Experimental and clinic-opathologic study on the relationship between transcription factor Egr-1 and esophageal carcinoma

Abstract

AIM: To observe the growth suppression effect of exogenous introduction of early growth response gene-1 (Egr-1 gene) on esophageal carcinoma tissue as well as on esophageal carcinoma cell line Eca109 and to explore the potential application of Egr-1 gene in gene therapy of tumor.

METHODS: Eukaryotic expression vector of PCMV-Egr-1 plasmid was introduced into Eca109 cell line which expressed no Egr-1 protein originally with lipofectamine transfection method. The introduction and expression of PCMV-Egr-1 plasmid into Eca109 cell line was confirmed by G418 selection culture, PCR amplification of neogene contained in the vector, Western blot analysis and immunocytochemical analysis. The cell growth curve, soft agar colony formation rate and tumorigenicity in SCID mice were examined to demonstrate the growth suppression effect of exogenous Egr-1 gene on Eca109 cell line. The Egr-1 mRNA and Egr-1 protein were also detected in 50 surgical specimens of esophageal carcinoma by in situ hybridization and immunohistochemistry.

RESULTS: Exogenous Egr-1 gene was introduced successfully into Eca109 cell line and expressed Egr-1 protein stably. The transfected Eca109 cell line grew more slowly than control Eca109 as shown by cell growth curves, the soft agar colony formation rate (4.0% vs 6.9%, P < 0.01) and the average growth rate of tumor in SCID mice (35.5 ± 7.6 vs 65.8 ± 7.6, P < 0.05). The expression level of Egr-1 mRNA and protein significantly increased in dysplastic epithelia adjacent to cancer rather than in cancer tissues (65.8% vs 20.0% by ISH and 57.9% vs 14.0% by IHC, P < 0.01).

CONCLUSION: Exogenous Egr-1 gene shows the strong effect of growth inhibition in Eca109 cell line. Egr-1 in the cancer tissue shows down-regulated expression that supports the inhibited function of Egr-1 in cancer growth and suggests Egr-1 may have an important role in gene therapy of esophageal carcinoma.

Key Words: esophageal neoplasms/pathology; tumor cells, cultured; genes, immediate early; gene expression; transfection

INTRODUCTION

Esophageal carcinoma is one of the most common malignant tumors in China[1-5]. Its pathogenesis and development are closely related with some of proto-oncogenes and their products[6-17]. Early growth response gene-1 (Egr-1 gene) is known as a member of immediate early gene (IEG) family characterized by rapid and transient expression in response to stimulation, modulating gene transcription positively or negatively depending on the cell types and playing an important role in the early cell growth. But its relationship with esophageal carcinoma is not well understood so far. In order to evaluate the growth inhibition in the cell line of esophageal carcinoma (Eca109) and analyze the expression of Egr-1 in the cancer tissue of esophageal carcinoma, the relationship between Egr-1 and esophageal carcinoma is explored by the methods of gene transfection, tumorigenicity in severe combined immunodeficient (SCID) mice and in situ hybridization.

MATERIALS AND METHODS

Human esophageal carcinoma cell line Eca109

Eca109 cell line from Chinese Academy of Medical Sciences was cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 100mL·L^-1 fetal bovine at 37 °C and 50 mL·L^-1 CO₂ incubator. The cancer tissue with adjacent mucosa and the esophageal mucosa in the upper surgical margin was from the same surgical specimen of 50 patients with esophageal carcinomas who had received neither chemotherapy nor radiotherapy before surgery. These were fixed in 40 mL·L^-1 neutral formalin with 1/1000 diethyl pyrocarbonate (DEPC, Sigma Chemical Co, USA) and embedded in paraffin, the sections were cut in 5 μm.

Eukaryotic expression vector of PCMV-Egr-1 plasmid

The plasmid was donated by Dr RP Huang (Molecular Medicine, Northwest Hospital, WA, USA). The final construct contains the neogene (5.5 kb fragment) driven by the respiratory syncytial virus (RSV) promoter and the Egr-1 gene (2.1 kb fragment) driven by the human cytomegalovirus (CMV) promoter. The plasmid was confirmed by amplification, purification and tested by cutting endonuclease.

Gene transfection

The eukaryotic expression vector of PCMV-Egr-1 plasmid was transfected to human esophageal carcinoma cell line Eca109 by the lipofectamine (Gibcobrl Company, USA) according to the manufacturer’s instructions. After transfection, the cells were trypsinized and reseeded at 1:2 ratio for selection culture with G418 at 600 mg·L^-1. Four weeks later, the resistant colonies were formed.

