Original Paper

Pdf file on Synergy OPEN

omments

Acta Biochim Biophys Sin 2008, 40: 375-380

doi:10.1111/j.1745-7270.2008.00420.x

Interaction between angiogenin and fibulin 1: evidence and implication

 

Hui Zhang1&, Xiangwei Gao1, Chunhua Weng1,2, and Zhengping Xu1,2*

 

1 Research Center for Environmental Genomics, and 2 Bioelectromagnetics Laboratory, Zhejiang University School of Medicine, Hangzhou 310058, China

 

Received: January 20, 2008�������

Accepted: March 17, 2008

This work was supported by the grants from the National Natural Science Foundation of China (No. 30171035), the Program for New Century Excellent Talents in University (No. NCET-05-0521), and the Zhejiang Provincial Program for the Cultivation of High-level Innovative Health Talents

&Present address: Department of Biochemistry, Zhejiang Medical College, Hangzhou 310053, China

*Corresponding author: Tel, 86-571-88208164; Fax, 86-571-88208163; E-mail, [email protected]

 

Angiogenin is an angiogenic factor involved in tumorigenesis. However, the mechanism of angiogenin�s action remains elusive. In the present study, we identified fibulin 1, an extracellular matrix and plasma glycoprotein, as an angiogenin-interacting molecule by yeast two-hybrid screening. This interaction was further confirmed by two different approaches. First, fibulin 1 was co-immunoprecipitated with angiogenin by anti-angiogenin monoclonal antibody in vitro, suggesting angiogenin binds with fibulin 1 directly. Then fluorescence resonance energy transfer analysis showed that fibulin 1 interacted with angiogenin in COS-7 cells, showing that the binding could occur in a cellular context. As fibulin 1 plays an important role in cell proliferation, migration, adhesion, and stabilizes new-forming blood vessel wall, the interaction between fibulin 1 and angiogenin might underline one possible mechanism of angiogenin in angiogenesis and/or tumorigenesis.

 

Keywords� ������angiogenin; fibulin 1; protein interaction; FRET; angiogenesis

 

Angiogenin (ANG) was originally isolated from the conditioned medium of HT-29 human colon adenocarcinoma cells based solely on its angiogenic activity [1], and was up-regulated in many types of cancers [2,3]. Animal studies showed that its monoclonal antibody (mAb), binding protein, enzymatic inhibitors, and antisense RNA suppressed tumor growth or metastasis [4-9]. Although the roles of ANG in angiogenesis and tumorigenesis have been well recognized, its mechanism of action needs to be elucidated.

ANG plays a key role in tumor angiogenesis. It activates vessel-originated endothelial and smooth muscle cells through binding with membrane proteins and undergoing nuclear translocation, and induces a wide range of cellular responses, including cell migration, invasion, proliferation, and formation of tubular structures [10,11]. Results showed that ANG bound to actin or a putative 170 kDa protein, and activated extracellular signal-regulated kinase 1/2 in human umbilical vein endothelial cells [12] or stress-activated protein kinase/c-Jun N-terminal kinase in human umbilical artery smooth muscle cells [13]. Meanwhile, exogenous ANG could translocate to the nucleus through an unknown mechanism and bind to DNA, hence enhancing rRNA transcription [14-16]. It was reported that ANG was a general requirement for endothelial cell proliferation and angiogenesis induced by various other angiogenic proteins, including acidic and basic fibroblast growth factors, epidermal growth factor, and vascular endothelial growth factor, suggesting ANG-stimulated rRNA transcription in endothelial cells might serve as a crossroad in the process of angiogenesis [15]. Recently, data showed that ANG was constitutively translocated into the nucleus of HeLa cells where it stimulated rRNA transcription, indicating that ANG also played a role in cancer cell proliferation [2].

However, there are still many knowledge gaps in clarifying the mechanism of ANG in angiogenesis and tumorigenesis. We believe that establishing a protein interaction network should help in filling these gaps. Therefore, a yeast two-hybrid technique was used to screen ANG�s potential interacting molecules from a human cDNA library. In the present study, we report the identification of fibulin 1 as an ANG binding partner.

 

Materials and Methods

 

Plasmid construction

The full-length cDNA of ANG was amplified by polymerase chain reaction (PCR) (forward primer, 5'-ATATAGTC�G�A�C�TCAGGATAACTCCAGGT-3'; reverse primer, 5'-ATA�A�GAATGCGGCCGCTA�CGGACG�AC�GGAA-3'), digested with SalI and NotI, and inserted into the corresponding sites of the vector pDBLeu (Invitrogen, Shanghai, China). This construct was named pDBLeu-ANG and subsequently used in the yeast two-hybrid screening as described below.

