Yong- Ping Cui, Jian-Bo Wang, Xin-Yu Zhang, Mei-Xia Bi, Li-Ping Guo, Shih-Hsin Lu, Department of Etiology and Carcinogenesis, Tumor Hospital, Peking Union Medical College & Chinese Academy of Medical Sciences, Beijing 100021, China
Supported by G1998051204 the Major State Basic Research Development Program of China, No.G1998051204
Correspondence to: Shih-Hsin Lu, Ph.D, Department of Etiology and Carcinogenesis, Tumor Hospital, Peking Union Medical College & Chinese Academy of Medical Sciences, Beijing 100021, China. shlu@public.bta.net.cn
Telephone: +86-10-67712368    Fax: +86-10-67712368
Received: 2002-11-19    Accepted: 2003-01-16

Abstract
AIM: To identify esophageal cancer related gene2 (ECRG2) associated proteins and their possible interactions with ECRG2 gene.

METHODS: In the yeast forward two-hybrid system, ECRG2 was fused with the DNA-binding domain (DBD) of Gal4 and human fetal liver cDNA library was fused with the transcriptional activation domain (AD) of Gal4. We performed a high-stringency scale procedure to screen ECRG2 against human fetal liver cDNA library and characterized positives by sequence analysis.

RESULTS: We found the following 9 putatively associated proteins. They were metallothionein2A, metallothionein1H, metallothionein1G, ferritin, erythrocyte membrane protein band4.2, mitochondrial ribosomal protein S12, hypothetical protein FLJ10101, and a novel gene whose cDNA was found to have no strong homology to any other previously characterized gene whose DDBJ/EMBL/GenBank accession number is AF422192 mapped to human chromosome 14q31.

CONCLUSION: MT, a potential interaction partner for ECRG2, might be involved in the regulation of cell proliferation and apoptosis, and in various physiological processes. Determination of a reliability score for each single protein-protein interaction, especially interaction of ECRG2 and MT, permits the assignment of ECRG2 and unannotated proteins to biological pathways. A further understanding of the association between ECRG2 and MT should facilitate the functions of ECRG2 gene.

Cui YP, Wang JB, Zhang XY, Bi MX, Guo LP, Lu SH. Using yeast two-hybrid system to identify ECRG2 associated proteins and their possible interactions with ECRG2 gene. World J Gastroenterol  2003; 9(9): 1892-1896
http://www.wjgnet.com/1007-9327/9/1892.asp

INTRODUCTION
Esophageal cancer (EC) is one of the most common malignancies worldwide with the highest mortality and prevalencet in certain areas of China with a higher incidence^[1-2]. It has been found some of tumor suppressor genes and oncogenes are involved in the EC initiation and development^[3,4]. However, so far no gene directly related to EC has been identified. By comparing the differential gene expression between normal esophageal epithelia and esophageal cancer using the technique of mRNA differential display, we have cloned a new gene ECRG2 (GenBank Accession No. AF268198) whose expression was found to be down-regulated in some malignant tissues such as esophageal carcinoma tissue, colon cancer and brain tumor tissues, and most common in esophageal cancer tissues. Although SMART online (http://www.smart.embl-heidelberg.de) has shown that ECRG2 gene contains a characteristic KAZAL-type conserved domain and belongs to the KAZAL-type related serine proteinase inhibitor family, little has been known about its function in normal cellur activities, other than suppressing neoplasia. In order to further reveal its biological roles, we therefore tried to identify ECRG2 associated proteins in the present study by the GAL4-based yeast two-hybrid system using the full-length ECRG2 cDNA as a bait to screen the human fetal liver cDNA library.

MATERIALS AND METHODS
Yeast strains
Matchmaker GAL4 two-hybrid system 3 and vector pACT2 containing the human fetal liver cDNA library were obtained from Clontech. Yeast strain AH109 (MATa, trp 1-901, leu2-3, 112, ura 3-52, his3-200, gal4, gal80, LYS2: GAL1_UAS-GAL1_TATA-HIS3, GAL2_UAS-GAL2_TATA-ADE2, URA3: MEL1_UAS-MEL1_TATA-LacZ) was used to screen the library and to verify protein-protein interactions, which could eliminate false positives by using three reporters-ADE2, HIS3, and MEL1 (or LacZ)-under the control of GAL4 upstream activating sequences (_UASs) and _TATA boxes. Among these reporters ADE2 provided strong nutrition selection, HIS3 gave a selection reducing false positive incidence, and MEL1 or LacZ encoding b-galactosidase could be assayed on X-a-gal indicator plates.