Detection of transfected cell neogene with polymerase chain reaction

Extracting the cell DNA followed by performing polymerase chain reaction (PCR) to amplify 327 bp of neogene with primer 1 (5’ACAAGATGGATTGCACGCAGG3’) and primer 2 (5’TTCTCGGCAGGAGCAAGGTGAG3’). The cycling procedure was: denatured at 95 °C for 1 min, annealed at 55 °C for 1.5 min, extended at 72 °C for 1 min, and after 30 cycles, lengthened at 72 °C for 5 min. The untransfected Eca 109 cell line was used as a negative control.

Western blot and immunocytocheimstry

Cells (3 × 10⁵) were harvested and dissolved by the addition of sodium dodecyl sulfate (SDS)-containing lysis buffer. The lysate was used for SDS-polyacrylamide gel electrophoresis on a 75 g·L^-1 gel. The proteins were electrophoretically transferred from the gel to NC membrane. The transferred membrane was treated with polycolonal antiserum against Egr-1 protein (1:200, Santa Cruz Biot Co, USA) followed by detection with peroxidase-labeled goat anti-rabbit (1:1000, Dako, USA). The resulting complexes were detected with the ECL reagent (Amersham Company, UK) according to the manufacturer’s instructions. The Egr-1 protein was detected by immunocytochemistry. The untransfected Eca 109 cell line was used as a negative control.

Detection of biological feactures in transfected cells

Growth curve assays Cells (1 × 10⁴) were seeded in each well of a 24-well plate and were allowed to grow for varying periods. The trypsinized cells were counted by cell counter plate. The growth assays were made by counting in triplicate on every other day of culture for up to 7 d followed by constructing the cell growth curves.

Soft agar culture The anchorage independent growth was examined by seeding 1000 cells in 3 g·L^-1 agar medium into 60 mm plates previously lined with 5 g·L^-1 agar medium. The plates (in triplicate and repeated twice) were cultured at 37 °C, 50 mL·L^-1 CO₂ incubator for 14 d. The average number of colony formation (any colony containing > 50 cells was counted as a colony) and the colony formation rate (the number of colonies/the number of seeded cells) were calculated.

Tumorigenicity in SCID mice Two groups were divided randomly. The experimental group that consisted of 6 mice was used for Eca 109 cells with transfected Egr-1 injection. The control group of 4 mice was used for Eca 109 cells without transfected Egr-1 injection. Trypsinized cells (5 × 10⁶) were injected into the subcutaneous sites on the shoulders of SCID mice. Animals were inspected at regular intervals for the appearance of visible tumors to measure the time of first appearance. Thirty days later, the mice were sacrificed and the tumors were carefully removed by blunt dissection. The tumors were weighed and their average growth rates were measured as mg·d^-1.

Egr-1 in situ hybridization

The expression of Egr-1 was detected by digoxigenin-labeled gene probe from a commercial kit (Boster Company, China) according to the manufacturer’s instructions. The human breast tissue and the mouse brain tissue were used as the positive control. Either the sections detected with incubation solution instead of the probe or the sections digested with ribonucleases (RNase) (10 mg·L^-1) before Egr-1 detection was designed for the negative control. The positive expression showed the brown staining signal in the cytoplasm.

Immunohistochemistry

Egr-1 was analyzed by using Egr-1 rabbit polyclonal antiserum with the SABC method according to the manufacturer’s instructions (Boster Company, China) following treatment with 3.3’-diaminobenzidine (DAB) staining. The human breast tissue and the mouse brain tissue were used as the positive control. Negative control was designed using phosphate-buffered saline (PBS) instead of Egr-1 rabbit polyclonal antiserum in detection. The positive expression showed the brown staining signal in nuclei.

Statistical analysis

The data were statistically analyzed using Student’s t test and the difference of results was analyzed by U test and χ² test.

RESULTS

Identification of plasmid

The eukaryotic expression vector of PCMV-Egr-1 plasmid was tested by PCR amplification, purification and restriction endonuclease and confirmed to be consistent with the plasmid map. With single digestion of Sac II, a 7.6 kb band was obtained which represents the whole lengh of the plasmid DNA. With double digestion of Sac II combining with Sma I, both 5.5 kb and 2.1 kb bands were obtained which represented the fragment of the vector and Egr-1 gene fragment respectively (Figure 1).