For in vitro co-immunoprecipitation analysis, fibulin 1, identified as an ANG binding molecule in the yeast two-hybrid screening, was amplified by PCR with primers 5'-CG�GAATTCATGGCAGTGGGCTCTTTCC-3' (forward) and 5'-AATGCGGCCGCAGGGTTAGGAAAGCAAC-3' (reverse). Then plasmid pCMVTNT-fibulin 1 was constructed by inserting the amplified fibulin 1 gene into vector pCMVTNT (Promega, Madison, USA), which was digested with EcoRI and NotI, and used in the in vitro transcription/translation system.

For molecular imaging, fibulin 1 was amplified by PCR with primers 5'-GAAGATCTCCAAAAAAGAGGGTGGG-3' (forward) and 5'-CGGAATTCA�GCTCGACGTC�TTAC�T��T-3' (reverse), and fused to cyan fluorescent protein (CFP) by cloning into the BglII and EcoRI sites of vector pECFP-C1 (Clontech, Palo Alto, USA) to generate pCFP-fibulin 1. ANG was amplified by PCR with primers 5'-C�G�GAA�TTCACCATGCAGGATAACTCCAGGTAC-3' (forward) and 5'-CGGGATCCGGACGGAAAATTG-3' (reverse), and digested with EcoRI and BamHI. The purified fragment was inserted into pEYFP-N1 (Clontech, Palo Alto, USA) at the corresponding sites to generate pYFP-ANG.

All constructs were confirmed by DNA sequencing.

 

Yeast two-hybrid screening

A yeast two-hybrid screening was carried out according to the manufacturer�s instructions (Invitrogen). Briefly, after passing the self-activation assessment and determining the basal expression level of the his3 reporter gene, the bait plasmid pDBLeu-ANG was transformed into the yeast strain MaV203 together with a cDNA library prepared from human heart in the prey vector pPC86 by a lithium acetate method. Candidate clones were selected by plating cotransformants on selective medium (Trp-, Leu-, His- plus 25 mM 3-amino-1,2,4-triazole) and confirmed by b-galactosidase activity analysis. Plasmids were isolated from candidate clones and retransformation assays were carried out to verify the interactions. The selected plasmids were sequenced by dideoxy sequencing. The insert sequences obtained were subjected to blast (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi).

 

In vitro co-immunoprecipitation analysis

Fibulin 1 was in vitro transcribed and translated in the TNT T7 Quick coupled transcription/translation system (Promega) using pCMVTNT-fibulin1 as a template and pCMVTNT as a control. In each reaction, 3 mg plasmid DNA was added to a 50 ml reaction system containing 40 ml TNT Quick Master Mix, 2 ml [35S]-methionine (3.7�103 Bq/mmol at 37Bq/ml; Amersham Pharmacia, Piscataway, USA). After incubation at 30 �C for 90 min, 10 ml in vitro translated products was mixed with 10 mg ANG (a gift from Dr. Guofu Hu at Harvard Medical School, Cambridge, USA) at 4 �C for 2 h. Then 52.5 ml co-immunoprecipitation buffer (20 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA, and 0.5% NP-40), 20 ml Protein A agarose beads (Santa Cruz Biotechnology, Santa Cruz, USA), and 5 mg ANG mAb 26-2F were added and incubated for 2 h at 4 �C with continuous rocking. After centrifugation at 5000 g for 15 s, the pellet was rinsed five times with 100 ml co-immunoprecipitation buffer each to eliminate non-specific binding. The final pellet was added with 15 ml sodium dodecyl sulfate (SDS) loading buffer, heated at 95 �C for 5 min, and separated by SDS-polyacrylamide gel electrophoresis. After electrophoresis, the gel was fixed and dried at 80 �C under constant vacuum, and subjected to autoradiography.

 

Cell culture and transfection

COS-7 cells were maintained in Dulbecco�s modified Eagle�s medium (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone, Logan, USA) at 37 �C in 5% humidified CO2. To carry out fluorescence resonance energy transfer (FRET) analysis, glass coverslips were placed in 6-well tissue culture plates and each well was seeded with 1-3�105 cells in 2 ml complete medium. After incubation for 24 h, the transfection was carried out using PolyFect transfection reagent (Qiagen, Hilden, Germany) according to the manufacturer�s protocol. In the cotransfection experiment, the DNA ratio of the two plasmids was adjusted to 1 to obtain approximately equal amounts of expressed proteins.

