protein in representative normal para-carcinoma tissue and laryngeal squamous cell carcinoma as detected by Western blot; data were normalized to β-actin.

These findings were confirmed on immunohistochemistry. PDCD5 protein was located in the cytoplasm but not in the nucleus. The proportion of cells expressing PDCD5 protein was significantly lower in carcinoma tissue than in normal tissue (P = 0.0097) (Table 1; Figure 3). Comparison of staining scores in the 29 paired samples showed that PDCD5 protein expression was significantly higher in normal tissue than in carcinoma tissue (P = 0.0079) (Table 2).

Figure 3.

Expression of programmed cell death 5 (PDCD5) protein in representative (A) normal para-carcinoma tissue and (B) laryngeal squamous cell carcinoma as detected by immunohistochemistry. Magnification × 400.

Correlation between PDCD5 mRNA and PDCD5 protein expression

A significant association was identified between PDCD5 mRNA expression and PDCD5 protein expression in laryngeal squamous cell carcinoma using the Pearson test (r = 0.695, P < 0.01).

Correlation between PDCD5 mRNA or PDCD5 protein expression and clinicopathological features

Associations between PDCD5 mRNA expression or PDCD5 protein expression and clinicopathological features in laryngeal squamous cell carcinoma were analysed using the χ² test. PDCD5 protein expression determined by Western blot or immunohistochemistry was not significantly associated with age, sex, location of primary or lymph node metastasis, but was significantly correlated with TNM stage and histological grade (Table 3). Moreover, PDCD5 mRNA expression was also associated with TNM stage and histological grade (Table 3).

Table 3.

Expression of programmed cell death 5 (PDCD5) mRNA and PDCD5 protein according to clinicopathological features in laryngeal squamous cell carcinoma specimens.

Parameter	n	PDCD5 mRNA		PDCD5 protein
		PDCD5 mRNA		Western blot		Immunohistochemistry
		Expression ratio^a	Statistical significance^b	Expression ratio^a	Statistical significance^b	Staining score	Statistical significance^b
Age, years
≤60	19	0.61 ± 0.32	NS	0.28 ± 0.17	NS	4.19 ± 2.50	NS
>60	22	0.53 ± 0.33	0.32 ± 0.19	3.72 ± 2.28
Sex
Male	34	0.41 ± 0.26	NS	0.31 ± 0.21	NS	3.85 ± 2.40	NS
Female	7	0.54 ± 0.29	0.38 ± 0.17	4.07 ± 2.35
Location of primary tumour
Supraglottic	10	0.41 ± 0.26	NS	0.37 ± 0.16	NS	4.15 ± 2.72	NS
Glottis	25	0.54 ± 0.29	0.43 ± 0.18	3.62 ± 2.08
Subglottic	6	0.41 ± 0.26	0.34 ± 0.11	3.43 ± 2.13
Histological grade
I	14	0.63 ± 0.34	P = 0.032	0.61 ± 0.32	P = 0.027	4.65 ± 2.69	P = 0.014
II	18	0.49 ± 0.29	0.43 ± 0.25	3.52 ± 2.76
III	9	0.24 ± 0.21	0.31 ± 0.12	2.12 ± 1.73
TNM stage
I + II	23	0.51 ± 0.27	P = 0.021	0.42 ± 0.22	P = 0.016	4.69 ± 2.578	P = 0.021
III + IV	18	0.34 ± 0.19	0.29 ± 0.17	2.93 ± 1.37
Metastases in nodes
No	20	0.41 ± 0.26	NS	0.43 ± 0.21	NS	4.39 ± 2.23	NS
Yes	21	0.54 ± 0.29	0.47 ± 0.26	3.52 ± 2.43

Data presented as mean ± SD.

Relative to β-actin.

Using χ² test.

NS, no statistically significant differences (P ≥ 0.05).