Figure 1 Identification of PCMV-Egr-1 plasmid. 1. Marker; 2. Uncut plasmid; 3. Cut with Sac II, showing 7.6 kp fragment of whole plasmid; 4. Cut with Sac II and Sma I, showing 5.5 kp fragment of vector and 2.1 kp fragment of Egr-1 gene.

Gene transfection

Eca109 was transfected with PCMV-Egr-1 plasmid by lipofectamine transfection method. Four weeks after G418 selection culture, the resistant colonies were formed. PCR indicated that 327 bp of neogene was shown in transfected Eca109 compared to the negative band in control Eca109 (Figure 2), demonstrating that PCMV-Egr-1 plasmid had been introduced into Eca109 cell and integrated into the genomic DNA.

Figure 2 PCR amplification of neogene. 1. Marker; 2-5. For transfected Eca109, showing 327 bp of neo-gene; 6. Negative control Eca109, no band was shown.

Detection of Egr-1 protein in transfected Eca109

Western blot The 80 kd of Egr-1 protein was shown for transfected Eca109 compared to the negative band of the control Eca109.

Immunocytochemistry The transfected Eca109 was positively brown stained in the nucleus compared to the negative staining for control Eca109 (Figure 3). The results demonstrated that exogenous Egr-1 gene introduced into Eca109 expressed high level of Egr-1 protein.

Figure 3 Positive Egr-1 protein in nuclei of transfected Eca109 cells. ICC × 400

Growth feature of transfected Eca109

Growth curves The transfected Eca109 grew much more slowly in DMEM medium than the control Eca109 cell (Figure 4).

Figure 4 Cell growth curves showing lower growth rate in transfected Eca109 cell than in the control Eca109 cell.

Soft agar growth assay Small and few colonies were formed slowly in transfected Eca109 compared to large and numerous colonies formed quickly in control Eca109. The colony formation rates were 4.0% and 6.9% in transfected Eca109 and in control Eca109 respectively (Table 1) which demonstrated that the anchorage independent growth ability of Egr-1 expressing Eca109 was reduced.

Table 1 Soft agar assays in Eca109 cell line.

Cell line	No. of seeded cell	No. of colonies				CFR
		1	2	3	Mean
Eca109	1000	65	71	73	69 ± 2.4	6.9%
Egr-1-Eca109	1000	40	43	39	40 ± 1.2	4.0%b

^bP < 0.01, vs Eca109, U test.

Tumorigenicity in SCID mice The tumorigenicity test showed that the tumors started to appear on the 21st day after injection in the six SCID mice injected with transfected Eca109, and on the 14th day in four SCID mice injected with control Eca109. The tumors of Egr-1 expressing Eca109 grew slowly with an average growth rate of 35.5 mg·d^-1, and the tumors of control Eca109 grew fast with an average growth rate of 65.8 mg·d^-1 (Figure 5, Table 2). The results demonstrated that the tumorigenicity in SCID mice of Egr-1 transfected Eca109 was inhibited.

Table 2 Tumorigenicity assays in Eca109 cell line.

Cell line	Tumorigenicity rate	Growth rate of tumors (mg·d-1)
		1	2	3	4	5	6	Mean
Eca109	4/4	83.3	70.0	46.6	63.3			65.8 ± 7.6
Egr-1-Eca109	6/6	20.0	13.3	33.3	66.6	43.0	36.7	35.5 ± 7.6a

^aP < 0.05, vs Eca109, Student’s t test.

Figure 5 The tumors of transfected Eca109 are smaller than that of control Eca109, in vitro, 30 d after injection in tumorigenicity test in SCID mice.

Egr-1 expression on esophageal tissues

In simple hyperplastic epithelia of esophageal mucosa, the expression of Egr-1 was found in the basal layer of the mucosa (Figure 6). In dysplastic epithelia, the expression of Egr-1 increased but significantly decreased in cancer tissues in which only a few cases of well-differentiated squamous cell carcinoma had the Egr-1 expression (Figure 7). The expression of Egr-1mRNA and proteins in the various pathological changes of esophagus are shown in Table 3.

Table 3 Expression of Egr-1mRNA and proteins in various pathological changes of esophagus, n (%).

