- Email: [email protected]
Identification of derlin-1 as a novel growth factor-responsive endothelial antigen by suppression subtractive hybridization
BBRC Biochemical and Biophysical Research Communications 348 (2006) 1272–1278 www.elsevier.com/locate/ybbrc
Identification of derlin-1 as a novel growth factor-responsive endothelial antigen by suppression subtractive hybridization q Yuliang Ran a, Yangfu Jiang b, Xing Zhong a, Zhuan Zhou a, Haiyan Liu a, Hai Hu a, Jin-Ning Lou c, Zhihua Yang a,* a
b
Department of Cellular and Molecular Biology, Cancer Institute (Hospital), Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, PR China Division of Molecular Oncology, State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, Sichuan 610041, PR China c Institute of Clinical Medical Sciences, China–Japan Friendship Hospital, Beijing 100029, PR China Received 20 July 2006 Available online 7 August 2006
Abstract Endothelial cells play an important regulatory role in embryonic development, reproductive functions, tumor growth and progression. In the present study, the suppression subtractive hybridization (SSH) method was employed to identify differentially expressed genes between non-stimulated endothelial cells and activated endothelial cells. Following mRNA isolation of non-stimulated and hepatocellular carcinoma homogenate-stimulated cells, cDNAs of both populations were prepared and subtracted by suppressive PCR. Sequencing of the enriched cDNAs identified a couple of genes differentially expressed, including derlin-1. Derlin-1 was significantly up-regulated by tumor homogenates, VEGF, and endothelial growth supplements in a dose-dependent manner. Knock-down of derlin-1 triggered endothelial cell apoptosis, inhibited endothelial cell proliferation, and blocked the formation of a network of tubular-like structures. Our data reveal that derlin-1 is a novel growth factor-responsive endothelial antigen that promotes endothelial cell survival and growth. 2006 Elsevier Inc. All rights reserved. Keywords: Angiogenesis; Derlin-1; Cancer
Angiogenesis is an important biological event for a variety of physiological and pathological processes, including embryonic development, reproductive functions, tumorigenesis, and other proliferative processes [1]. Tumor angiogenesis is essential for tumor growth and metastasis, which is promoted by cytokines or chemokines that have mitogenic or chemotactic effects on vascular endothelial cells [2,3]. Over the last decade, much progress has been made in the identification of the regulators of vasculogenesis or q Contract grant sponsor: National Natural Science Foundation of China (Key Program) (Contract Grant No. 30230150) and National High Technology Research and Development Program of China (863 Program) (Contract Grant No. 2004AA221140). * Corresponding author. E-mail address: [email protected] (Z. Yang).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.07.186
angiogenesis [4]. VEGF and Tie 2 ligands play a central role in endothelial proliferation and in the assembly of vessel wall [5]. The dependence of tumor growth and metastasis on angiogenesis has led to the concept of antiangiogenic therapy for malignant tumors [6]. Inhibition of tumor angiogenesis as an anticancer strategy has generated much excitement among cancer researchers and clinicians [7,8]. However, realization of the full potential of antiangiogenic approaches will require a better understanding of the molecular differences between normal and tumor vessels in a variety of normal and abnormal circumstances. Many evidence indicated that there was big difference in gene expression profile between normal vascular endothelial cells and tumor endothelial cells, as well as the significant difference in vascular morphology and function [9,10].