Discussion

Loss of signal transduction pathways is a frequent occurrence in cancer cells and several key components in cell apoptosis were first identified through their effects on pro-cancer transformation or mutation in cancer tissues.^13–15 Several studies have shown that PDCD5 promotes cell apoptosis mediated by various factors and inhibits cell proliferation.^3,5,16 The use of a variety of technologies has confirmed that PDCD5 is downregulated in many diseases, especially cancers. Spectrum analysis of gene expression in hereditary breast cancer tissues has shown that PDCD5 expression is increased in BRCA1-mutation cancers and decreased in BRCA2-mutation cancers.¹⁷ In addition, a study of differential gene expression using DNA chip technology demonstrated that some cell apoptosis-related genes, including PDCD5, were downregulated in hepatocellular carcinoma.¹⁸ It has also been reported that expression of PDCD5 in ovarian epithelial cancer tissue is lower than in normal ovary or benign ovarian tumour; moreover, PDCD5 expression decreased with increasing histological grade.⁴

In the present study, expression of both PDCD5 mRNA and PDCD5 protein in laryngeal squamous cell carcinoma was significantly lower than that in normal tissue (P = 0.0038 and P = 0.0012, respectively). These results were confirmed by immunohistochemical analysis. Further evaluation of associations between PDCD5 protein expression on Western blot and the pathological features of laryngeal squamous cell carcinoma showed that the reduction in expression was not associated with age, sex, location of the primary or the presence of lymph node metastases, but was associated with histological grade and TNM stage (P = 0.027 and P = 0.016, respectively). Moreover, the reduction in PDCD5 mRNA expression was also associated with histological grade and TNM stage (P = 0.032 and P = 0.021, respectively).

There are currently few studies on the apoptosis-related gene PDCD5 and the exact mechanisms of its role in apoptosis regulation are unknown. Fernandez et al.¹⁹ reported that when normal human breast epithelial cells were exposed in vitro to bisphenol A, an endocrine disruptor, expression of many genes associated with DNA repair were increased but expression of PDCD5 and BCL2L11, both of which are involved in apoptosis, were downregulated. In the osteosarcoma cell line MG-63, PDCD5 has been shown to induce apoptosis and G2 phase arrest.²⁰ In addition, expression of PDCD5 inhibited the Ras/Raf/MEK/ERK signalling pathway, leading to inhibition of cyclin B and cyclin-dependent kinase 1, with downregulation of ERK resulting in activation of caspase 3 and 9.²⁰ Furthermore, caspase 3 regulates Bax translocation from cytoplasm to mitochondria, which plays an essential role in apoptosis.²¹ In another study, PCDC5 was identified as a substrate for casein kinase 2 (CK2); PDCD5 was phosphorylated in vitro by both the CK2α subunit and the CK2 holoenzyme at residue S118.²² Transfection of the non-phosphorylatable mutant (S118A) impaired PDCD5 acceleration of either doxorubimicin- or ultraviolet-induced apoptosis in U2OS cells.²² On the other hand, when PDCD5 was transfected into gastric cancer cell line BG823, the stable transfection resulted in G2/M cell cycle arrest, increased apoptosis and nuclear translocation of PDCD5 and p53 after cisplatin treatment, indicating that regulation of PDCD5 may be a novel strategy for improving chemotherapeutic effects in cancer.²³

Expression of PDCD5 protein has been shown to be significantly associated with expression of p53.²⁴ As the pathway mediated by p53 is important in cancer-cell apoptosis, this association is consistent with PDCD5 having a role in apoptosis. It has been shown that PDCD5 can bind with p53 within residues 41–56 of the p53 TAD2 subdomain, while p53 binds preferentially to the positively charged surface region around the C-terminals of helices α3 and α5 and the N-terminal of helix α4 of PDCD5.²⁵

In conclusion, the present study found that expression of both PDCD5 mRNA and PDCD5 protein was significantly decreased in laryngeal squamous cell carcinoma, indicating that PDCD5 may have an important role in the pathogenesis and development of this cancer as a regulating agent in cell apoptosis. However, the mechanism of loss of expression of this pro-apoptosis gene in laryngeal squamous cell carcinoma is still unknown. In addition, the pathways via which PDCD5 exerts its functions remain to be elucidated.

Footnotes

Declaration of conflicting interest

The authors declare that there are no conflicts of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Sipaul

Birchall

Corfield

. What role do mucins have in the development of laryngeal squamous cell carcinoma? A systematic review. Eur Arch Otorhinolaryngol 2011; 268: 1109–1117.

Loyo

Pai

. The molecular genetics of laryngeal cancer. Otolaryngol Clin North Am 2008; 41: 657–672.

Liu

Wang

Zhang

. TFAR19, a novel apoptosis-related gene cloned from human leukemia cell line TF-1, could enhance apoptosis of some tumor cells induced by growth factor withdrawal. Biochem Biophys Res Commun 1999; 254: 203–210.

Zhang

Wang

Song

. Clinical and prognostic significance of lost or decreased PDCD5 expression in human epithelial ovarian carcinomas. Oncol Rep 2011; 25: 353–358.