Plasmid constructs
Vector pGBkT7 expressing proteins were fused with amino acids 1-147 of the GAL4 DNA binding domain (DNA-BD), pGADT7 expressing proteins were fused with amino acids 768-881 of the GAL4 activation domain (AD). The control plasmids pGBKT7-53, pGBKT7-lam, pGADT7-T, pCL1 were from Clontech. Plasmid pGBKT7-ECRG2 encoding full-length (85 amino acids) ECRG2 gene fused in frame with the GAL4 DNA binding domain was constructed by inserting the PCR-generated fragment into the EcoRI and BamHI sites of pGBKT7. Plasmid pGADT7-ECRG2 encoding full-length ECRG2 gene fused in frame with GAL4 activation domain was constructed by inserting the PCR-generated fragment into the EcoRI and BamHI sites of pGADT7.

Library tittering and amplification
For tittering library, an aliquot of the library plasmid was thawed and mixed on ice by gentle vortexing. Then 1ml of library plasmid was transferred to 1 ml of LB broth, which was mixed and named dilution A (1:10³). By the same way, another library plasmid diluent dilution B (1:10⁶) was obtained. The prewarmed LB/amp plates were prepared by inoculating them in 1 ml of dilution A mixed with 50 ml of LB broth, or 50 ml and 100 ml of dilution B where respectively incubated on LB/amp plates at 30 ℃ for 48 h. The library titer was calculated as follows: Dilution A: counted colonies×10³×10³=cfu/ml, Dilution B: (counted colonies/plating volume)×10³×10³×10³ =cfu/ml. For the library amplification, LB/amp plates were inoculated with the library plasmids at such a high density that the resulting colonies could be easily confluent (-20 000-40 000 cfu per 150-mm plate) and reach at least 2-3 times as the library original plasmid number to ensure the better screening results. After incubated at 30 °C for 48 h, colonies were scraped, mixed with adequate volume of LB broth and shake-cultured in a flask at 30 °C for 2 h. One-third of the library culture was set aside for plasmid preparation.

Verification of activation of reporter gene by pGBKT7-ECRG2
Plasmid pGBKT7-ECRG2 was independently transformed into strain AH109. The transformants were assayed for b-galactosidase activity by selecting them on SD/-Trp/-Leu/X-a-Gal. Positive control, pCL1, was performed in parallel.

Library transformation and screening
Plasmid pGBkT7-ECRG2 was used as a bait in two-hybrid screening of human fetal liver cDNA libraries by MATCHMAKER Two-Hybrid System 3 protocol (Clontech). The yeast strain AH109 was sequencely transformed with pGBKT7-ECRG2 and a human fetal liver cDNA library in the pACT2 vector (Clontech) was obtained by the lithium acetate method. Transformants expressing both the bait and interacting prey proteins were selected on SD/-Trp/-Leu/-His/-Ade and incubated at 30 °C for 5-7 days. b-galactosidase activity was tested using the filter lift assay to identify the positive colones.

Isolating plasmid DNA from putatively positive yeast clones, rescuing AD/library plasmids and retesting protein interaction in yeast
Approximately 3×10⁶ colonies were screened and 146 positive clones were identified. cDNA inserts of the positive clones were amplified by PCR using primers complementary to the sequence of pACT2 vector (5'TACC ACT ACA ATG GAT3' and 5'GTG AAC TTG CGG GGT TTT TCA GTA TCT ACG A3'). Subsequently, pACT2-cDNA constructs were isolated from positive yeast colonies, as recommended by the supplier, transformed into super-competent E. coli DH5a by electroporation, grown under selection, re-isolated and analyzed by restriction digests. The uniquely purified constructs were then re-tested against the original pGBKT7-ECRG2 bait construct. To ensure the interactions were specific, the positive clones were also tested against an irrelevant bait protein laminC and grown on SD/-Ade/-His/-Leu/-Trp/X-a-Gal to test the specificity of interactions.

Sequence Ad/library inserts and blast online (http://www.ncbi.nlm.nih.gov/blast)
The positive inserts were sequenced and analyzed by comparison to the GenBank sequence data bank.

RESULTS
Library tittering and amplification
The titer of human fetal liver cDNA library was 3.17×10⁹cfu/ml counted by dilution method. The number of separate colonies was 1.05×10⁷. We amplified it in 150 mm500 plates, about 1108 (3.5 ml original cDNA library) colonies and the library plasmids were successfully isolated.