 

Live imaging of proteins and FRET

Cells were imaged in complete medium at room temperature. The experiments were carried out with a confocal imaging spectrophotometer system (TCS SP; Leica, Wetzlar, Germany) attached to a Leica DMIRBE inverted epifluorescence microscope and equipped with an argon laser (458, 488, 514, and 543 nm lines) and acousto-optic tunable filter to attenuate individual visible laser lines. In all cases a 40�NA oil immersion objective lens was applied. Quantification of FRET was determined by calculating the ratio of intensity of the fluorescence emission signals at 525 nm/475 nm while excited at the CFP excitation wavelength (458 nm).

In the photobleaching study, CFP-fibulin 1 was excited at 458 nm and detected at 470-500 nm. For ANG-yellow fluorescent protein (YFP), an excitation wavelength of 514 nm and an emission wavelength of 530-550 nm was applied. CFP-fibulin 1 and ANG-YFP images were collected at 10 s intervals for five times. After that, YFP was photobleached with the full power of the 514 nm line laser, and the emission intensities of CFP and YFP were monitored an additional five times.

 

Results

 

Fibulin 1 interacted with ANG in yeast cells

To identify putative ANG-interacting proteins, a human heart cDNA library was screened by the bait protein ANG following the standard yeast two-hybrid screening procedure. Among 46 truly positive clones [17], two coded for human fibulin 1 isoform C (GenBank accession No. NM_001996; Table 1), showing that fibulin 1 was an ANG binding protein in yeast cells. The longest ANG-binding fibulin 1 fragment was from amino acid 211 (corresponding to the nucleotide 707 bp in mRNA). This peptide was used in the following study.

 

Fibulin 1 interacted with ANG in vitro

Fibulin 1 was in vitro synthesized and labeled with [35S]-Met, and subsequently incubated with ANG. The mixture was immunoprecipitated with anti-ANG mAb 26-2F in the presence of Protein A agarose beads, and the co-immunoprecipitated products were subjected to SDS-polyacrylamide gel electrophoresis followed by autoradiography. Fig. 1 clearly shows that fibulin 1 could be co-immunoprecipitated by 26-2F (Fig. 1, lane 3), indicating that fibulin 1 interacted with ANG in vitro.

 

Fibulin 1 interacted with ANG in vivo detected by FRET analysis

To confirm the binding between fibulin 1 and ANG in mammalian cellular content, we investigated the interaction between the two proteins in COS-7 cells using FRET analysis, a molecular imaging approach. Spectra of CFP-tagged and YFP-tagged proteins were first determined in cells transfected with each single construct and excited with the appropriate laser line (458 nm for CFP and 514 nm for YFP), which showed the emission spectra for CFP-fibulin 1 and ANG-YFP were rather distinct with emission peaks at approximately 475 and 525 nm, respectively [Fig. 2(A,B)]. For the measurement of FRET, cells expressing both CFP-fibulin 1 and ANG-YFP were excited at 458 nm. If FRET occurred between CFP-fibulin 1 and ANG-YFP, a shift in the emission spectrum of CFP (475 nm) to YFP (525 nm) would be observed. As expected, a decrease in the 475 nm emission and an increase in the 525 nm emission occurred [Fig. 2(C)], indicating that FRET did take place between CFP-fibulin 1 and ANG-YFP in COS-7 cells.

YFP could be excited by 458 nm light at a level approximately 30% of the intensity induced by 514 nm laser. Such �spectra cross-talk� will cause a false FRET that is dependent on the expression level of the YFP construct relative to the CFP construct. Thus, if the expression of the YFP protein was to be greatly in excess of the CFP protein, discriminating between such �spectra cross-talk� and real FRET would not be possible. For this reason, the experimental conditions were optimized to ensure that the expression level of the YFP construct was approximate to that of the CFP construct, achieved by adjusting the plasmid ratio to 1:1 while carrying out double transfection. Furthermore, energy transfer might occur due to the physical proximity of two proteins crowded together due to overexpression. To eliminate these possibilities, CFP-fibulin 1 and YFP, or CFP and ANG-YFP were expressed in a cell simultaneously and the emission pattern of each pair was determined by spectrum scanning. Although there was an emission peak at 525 nm in both control groups [Fig. 2(D,E)], significant differences existed between the two control groups and cells expressing CFP-fibulin 1 plus ANG-YFP. The ratios of 525 nm to 475 nm emission intensity of the two control groups were 0.80�0.10 (n=8) and 0.81�0.12 (n=8), respectively; the ratio of the cells expressing both CFP-fibulin 1 and YFP-ANG was 2.90� 1.19 (n=8).

