Introduction

Esophageal squamous cell carcinoma (ESCC) is one of the main reasons of cancer-related deaths in Iran and ranks as the fifth and sixth of the most common cancers in women and men, respectively [1, 2]. Developmental cell signaling pathways such as Wnt and notch play important roles in cell fate decision during development [3–5]. Notch family genes encode cell surface receptors. In a ligand binding-dependent manner, the receptors are activated through cleavages by some of the metalloproteases such as c-secretase and tumor necrosis factor-α-converting enzyme (TACE), releasing notch intracellular domain (NICD) into the cytoplasm. The NICD translocates into the nucleus where it functions as a transcription activator of notch signaling target genes. Targeted to the CSL (CBF1, Suppressor of Hairless or Lag-1) complex, NICD releases corepressors and recruits MAML1 coactivator, triggering the activation of notch signaling transcription complex [6, 7]. Hairy enhance of split-1 (HES1), cyclin D1, p21, p27, and c-Myc are some of the important notch target genes [8]. In the absence of NICD, the CSL complex binds to the regulatory region of target genes and recruits corepressors such as CBF1 and hairless/split ends (SPEN) [9, 10], which are associated with histone deacetylases to conserve the chromatin structure in a transcriptional inactive status.

Crosstalk between notch and other important cell signaling pathways such as Wnt and Sonic hedgehog (Shh) plays key roles in tumorigenesis [11–13]. The tumor growth and development is related to cancer stem-like cells (CSCs) where deregulated notch signaling leads to uncontrolled proliferation of CSCs to create a tumor mass [14]. The notch pathway keeps balance between cell proliferation and apoptosis. Therefore, it is clear that deregulation of the notch pathway may result in tumorigenesis [15]. The notch signaling pathway acts as a double-edged sword in tumorigenesis based on the cell context. In some malignancies such as lung, cervical, and colon cancers, it functions as an oncogene, while in some others such as hepatocellular carcinoma, it plays as a tumor suppressor [6, 7, 16]. Suppression of the notch signaling pathway will result in differentiation of cancer stem cells, and therefore, the notch pathway is suggested as an important target for cancer treatment [17].

HES proteins as basic helix-loop-helix (bHLH) transcription factors are one of the most important gene families which are targeted by the notch signaling pathway. This family consists of HES1–6 as members of the bHLH transcription factor repressors which play their roles through expressional regulation of cell cycle mediators such as p21 and cyclin D1. The HES1 has the most important role during the developmental processes [18, 19]. Although the expressional studies of HES1 in different malignancies have been reported, the impact of HES1 messenger RNA (mRNA) expression in ESCC progression and development is not elucidated yet. Therefore, in this study, we aimed to reveal the level of HES1 mRNA expression in ESCC and to evaluate its correlation with different clinicopathological features of the patients.

Materials and Methods

Tissue Samples

Fifty ESCC patients from the Emam Reza and Omid Hospitals of Mashhad University of Medical Sciences (MUMS) were enrolled in this study. RNA extraction from the tumoral and corresponding normal tissues was performed using the RNeasy Mini Kit (Qiagen, Hilden, Germany). All the cases had not received any chemo- and radiotherapy before the surgery. The ethic committee of MUMS confirmed the study, and all the patients gave their written consent to be enrolled. The normal and tumoral microdissected samples were conserved in RNAlater solution (Qiagen, Hilden, Germany) and histologically examined by expert pathologists based on the last TNM classification guideline [20].

RNA Extraction, cDNA Synthesis, and Quantitative RT-PCR

cDNA synthesis was performed after RNA extraction from the normal and tumoral tissues as described before [21]. Assessment of the level of HES1 mRNA expression was done using the primer sequences presented in Table 1 through a duplicate comparative real-time PCR (Stratagene Mx3000P, La Jolla, CA) SYBR Green (GENETBIO, Korea) method. The data were normalized using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) which has the lowest variability between the normal and tumor tissues of ESCC in comparison with the other housekeeping genes [22, 23]. The thermal profile involved an initial denaturation step of 95 °C for 10 min followed by 40 cycles of 95 °C (15 s), 52 °C (30 s), and 72 °C (30 s). Fold changes in HES1 mRNA expression were calculated using the ΔΔCt method [24], and overexpression and underexpression of the gene were defined for the tumors with more than +2-fold and less than −2-fold in the level of HES1 mRNA expression, respectively. The range in between was considered as normal expression.