Verification of activation reporter genes by pGBKT7-ECRG2
Plasmid pGBKT7-ECRG2 was independently transformed into AH109. After b-galactosidase activity assay, the AH109 transformed pGBKT7-ECRG2 did not appear blue colonies, but the positive control, pCL1, appeared as is shown in Figure 1. It was verified that pGBKT7-ECRG2 construct did not activate reporter genes and was suitable for the yeast two-hybrid system (Figure 1).

Figure 1(PDF) Colony-lift filter assay for b-galactosidase activity. Left: b-galactosidase activity of AH109 transformed with pGBKT7-ECRG2. Right: b-galactosidase activity of AH109 transformed with pCL1 (positive control). The results showed that pCL1 could activate reporter genes, but pGBKT7-ECRG2 could not.

Yeast two-hybrid screen of cellular proteins interacting with ECRG2 protein
To identify proteins associated with the ECRG2 protein, ECRG2's ORF was synthesized as a translational fusion of a DNA-binding domain (DBD) and used as the bait for screening of a human fetal liver yeast two-hybrid cDNA library. Of 3×10⁶ transformants screened, 146 clones grew in the absence of tryptophan, lencine, histidine, adenine and expressed b-galactosidase activity. pACT2/cDNA plasmids were successfully isolated and duplicates were eliminated by HaeIII digestion (Figure 2). After elimination, 26 uniquely positive clones were further retested for specificity of b-galactosidase expression. After retransformation, 9 independent positive clones were identified and sequenced (Table 1).

Figure 2(PDF) Positive colonies screened by yeast two-hybrid system using full-length cDNA of ECRG2 as baits. A: PCR products of positive clones digested with HaeIII restriction enzyme. Lane 1: lDNA/EcoRI+HindIII Marker, Lanes 2-16: positive colonies, B: Trp⁺/Leu⁺/His⁺/Ade⁺ positive clone growing on the SD/Trp^-/Leu^-/His^-/Ade^- plate, C: colony-lift filter assay for b-galactosidase activity.

Table 1  Yeast two-hybrid screening using ECRG2 as baits

^a: Transformation efficiency (transformants/µg) = transformation yield ÷ amount of library DNA in µg). ^b: Transformation yield (total transformants) = [(colonies/plate) ÷ (volume/plate)] ×[(volume of total reaction) ÷ (dilution factor)]. ^c: b-galactosidase activity of the positive colonies assayed by b-galactosidase filter lift assay.

Figure 3(PDF) Blast results of MT2A.
Figure 4(PDF) Blast results of AF422192.

Bioinformatics
Blast online (http://www.ncbi.nlm.nih.gov/blast) showed the genes identified by the yeast two-hybrid approach were metallothionein 2A(MT2A), metallothionein 1H, metallothionein 1G, ferritin, erythrocyte membrane protein band 4.2, hemoglobin, mitochondrial ribosomal protein S12, hypothetical protein FLJ10101, and a novel gene whose cDNA was found to have no strong homology to any other previously characterized gene whose GenBank accession No.is AF422192.

DISCUSSION
Protein-protein interactions played important roles in almost all events that took place in a cell^[6]. Because proteins often assembled into large complexes to perform discrete activities, the characterization of the interaction pattern of a protein could provide considerable assistance in the elucidation of the functions of that protein^[7]. The availability of complete genome sequences now permits the development of tools for functional biology on a proteomic scale. Among those, the yeast two-hybrid system is the choice to detect protein-protein interactions^[8,9]. The ECRG2 gene is a novel candidate of tumor suppressor gene identified from human esophageal carcinoma. Identifying the function of ECRG2 gene product may provide opportunities to elucidate the esophageal cancer mechanisms and its role in tumor development and progression. The yeast-two-hybrid approach could find novel partners for known function proteins and identify the function of a novel protein by identifying well-characterized interacting partners^[10]. So, we searched for associated proteins with a yeast-two-hybrid system using the ECRG2 cDNA fragment as baits.
    Our results showed that ECRG2 gene did not activate transcription by itself and was suitable for yeast two-hybrid. On screening a human fetal liver cDNA library, we identified 9 putative clones as associated proteins, which included metallothionein 2A (MT2A), metallothionein 1H, metallothionein 1G. Metallothioneins (MTs) are a family of low molecular weight, cysteine-rich, metal ion-binding proteins which are wildly distributed in various species. MTs were thought to be involved in heavy-metal detoxification, intracellular trace elements storage and scavenging free radicals. Recently, emerging data suggested that MTs had a close relationship with tumors. They might play important roles in carcinogenic and apoptotic process and differentiation of tumor cells^[11-23]. In addition, MTs were attributed to affording tumor cell resistance to some important chemotherpeutic agents^[24]. Using immunohistochemical-staining method, MTs have been localized intensively in various types of human tumors in organs and tissues such as skin, kidneys, prostate, tests, gallbladder, colon, breast and endometrium^[25-34]. Since human metallothioneins are closely linked with tumor, it is possible for us to understand the cellular functions of the ECRG2 protein through its linkage to MT2A. The other associated clone is a novel gene whose cDNA was found to have no strong homology to any other previously characterized gene whose GenBank Accession No.is AF422192. There are also 5 clones including ferritin, erythrocyte membrane protein band 4.2, hemoglobin, mitochondrial ribosomal protein S12 and hypothetical protein FLJ10101. Determination of a reliability score for each single protein-protein interaction, especially interaction of ECRG2 and MTs, permits the assignment of ECRG2 and unannotated proteins to biological pathways. A further understanding of the association between ECRG2 and MT should facilitate the functions of ECRG2 gene.