Yang

Zhao

. Expression of programmed cell death 5 gene involves in regulation of apoptosis in gastric tumor cells. Apoptosis 2006; 11: 993–1001.

Spinola

Meyer

Kammerer

. Association of the PDCD5 locus with lung cancer risk and prognosis in smokers. J Clin Oncol 2006; 24: 1672–1678.

Nguyen

McCance

. Role of the retinoblastoma tumor suppressor protein in cellular differentiation. J Cell Biochem 2005; 94: 870–879.

American Joint Committee on Cancer. AJCC cancer staging manual, 7th edn. New York: Springer, 2010.

Sobin

Gospodarowicz

Wittekind

. International Union Against Cancer (UICC). TNM classification of malignant tumours, 7th edn. Oxford: Wiley-Blackwell, 2009.

10.

Sutton

Sutherland

Shnyder

. Improved preparation and detection of cytochrome P450 isoforms using MS methods. Proteomics 2010; 10: 327–331.

11.

Ryu

Lee

Chi

. Increased expression of cFLIP(L) in colonic adenocarcinoma. J Pathol 2001; 194: 15–19.

12.

Zhou

Liu

. Overexpression of cellular FLICE-inhibitory protein (FLIP) in gastric adenocarcinoma. Clin Sci (Lond) 2004; 106: 397–405.

13.

McCabe

Dlamini

. The molecular mechanisms of oesophageal cancer. Int Inununopharmacol 2005; 5: 1113–1130.

14.

Kountouras

Zavos

Chatzopoulos

. Apoptotic and anti-angiogenic strategies in liver and gastrointestinal malignancies. J Surg Oncol 2005; 90: 249–259.

15.

Yang

Matthews

Clair

. Tumorigenesis suppressor Pdcd4 down-regulates mitogen-activated protein kinase kinase kinase kinase 1 expression to suppress colon carcinoma cell invasion. Mol Cell Biol 2006; 26: 1297–1306.

16.

Rui

Chen

Zhang

. Transfer of anti-TFAR19 monoclonal antibody into HeLa cells by in situ electroporation can inhibit the apoptosis. Life Sci 2002; 71: 1771–1778.

17.

Hedenfalk

Duggan

Chen

. Gene-expression profiles in hereditary breast cancer. N Engl J Med 2001; 344: 539–548.

18.

Huang

. Insight into hepatocellular carcinogenesis at transcriptome level by comparing gene expression profiles of hepatocellular carcinoma with those of corresponding noncancerous liver. Proc Natl Acad Sci U S A 2001; 98: 15089–15094.

19.

Fernandez

Huang

Snider

. Expression and DNA methylation changes in human breast epithelial cells after bisphenol A exposure. Int J Oncol 2012; 41: 369–377.

20.

Han

Sun

Bai

. The anti-tumor role and mechanism of integrated and truncated PDCD5 proteins in osteosarcoma cells. Cell Signal 2012; 24: 1713–1721.

21.

Zhuge

Chang

. PDCD5-regulated cell fate decision after ultraviolet-irradiation-induced DNA damage. Biophys J 2011; 101: 2582–2591.

22.

Salvi

Chen

. Programmed cell death protein 5 (PDCD5) is phosphorylated by CK2 in vitro and in 293T cells. Biochem Biophys Res Commun 2009; 387: 606–610.

23.

Chen

Pan

. Transfection of PDCD5 effect on the biological behavior of tumor cells and sensitized gastric cancer cells to cisplatin-induced apoptosis. Dig Dis Sci 2012; 57: 1847–1856.

24.

Wang

Fan

. Cellular uptake of exogenous human PDCD5 protein. J Biol Chem 2006; 281: 24803–24817.

25.

Yao

Feng

Zhou

. NMR studies of the interaction between human programmed cell death 5 and human p53. Biochemistry 2012; 51: 2684–2693.

Expression and clinical significance of the programmed cell death 5 gene and protein in laryngeal squamous cell carcinoma

Abstract

Objective

Methods

Results

Conclusion

Keywords

Introduction

Patients and methods

Patients

Specimens

RT–PCR

Western blot analysis

Immunohistochemistry

Statistical analyses

Results

Expression of PDCD5

Expression of PDCD5 protein

Correlation between PDCD5 mRNA and PDCD5 protein expression

Correlation between PDCD5 mRNA or PDCD5 protein expression and clinicopathological features

Discussion

Footnotes

Declaration of conflicting interest

Funding

References