Table 1 Primer set sequences used in real-time PCR
Full size table

Statistical Analysis

Statistical analysis was done using SPSS 16.0 (SPSS, Chicago, IL). Due to normal distribution of the samples, assessment of correlations between the level of HES1 mRNA expression and clinicopathological features was performed by the Pearson test, independent sample t test, and ANOVA. Statistical tests were considered significant with a p value <0.05.

Results

Study Population

Fifty ESCC patients including 26 males (52 %) and 24 females (48 %) were enrolled in this study. The patients’ age was scattered between 37 and 83 years (mean ± SD 62.80 ± 10.89). The tumor size ranged from 1 to 12 cm (mean ± SD 4.12 ± 1.95). The tumors were almost equally distributed in the lower 23 (46 %) and middle 27 (54 %) parts of the esophagus. Most of the tumors (49/50, 98 %) were classified into tumor stages of ΙΙ/ΙΙΙ, and 40 out of 50 tumor samples (80 %) were invaded to adventitia (T3). The majority of cases were moderately differentiated (32/50, 64 %). The number of tumors without any metastasis to the regional lymph nodes was more than the tumors with lymph node metastasis (28 vs. 22 cases, respectively).

Significant Downregulation of HES1 in ESCC

Assessment of HES1 expression was performed using real-time PCR (SYBR Green method), in which the level of HES1 expression in tumor samples was compared to the corresponding normal (tumor-free) esophageal tissues. The fold changes of HES1 mRNA expression in patients are shown as a scatterplot in Fig. 1. Thirteen out of 50 cases (26 %) had significant HES1 underexpression (P = 0.001), while HES1 overexpression was observed only in 4 (8 %) samples, and the other tumor tissues showed normal expression of HES1 mRNA (mean ± SD 0.30 ± 0.20). Figure 2 compares the different levels of HES1 mRNA expression in three groups (underexpressed, normal, and overexpressed samples), schematically.

Fig. 1
figure 1

Scatterplot represents descriptive analysis of relative gene expression of HES1 in ESCC patients

Full size image
Fig. 2
figure 2

Box plot represents relative mRNA expression of HES1 in ESCC patients. Median, lower/upper quartile, and highest/lowest observations of fold changes in patients with underexpressed, normal expressed, or overexpressed HES1 are shown in the box plot

Full size image

Correlations Between HES1 mRNA Expression and Clinicopathological Features

Correlations between the level of HES1 mRNA expression and clinicopathological features including the depth of tumor invasion, lymph node metastasis, tumor stage, grade, and location are shown in Table 2. There was a significant correlation between the depth of tumor invasion (T) and HES1 mRNA expression in which underexpressed tumors in T3 had lower levels of HES1 mRNA expression in comparison with the underexpressed tumors in T2 (−4.23 ± 0.763 vs. −2.55 ± 0.304, fold change) (p = 0.035). Most of underexpressed cases (9 out of 13, 69.2 %) were invaded to the adventitia classified as T3, histologically. There was not any significant correlation between the HES1 underexpression and stage of tumor. Although there was not any significant correlation between the tumor grade and HES1 mRNA expression (p = 0.45), most of HES1 underexpressed tumors were moderately differentiated (9 out of 13, 69.2 %). The underexpressed tumors were observed in the lower and middle esophagus equally. There was not any significant correlation between the level of HES1 mRNA expression and sex, tumor size, location, and metastasis status. HES1 overexpressed tumors had higher mean size in comparison with the underexpressed tumors (mean ± SD 5.63 ± 2.13 vs. 3.97 ± 0.50 cm), and the biggest tumor sample in size (12 cm) belonged to the HES1 overexpressed tissue.