REFERENCES
1    Lu SH. Alterations of oncogenes and tumor suppressor genes in esophageal cancer in China. Mutat Res 
      2000; 467: 343-353
2    Zhou J, Zhao LQ, Xiong MM, Wang XQ, Yang GR, Qiu ZL, Wu M, Liu ZH. Gene expression profiles at different stages 
      of human esophageal squamous cell carcinoma. World J Gastroenterol 2003; 9: 9-15
3    Montesano R, Hollstein M, Hainaut P. Genetic alterations in esophageal cancer and their relevance to etiology 
      and pathogenesis: a review. Int J Cancer 1996; 69: 225-235
4    Xiong XD, Xu LY, Shen ZY, Cai WJ, Luo JM, Han YL, Li EM. Identification of differentially expressed proteins 
      between human esophageal immortalized and carcinomatous cell lines by two-dimensional electrophoresis 
      and MALDI-TOF-mass spectrometry. World J Gastroenterol 2002; 8: 777-781
5    Su T, Liu HL, Lu SH. Cloning and identification of cDNA fragments related to human esophageal cancer. 
      Zhonghua Zhongliu Zazhi 1998; 20: 254-257
6    Blackstock WP, Weir MP. Proteomics: quantitative and physical mapping of cellular proteins. Trends Biotechnol 
      1999; 17: 121-127
7    Emmert-Buck MR, Gillespie JW, Pawletz CP, Ornstein DK, Basrur V, Appella E, Wang QH, Huang J, Hu N, Taylor P, 
      Petricoin EE 3rd. An approach to proteomic analysis of human tumors. Mol Carcinog 2000; 27: 158-165
8    Fromont-Racine M, Rain JC, Legrain P. Toward a functional analysis of the yeast genome through exhaustive two-
      hybrid screens. Nature Genet 1997; 16: 277-282
9    Toby GG, Golemis EA. Using the yeast interaction trap and other two-hybrid-based approaches to study protein-
      protein interactions. Methods 2001; 24: 201-217
10  Chambers G, Lawrie L, Cash P, Murray GI. Proteomics: a new approach to the study of disease. J Pathol 
      2000; 192:280-288
11  Zhang XH, Takenaka I. Incidence of apoptosis and metallothionein expression in renal cell carcinoma. Br J Urol 
      1998; 81: 9-13
12  Tan Y, Sinniah R, Bay BH, Singh G. Metallothionein expression and nuclear size in benign, borderline, and 
      malignant serous ovarian tumours. J Pathol 1999; 189: 60-65
13  Takaba K, Saeki K, Suzuki K, Wanibuchi H, Fukushima S. Significant overexpression of metallothionein and cyclin D1 
      and apoptosis in the early process of rat urinary bladder carcinogenesis induced by treatment with N-butyl-N-
      (4-hydroxybutyl) nitrosamine or sodium L-ascorbate. Carcinogenesis 2000; 21: 691-700
14  Jayasurya A, Bay BH, Yap WM, Tan NG, Tan BK. Proliferative potential in nasopharyngeal carcinoma: correlations 
      with metallothionein expression and tissue zinc levels. Carcinogenesis 2000; 21: 1809-1812
15  Hiura T, Khalid H, Yamashita H, Tokunaga Y, Yasunaga A, Shibata S. Immunohistochemical analysis of metallothionein 
      in astrocytic tumors in relation to tumor grade, proliferative potential, and survival. Cancer 1998; 83: 2361-2369
16  Hishikawa Y, Kohno H, Ueda S, Kimoto T, Dhar DK, Kubota H, Tachibana M, Koji T, Nagasue N. Expression 
      of metallothionein in colorectal cancers and synchronous liver metastases. Oncology 2001; 61: 162-167
17  Abdel-Mageed AB, Agrawal KC. Activation of nuclear factor kappaB: potential role in metallothionein-mediated 
      mitogenic response. Cancer Res 1998; 58: 2335-2338