 

Photobleaching experiments confirm FRET measurements

To be more certain that the evidence for FRET described above was valid, we employed photobleaching of the YFP using a high-intensity exposure to the YFP excitation light (514 nm laser) to block energy transfer from CFP to YFP, and recorded the energy redistribution of CFP. If FRET occurs, photobleaching of ANG-YFP should lead to an increase in the 480 nm emission of CFP-fibulin1. Experimental observation showed that this was indeed the case (Fig. 3). Real-time imaging showed the emission intensity of CFP-fibulin 1 significantly increased following bleaching of ANG-YFP. The photobleaching results further suggested that the apparent FRET phenomenon was due to the interaction between fibulin 1 and ANG and not triggered by �spectra cross-talk� between CFP and YFP or overexpression-induced proximity of two proteins.

 

Discussion

 

Fibulin 1 belongs to a six-member family of extracellular matrix (ECM) protein named fibulin [18,19]. Although its actual function remains to be explored, fibulin 1 has been suggested to be involved in cell adhesion, migration, proliferation, and malignant transformation through binding to various ECM proteins, receptors and growth factors [18-20]. Of particular interest is that fibulin 1 might be involved in angiogenesis. It was reported that fibulin 1 bound with angiogenesis inhibitor endostatin [21]. Targeted inactivation of fibulin 1 gene in mice caused dilation and ruptures in the endothelial lining of small blood vessels [22], indicating that fibulin 1 was important in the stabilization of blood vessel walls. Therefore, it is possible that ANG induces the essential cellular processes of angiogenesis through interaction with fibulin 1.

As ANG is a key player in angiogenesis [15], the interaction between ANG and fibulin 1 might interfere with the normal physiological or pathological functions of fibulin 1, and affect new blood vessel formation. To obtain more information about fibulin 1 in angiogenesis, we carried out bioinformatics analysis using PathwayAssist 3.0 software (Invitrogen). Data mining suggested that the reported binding partners of fibulin 1 include fibronectin, fibrinogen, laminin, nidogen, tropoelastin, aggrecan, versican, b-amyloid precursor protein, connective-tissue growth factor, heparin-binding epidermal growth factor (HB-EGF), and the angiogenesis inhibitor endostatin. Fibulin 1 might negatively regulate cell motility and proliferation. Binding of fibulin 1 with fibronectin could generate a new anti-adhesive site, that repulses cellular interactions rather than promotes them, thus inhibiting cell adhesion to, and migration on, fibronectin [23]. Versican was described as negatively regulating cell movement [24]. Fibulin 1 might suppress cell migration through binding to aggrecan or versican. ANG is known to promote cell proliferation, motility, and adhesion [10], contrary to the functions of fibulin 1. Therefore, it is sound to speculate that one possible mechanism of action of ANG is to disrupt the suppression of fibulin 1 on these cell behaviors by interfering fibulin 1�s binding to fibronectin, laminin, nidogen, versican, and aggrecan (Fig. 4). However, ANG was discovered to interact with cytoskeletal protein α-actinin 2 [17], and fibulin 1 was reported� to control the cytoskeletal organization of endothelial cells [22]. Thus, the interaction between ANG and fibulin 1 might also modulate the cellular cytoskeletal system in favor of angiogenesis.

Taken together, we identified fibulin 1 as an interacting partner of ANG, and proposed that the interaction might partly account for the molecular mechanism underlining ANG-induced angiogenesis. It will be of high priority to clarify the pathophysiological significance of this interaction, which could be helpful in elucidating the molecular mechanism and provide a new interfering target for anti-angiogenesis and antitumor therapy.