Table 2 Correlation between the level of HES1 mRNA expression and different clinicopathological features in ESCC patients
Full size table

Discussion

The notch signaling pathway is one of the most important developmental signaling cascades involved in cell fate decision and differentiation in both embryonic and adult stem cells [25, 26]. It functions via the basic helix-loop-helix (bHLH) transcription factors such as HES gene family [27]. Among the members of HES family, HES1 has a key role in embryogenesis and regulation of differentiation in blood and a variety of tissues such as muscles [28–30]. Therefore, it can be expected that deregulation of HES1 expression may lead to disorganization of normal cell growth program [31]. It has been shown that HES1 functions as a tumor suppressor in some malignancies such as lung cancer and hepatocarcinoma [32]; however, different reports have showed the role of notch signaling in the progression and development of solid tumors including colon, prostate, and cervix cancers [33–35].

In this study, it was observed that the level of HES1 mRNA expression in ESCC was significantly lower compared to the corresponding margin normal esophageal tissues, suggesting that HES1 underexpression is probably related to ESCC progression. Such finding may ignore the function of the notch pathway in tumor progression through the HES1 gene expression [36]. Recently, we showed significant overexpression of MAML1 in ESCC and its correlation with aggressiveness and metastasis of the tumor cells [37]. Since MAML1 is the main coactivator of notch signaling transcription machinery, its overexpression leads us to hypothesize that notch signaling is active in ESCC. Therefore, underexpression of HES1, as one of the main notch signaling target genes, is apparently in contrast with the activity of this pathway in ESCC, suggesting that HES1 is not an ideal indicator of notch signaling activation in the disease. In this study, there was not only a significant correlation between the level of HES1 expression and depth of tumor invasion but also the number of HES1 underexpressed samples was more than the overexpressed cases. Since our previous report regarding overexpression of MAML1 [37] was done on the same study population, we compared the results and interestingly found that most of the HES1 underexpressed cases also showed normal or underexpression of MAML1. Having combined these data, we may extrapolate that HES1, as a suppressor transcription factor, normally suppresses some involved genes in ESCC progression and development, and when HES1 is underexpressed through deregulation of notch signaling pathway, the HES1 target genes experience uncontrolled expression leading to ESCC progression through advanced stages. Another interesting issue that can complicate judgment about HES1 gene expression status is the controlling role of other cell signaling pathways in HES1 expression. It has been shown that the HES1 expression is also regulated by the hedgehog signaling pathway, suggesting that the notch activation via the MAML1 is not lonely a justified guaranty for the HES1 overexpression and activation [38]. In hedgehog cell signaling pathway, HES1 gene expression is positively regulated by Gli1/2 in a notch-independent manner, relying to the complicated crosstalk between the signaling pathways [39, 40].

Our results showed that there was an inverse correlation between the HES1 gene expression and depth of tumor invasion especially in advanced stages of the disease. Therefore, such underexpression of HES1 in aggressive ESCC may be reflected from malfunction of the Shh pathway in these patients. Moreover, expression of HES1 is strictly dependent on the microRNA (miRNA) signature. It has been shown that several miRNAs such as miR-124 and miR-23b can regulate HES1 gene expression [41, 42]. Such regulation may be associated to the level of HES1 mRNA expression in ESCC.

In conclusion, we revealed the significant underexpression of HES1 in ESCC samples and showed its significant association with the depth of tumor invasion. To the best of our knowledge, this is the first report elucidating the expressional status of HES1 and its correlation with indices of poor prognosis in ESCC patients. Although these results cannot confirm the role of notch signaling in ESCC progression, definitely it cannot be ignored due to existent crosstalk between the notch and hedgehog cell signaling pathways to control HES1 gene expression as well as miRNA signature. Despite the lack of a consistent relation between the HES1 expression and notch pathway in ESCC patients, our study introduces HES1 underexpression as an independent prognostic marker for the disease.