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
The formation of malignant cells that lack the ability to switch on angiogenesis rarely leads to clinically detectable cancers. Experimental evidence to support the concept of the ‘angiogenic switch’ already exists [3]. Accumulation of genetic alterations in tumor cells might turn on the expression or release of angiogenic factors, which may stimulate endothelial cell proliferation, migration, and tube formation. Suppression subtractive hybridization methodology has been successfully used to identify the differentially expressed genes between two reciprocally subtracted libraries [11]. We took advantage of this approach to identify the differentially expressed genes between quiescent and tumor homogenate-activated endothelial cells. Here, we report on the identification of a novel regulator of angiogenesis, derlin-1, by suppression subtractive hybridization. Derlin-1 reportedly participated in the dislocation of misfolded proteins from ER, mediates the retro-translocation of proteins from ER lumen into cytosol [12,13]. We demonstrated here that derlin-1 was regulated by VEGF, and promoted endothelial cell survival. Materials and methods Cell culture. The human vascular endothelial cells derived from hepatocellular carcinoma (HCVEC) were cultured as described previously [14]. The dishes for HCVEC culture were coated with 2% of gelatin. The coated dishes were incubated at 37 C for 15–20 min before cell plating. The HCVEC was maintained in DMEM containing 20% of fetal bovine serum, 100 lg/ml of endothelial cell growth supplements (ECGS), 2 mmol/L of L-glutamine, 100 lg/ml of sodium heparsulfate, 40 lU/ml of insulin, 100 U/ml of penicillin, and 100 lg/ml of streptomycin. Human umbilical vascular endothelial cell (HUVEC) was isolated from fresh neonatal umbillium by a recommended procedure [15]. Activation of HUVEC and suppression subtractive hybridization (SSH). To prepare the homogenate of human hepatocellular carcinoma, fresh tumor tissue was washed with sterilized PBS, cut into small pieces, and homogenized in DMEM. The homogenate was filtered and stored at 80 C. To activate HUVECs, HUVECs were treated with bFGF, VEGF, and the homogenate of HCC. The mRNA of activated HUVECs was isolated as described. The cDNA library for activated HUVECs was constructed. Briefly, the first and second strand of cDNA were synthesized by THERMOScript RT-PCR system, cDNA was ligated into Uni-ZAP XR vector, packaged by Gigapack III gold packaging extract. The titer of primary phage cDNA library was amplified. The differentially expressed gene fragments between HUVECs and activated HUVECs were identified by suppressive subtractive hybridization (SSH) as described [11,16,17]. Briefly, the SSH was performed between unstimulated HUVEC (driver) and HUVEC stimulated with HCC homogenate, bFGF, and VEGF (tester). For preparation of polyadenylated RNA, total RNA was extracted with the TRIzol reagent and purified with the Poly(A)Quick kit (Stratagene, Heidelberg, Germany). The further steps were performed according to the PCR-Select1cDNA Substraction kit (Clontech, Palo Alto, CA, USA). It is primarily based on a technique called suppression PCR and combines normalization and subtraction in a single procedure. The differentially expressed gene fragments were amplified by PCR and labeled with 32p-dATP and 32p-dCTP by Primer-a-Gene Labeling system (Promega). The labeled gene fragments were used as probes to screen the established HUVECs’ cDNA library. Preparation of polyclonal antibodies. Anti-derlin-1 antisera were generated by immunizing rabbits with peptides coupled to keyhole-limpet haemocyanin through an added Cys residue. The derlin-1 sequence used was (C)RHNWGQGFRLGDQ. The titer for anti-derlin-1 antisera was
1273
more than 1 · 109. Antibodies specific to the C-terminus of human derlin1 were affinity-purified with Sepharose 4B which was conjugated with the C-terminus peptide of derlin-1 as described [18]. The affinity-purified antiderlin-1 polyclonal antibody was used for immunohistochemical and Western blot analysis. Western blot analysis. Cells were washed twice with phosphate-buffered saline and lysed with cold RIPA lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40, 1% Triton X-100, 0.1% SDS, 1% sodium deoxycholate, 1 mM EDTA, 50 mM NaF, 10 mM sodium pyrophosphate, and 0.5 mM DTT) containing protease inhibitors (phenylmethylsulfonyl fluoride (PMSF) 1 mmol/L and leupeptin 0.1 g/L). Cell lysates were collected from culture plates using a rubber policeman, and protein collected by centrifugation. Protein concentrations were determined by BCA protein assay (Pierce, Rockford, IL). Aliquots of 20 lg proteins were boiled in 2· loading buffer (0.1 M Tris–Cl, pH 6.8, 4% SDS, 0.2% bromophenyl blue, and 20% glycerol) for 10 min, then loaded into 10% Tris–HCl– polyacrylamide gels (Bio-Rad, Hercules, CA), and transferred electrophoretically to PVDF membrane (Pierce). Membranes were incubated with primary antibodies against