 

References

 

 1�� Fett JW, Strydom DJ, Lobb RR, Alderman EM, Bethune JL, Riordan JF, Vallee BL. Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells. Biochemistry 1985, 24: 5480-5486

 2�� Tsuji T, Sun Y, Kishimoto K, Olson K, Liu S, Hirukawa S, Hu GF. Angiogenin is translocated to the nucleus of HeLa cells and is involved in ribosomal RNA transcription and cell proliferation. Cancer Res 2005, 4: 1352-1360

 3�� Kuroi K, Toi M. Circulating angiogenesis regulators in cancer patients. Int J Biol Markers 2001, 16: 5-26

 4�� Olson KA, Fett JW, French TC, Key ME, Vallee BL. Angiogenin antagonists prevent tumor growth in vivo. Proc Natl Acad Sci USA 1995, 92: 442-446

 5� Piccoli R, Olson KA, Vallee BL, Fett JW. Chimeric anti-angiogenin antibody cAb 26-2F inhibits the formation of human breast cancer xenografts in athymic mice. Proc Natl Acad Sci USA 1998, 95: 4579-4583

 6�� Olson KA, Byers HR, Key ME, Fett JW. Inhibition of prostate carcinoma establishment and metastatic growth in mice by an antiangiogenin monoclonal antibody. Int J Cancer 2002, 98: 923-929

 7�� Olson KA, Byers HR, Key ME, Fett JW. Prevention of human prostate tumor metastasis in athymic mice by antisense targeting of human angiogenin. Clin Cancer Res 2001, 7: 3598-3605

 8�� Kao RY, Jenkins JL, Olson KA, Key ME, Fett JW, Shapiro R. A small-molecule inhibitor of the ribonucleolytic activity of human angiogenin that possesses antitumor activity. Proc Natl Acad Sci USA 2002, 99: 10066-10071

 9�� Kumar K, Brady M, Shapiro R. Selective abolition of pancreatic RNase binding to its inhibitor protein. Proc Natl Acad Sci USA 2004, 101: 53-58

10� Riordan JF. Angiogenin. Methods Enzymol 2001, 341: 263-273

11� Xu Z, Monti DM, Hu G. Angiogenin activates human umbilical artery smooth muscle cells. Biochem Biophys Res Commun 2001, 285: 909-914

12� Liu S, Yu D, Xu ZP, Riordan JF, Hu GF. Angiogenin activates Erk1/2 in human umbilical vein endothelial cells. Biochem Biophys Res Commun 2001, 287: 305-310

13� Xu ZP, Tsuji T, Riordan JF, Hu GF. The nuclear function of angiogenin in endothelial cells is related to rRNA production. Biochem Biophys Res Commun 2002, 294: 287-292

14� Xu ZP, Tsuji T, Riordan JF, Hu GF. Identification and characterization of an angiogenin-binding DNA sequence that stimulates luciferase reporter gene expression. Biochemistry 2003, 42: 121-128

15� Kishimoto K, Liu S, Tsuji T, Olson K, Hu GF. Endogenous angiogenin in endothelial cells is a general requirement for cell proliferation and angiogenesis. Oncogene 2005, 3: 445-456

16� Li LY, Zhang YZ, Zheng YF, Hu GF, Xu ZP. An in vitro nuclear run-on assay for nuclear function of angiogenin. Acta Biochim Biophys Sin 2003, 35: 350-354

17� Hu H, Gao X, Sun Y, Zhou J, Yang M, Xu Z. α-Actinin-2, a cytoskeletal protein, binds to angiogenin. Biochem Biophys Res Commun 2005, 329: 661-667

18� Timpl R, Sasaki T, Kostka G, Chu ML. Fibulins: a versatile family of extracellular matrix proteins. Nat Rev Mol Cell Biol 2003, 4: 479-489

19� Argraves WS, Greene LM, Cooley MA, Gallagher WM. Fibulins: physiological and disease perspectives. EMBO Rep 2003, 4: 1127-1131

20� Chu ML, Tsuda T. Fibulins in development and heritable disease. Birth Defects Res C Embryo Today 2004, 72: 25-36

21� Sasaki T, Fukai N, Mann K, Gohring W, Olsen BR, Timpl R. Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin. EMBO J 1998, 15: 4249-4256

22� Kostka G, Giltay R, Bloch W, Addicks K, Timpl R, Fassler R, Chu ML. Perinatal lethality and endothelial cell abnormalities in several vessel compartments of fibulin-1-deficient mice. Mol Cell Biol 2001, 21: 7025-7034

23� Twal WO, Czirok A, Hegedus B, Knaak C, Chintalapudi MR, Okagawa H, Sugi Y et al. Fibulin-1 suppression of fibronectin-regulated cell adhesion and motility. J Cell Sci 2001, 114: 4587-4598

24� Landolt RM, Vaughan L, Winterhalter KH, Zimmermann DR. Versican is selectively expressed in embryonic tissues that act as barriers to neural crest cell migration and axon outgrowth. Development 1995, 121: 2303-2312