Group	n	ISH	IHC
Simple hyperplasia	30	12 (40.0)	9 (30.0)
Dysplasia	38	25 (65.8)	22 (57.9)a
Esophageal cancer	50	10 (20.0)	7 (14.0)b

^aP < 0.05, vs simple hyperplasia, χ² test;

^bP < 0.01, vs dysplasia, χ² test.

Figure 6 Egr-1 protein expression in basal mucosal layer in simple hyperplastic epithelia of esophagus. IHC × 200

Figure 7 Positive Egr-1 mRNA in cytoplasm of esophageal squamous cell carcinoma. ISH × 400

DISCUSSION

As we know, the oncogenes and tumor suppressor genes are involved in the pathogenesis and development of esophageal carcinoma[18-32]. Recent studies in dicate that Egr-1 gene as a tumor related-gene is correlated with the tumor development. It was reported that Egr-1 proteins were decreased and even disappeared in several kinds of cancer tissues. Furthermore, the Egr-1 mRNA expression was consistent with the expression level of Egr-1 protein which had been verified in the down-regulation of Egr-1 occurred at the transcription level in cancer cells[33]. Southern blot analysis indicated no deletion, no rearrangement or mutation of Egr-1 on DNA level. The exogenous introduction of Egr-1 could inhibit the growth of tumor cells accompanied in a dose-dependent manner, for example, the ability of anchorage independent growth and tumorigenicity in SCID mice of human HT1080 fibrosarcoma without original Egr-1 was significantly inhibited after the exogenous introduction of Egr-1. If the antisense Egr-1 was introduced to the cells, it would inhibit the endogenous Egr-1 expression and promote the malignant transformation of the cells. In the present study, eukaryotic expression vector of PCMV-Egr-1 plasmid was introduced into Eca109 cell line that expressed no Egr-1 protein originally with Lipofectamine transfection method. The introduction and expression of PCMV-Egr-1 plasmid into Eca109 cell line was confirmed by G418 selection culture by which colony formation persisted in next generations and by PCR amplification of neogene contained in the vector. Furthermore, the strong expression of Egr-1 protein in transfected Eca109 was detected by Western blot and immunocytochemistry, which verified the success of gene transfection. Growth inhibition of the transfected cells shown by the growth curves, the colony formation rates in the soft agar and tumorigenicity in SCID mice demonstrated that exogenous Egr-1 gene inhibited the growth of Eca109.

Recent studies suggest that the mechanism of suppression of tumor growth by Egr-1 is the Egr-1 protein with a zinc-finger domain which regulates the transcription of many downstream genes by binding to the GC rich element in the promoter region and modulates genes transcription and various biological effects[34-45]. Egr-1 can compete with transcription activator SP1 in binding to an overlapping consensus binding motif in the promoter region of GCE which commonly exists in oncogene and tumor suppressor gene and abolishes the function of SP1, leading to the transcription inhibition of downstream genes and growth inhibition of tumor cells[46,47]. In addition, the exogenous Egr-1 may inhibit the growth by binding to the GEE of TGF-β 1 and activating the transcription of TGF-β 1 and subsequently activating p21 gene or by down-regulating Bcl-2 gene to influence the tumor cells[48,49]. Some studies indicated that the activation of some oncogenes e.g. wild type p53 gene, cell apoptosis, the TNF-α and the concentration of calcium may be involved in the mechanism of growth inhibition property of Egr-1[50,51]. On the whole, various and complicated mechanisms may be involved in the suppressive property of Egr-1 for tumor growth.

Egr-1 is one of the immediate early gene family which regulates the cellular growth and differentiation by activating Cyclin D1 to promote the cells from the G₀/G₁ phase into the G2/M phase[52]. The mechanism of down-expression of Egr-1 in cancer tissues is not clear up to now. It was hypothesized that the high concentration of Egr-1 proteins produced by overstimulation of Egr-1 interacted with the promoter region of EBS and subsequently the gene transcription was inhibited, which was concordant with the present results that showed strong expression of Egr-1 in dysplastic cells adjacent to esophageal carcinoma and weekly expression in cancer cells.

Since only downregulation expression without gene mutation occurred in Egr-1 DNA level, it was different from some tumor suppressor genes such as p3, p16 and Rb with mutation gene on DNA level. It is more convenient to introduce normal exogenous gene into the tumor cells than to repair mutation gene in gene therapy. The present study substantiates that exogenous Egr-1 as a target gene has a potential application in gene therapy of esophageal carcinoma.