derlin-1 and appropriate HRP-secondary antibodies. Membranes were additionally probed with an antibody against actin (Sigma) to ensure equal loading of protein between samples. Detection was performed with enhanced chemiluminescence reagent (Pierce). Cell proliferation assay. For HCC homogenates assay, fresh HCC tissues were homogenized in DMEM and desterilized. HUVECs were seeded in 12-well plate at 15,000 cells/well. The cells were treated with or without HCC homogenates, VEGF, and bFGF. Cell number was counted for 6 days. For the in vitro cell proliferation assays, exponentially growing cells were seeded in quadruplicate at 1000 cells/well into 96-well plate. Cell numbers were determined every other day for 7 days by MTT assay. Briefly, 20 lL of MTT (5 g/L) was added into each well and cultured for another 4 h, the supernatant was discarded, then 100 lL DMSO was added. When the crystals were dissolved, the optical absorbent density (A) values of the slides were read on the minireader II at the wavelength of 490 nm. Each assay was repeated three times. Construction of shRNA expression vector. Two shRNAs targeting derlin-1 and one negative control shRNA were synthesized. shRNAs were cloned into pSilencer downstream of human U6 promotor. The expression cassets are as follows: Derlin-1shRNA1, 5 0 -GATCCCAGAGACATGA TTGTATCATTCAAGAGATGATACAATCATGTCTCTGTTTTTTG GAAA-3 0 ; derlin-1shRNA2, 5 0 -AGCTTTTCCAAAAAACACGATTTA AGGCCTGCTATCTCTTGAATAGCAGGCCTTAAATCGTGG-3 0 ; negative control shRNA, 5 0 -AGCTTTTCCAAAAAAAGGCTTAGGA ATCATACTATCTCTTGAATAGTATGATTCCTAAGCCTG-3 0 . Gene transfections. The expression vectors for derlin-1 shRNA and negative control shRNA were transfected into human HCVECs as described [19]. Briefly, Subconfluent proliferating cells in 12-well plate were incubated with 2 lg of expression vectors in 1 ml of serum-free medium containing LipoFECTAMINE for 5 h. Culture was washed to remove the excess medium to allow the expression of shRNAs. Transfected cells were selected with G418 for 2 weeks. Isolated clones were picked up. The expression of derlin-1 was checked by RT-PCR and flow cytometry. RT-PCR. Total RNA from HCVECs was isolated using Trizol reagent. Approximately 4 lg of total RNA was subjected to reverse transcription by M-MLV, followed by semiquantitative PCR analysis. The primer sequence for derlin-1 is as follows: sense, 5 0 -TCGGCAAAC TCGGCCTCATC-3 0 ; antisense, 5 0 -GAATGGCGGAGGCGGGAGA-3 0 . Gapdh was also amplified as internal control. Flow cytometry. Replicate cultures (n = 5) of 1 · 106 cells were plated in cell culture wells. The cells were harvested, washed with PBS, and fixed in 70% ethanol for 30 min at 4 C, then treated with 50 lg/ml RNase A (Sigma), stained with 50 lg/ml of propidium iodide for 20 min at 4 C without light, and analyzed by flow cytometry for DNA synthesis and cell cycle status. Tube formation assay. To prepare the gelatin-coated plates, 96-well plates were coated with DMEM supplemented with 3% of gelatin, incubated at 4 C overnight. 1 · 104 cells/well were plated in gelatin-coated
1274
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
plates. After cells attached the plates, 1 · 106 platelets/well were added into each well, followed by incubation at 37 C overnight. The tube formation of endothelial cells was checked under phase contrast microscope.
Results Stimulation of HUVEC proliferation by HCC homogenates To validate the optimal stimuli to mimic the in vivo activation of tumor endothelial cells, we investigated the effects of HCC homogenates, VEGF, and bFGF on HUVEC proliferation. Fresh HCC tissues were homogenized in DMEM and desterilized. HUVECs were seeded in 12-well plate at 15,000 cells/well. We found that HUVEC growth was significantly stimulated by HCC homogenates with a protein concentration between 250 ng/ml and 500 ng/ml. ELISA revealed that the concentration of VEGF and bFGF in 250 ng/ml HCC homogenates were 53.29 pg/ml and 11.43 pg/ml, respectively. To evaluate the effect of HCC homogenates on endothelial cells, We treated HUVEC with 250 ng/ml HCC homogenates, 53.29 pg/ml VEGF, 11.43 pg/ml bFGF, and a combination of 53.29 pg/ml VEGF and 11.43 pg/ml bFGF. The dosage of VEGF and bFGF was based on the amount of VEGF and bFGF in 250 ng/ml HCC homogenates, as detected by ELISA. Cell growth curve demonstrated that HCC homogenates stimulated HUVEC growth more profoundly than equivalent VEGF, whereas bFGF failed to stimulate HUVEC growth at this dosage (Fig. 1). These results indicated that tumor homogenates may serve as a desirable stimuli to activate endothelial cells.
Fig. 1. Stimulation of HUVEC growth by HCC homogenate. HUVECs were seeded in 12-well plate at 15,000 cells/well. The cells were treated with 250 ng/ml HCC homogenates, 53.29 pg/ml VEGF, 11.43 pg/ml bFGF, and a combination of 53.29 pg/ml VEGF and 11.43 pg/ml bFGF, respectively. The dosage of VEGF and bFGF was based on the amount of VEGF and bFGF in 250 ng/ml HCC homogenates, as detected by ELISA. Cell number was counted for 6 days.