18  Aloia TA, Harpole DH Jr, Reed CE, Allegra C, Moore MB, Herndon JE D Amico TA. Tumor marker expression is predictive 
      of survival in patients with esophageal cancer. Ann Thorac Surg 2001; 72: 859-866
19  Joseph MG, Banerjee D, Kocha W, Feld R, Stitt LW, Cherian MG. Metallothionein expression in patients with small 
      cell carcinoma of the lung: correlation with other molecular markers and clinical outcome. Cancer 2001; 92: 836-842
20  Ebert MP, Gunther T, Hoffmann J, Yu J, Miehlke S, Schulz HU, Roessner A, Korc M, Malfertheiner P. Expression 
      of metallothionein II in intestinal metaplasia, dysplasia, and gastric cancer. Cancer Res 2000; 60: 1995-2001
21  Jin R, Chow VT, Tan PH, Dheen ST, Duan W, Bay BH. Metallothionein 2A expression is associated with cell proliferation 
      in breast cancer. Carcinogenesis 2002; 23: 81-86
22  Jayasurya A, Bay BH, Yap WM, Tan NG. Correlation of metallothionein expression with apoptosis in 
      nasopharyngeal carcinoma. Br J Cancer 2000; 82: 1198-1203
23  Cherian MG, Howell SB, Imura N, Klaassen CD, Koropatnick J, Lazo JS, Waalkes MP. Role of metallothionein 
      in carcinogenesis. Toxicol Appl Pharmacol 1994; 126: 1-5
24  Jasani B, Schmid KW. Significance of metallothionein overexpression in human tumours. Histopathology 
      1997; 31: 211-214
25  Huang GW, Yang LY. Metallothionein expression in hepatocellular carcinoma. World J Gastroenterol 2002; 8: 650-653
26  Rossen K, Haerslev T, Hou-Jensen K, Jacobsen GK. Metallothionein expression in basaloid proliferations 
      overlying dermatofibromas and in basal cell carcinomas. Br J Dermatol 1997; 136: 30-34
27  Zhang XH, Jin L, Sakamoto H, Takenaka I. Immunohistochemical localization of metallothionein in human prostate 
      cancer. J Urol 1996; 156: 1679-1681
28  Giuffre G, Barresi G, Sturniolo GC, Sarnelli R, D扞nca R, Tuccari G. Immunohistochemical expression of metallothionein 
      in normal human colorectal mucosa, in adenomas and in adenocarcinomas and their associated metastases.
      Histopathology 1996; 29: 347-354
29  Kuo T, Lo SK. Immunohistochemical metallothionein expression in thymoma: correlation with histological types and 
      cellular origin. Histopathology 1997; 30: 243-248
30  Shukla VK, Aryya NC, Pitale A, Pandey M, Dixit VK, Reddy CD, Gautam A. Metallothionein expression in carcinoma of 
      the gallbladder. Histopathology 1998; 33: 154-157
31  Zelger B, Hittmair A, Schir M, Ofner C, Ofner D, Fritsch PO, Bocker W, Jasani B, Schmid KW. Immunohistochemically
      demonstrated metallothionein expression in malignant melanoma. Histopathology 1993; 23: 257-264
32  Goulding H, Jasani B, Pereira H, Reid A, Galea M, Bell JA, Elston CW, Robertson JF, Blamey RW, Nicholson 
      RA. Metallothionein expression in human breast cancer. Br J Cancer 1995; 72: 968-972
33  Douglas-Jones AG, Schmid KW, Bier B, Horgan K, Lyons K, Dallimore ND, Moneypenny IJ, Jasani B. 
      Metallothionein expression in duct carcinoma in situ of the breast. Hum Pathol 1995; 26: 217-222
34  Uozaki H, Horiuchi H, Ishida T, Iijima T, Imamura T, Machinami R. Overexpression of resistance-related 
      proteins (metallothioneins, glutathione-S-transferase pi, heat shock protein 27, and lung resistance-related 
      protein) in osteosarcoma. Relationship with poor prognosis. Cancer 1997; 79: 2336-2344

   Edited by Zhu L and Wang XL  

Reviews	Add
more>>