Upregulation of derlin-1 expression in proliferating endothelial cells In an effort to identify novel angiogenesis-related genes in HCC, the mRNA of HUVEC stimulated with HCC homogenate (tester) and non-stimulated HUVEC (driver) was isolated. We performed SSH on HUVECs exposed to HCC homogenate and isolated approximately 35 differentially expressed clones between HUVECs and activated HUVECs, including some known angiogenesis-related genes, such as osteonectin, fibronectin 1, annexin A2, and MMP-1. Interestingly, we also isolated some novel angiogenesis-related genes, one of which was identified to be MGC3067/derlin-1, which was located in 8q24.13 and contained 7 exons. Derlin-1 expression was up-regulated in the HCC homogenate-activated human umbilical vascular endothelial cells (HUVECs), compared with unactivated HUVECs (Fig. 2A). Next, we tested whether derlin-1 expression could be regulated by endothelial cell growth supplements, which contain growth factors, hormones, trace elements, and other proteins. Consistent with HCC homogenate’s upregulation of derlin-1 expression in HUVEC, the endothelial cell growth supplements (ECGS) upregulated derlin-1 expression in HUVEC in a dose-dependent manner (Fig. 2B). To test whether derlin-1 expression is regulated by the key angiogenic factor VEGF, which are frequently overexpressed in tumor cells, we treated the
Fig. 2. Upregulation of derlin-1 expression in endothelial cells. (A) Induction of derlin-1 expression in endothelial cells by HCC homogenates. HUVEC was cultured in the absence of ECGS and bFGF, and treated with different doses of HCC homogenates. Cell lysates were subjected to Western blot. (B) Stimulation of derlin-1 expression by ECGS. HCVEC was cultured in the absence of bFGF, and treated with different doses of ECGS. Cell lysates were subjected to Western blot. (C) Upregulation of derlin-1 expression by VEGF. HCVEC was cultured in the absence of ECGS, and treated with different doses of VEGF. Cell lysates were subjected to Western blot.
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
1275
quiescent endothelial cells with different doses of recombinant VEGF. Western blot analysis demonstrated that derlin-1 expression was significantly upregulated by VEGF in a dose-dependent manner (Fig. 2C). Derlin-1 knock-down leads to HCVEC apoptosis To determine the functions of derlin-1 in proliferating endothelial cells, we constructed three eukaryotic expression vectors, two of which could express shRNA targeting derlin-1 (derlin-1shRNA1 and derlin-1shRNA2). Another one served as a negative control, which was called derlin1shRNAN. Derlin-1 was overexpressed in a human tumor endothelial cell line that was derived from hepatocellular carcinoma (HCVEC), as detected by RT-PCR and Western blot (Fig. 3). We then transfected HCVEC with derlin1shRNA1, derlin-1shRNA2, derlin-1shRNAN, and pSilencer (empty vector), respectively. The expression of derlin-1 in these transfectants was checked by RT-PCR and Western blot analysis. Both derlin-1shRNA1 and derlin-1shRNA2 could dramatically inhibit the expression of derlin-1, whereas derlin-1shRNAN and pSilencer could not (Fig. 3). There was dramatic change in the morphology of HCVECs transfected with derlin-1shRNA1 and derlin1shRNA2, as compared with HCVECs transfected with derlin-1shRNAN and pSilencer. Derlin-1 knock-down resulted in vacuolar degeneration in HCVECs. There were obviously different morphological characteristics in the cells overexpressing derlin-1 shRNA compared with the control cells when studied by a transmission electron microscope. Control cells had intact membrane, organelles, and normal nuclear morphology. Apoptotic cells in the derlin-1 knock-down group showed homogeneous chromatin condensation within the nucleus (data not shown). Derlin-1 shRNA also significantly inhibited HCVEC proliferation, as analyzed by MTT assay (Fig. 4).
Fig. 4. Effect of derlin-1 knock-down on cellular proliferation. Derlin-1 silencing inhibited HCVEC proliferation. Absorbance at 490 nm showed a significant decrease in growth of derlin-1shRNA-transfected cells as compared with parental and vector control cells (P < 0.01).
To detect the effect of derlin-1 on cell cycles and apoptosis of HCVECs, we picked up two clones transfected with pSilencer, two clones transfected with derlin-1shRNAN, three clones with derlin-1shRNA1, and three clones with derlin-1shRNA2 to analyze cell cycle and apoptosis by flow cytometry (Fig. 5). No apoptotic peak was detected in both empty vector- and shRNAN-transfected clones. However, the apoptotic peaks were detected in cells transfected with derlin-1shRNA1 and derlin-1shRNA2. The average apoptotic rate in cells transfected with derlin-1shRNA1 and derlin-1shRNA2 was 32.8% and 18.2%, respectively. No significant change in G1, S, and G2 phase was detected. These data indicated that derlin-1 might promote HCVEC growth and prevent cell apoptosis. Derlin-1 knock-down inhibits the tube formation of HCVECs We previously found that HCVECs are capable of forming tubular network. To determine the effect of derlin-1 on the assembly of vascular endothelial cells, we investigated the tube formation of HCVECs in gelatincoated dishes. Both HCVECs transfected with empty vector and derlin-1shRNAN could form typical tubular structure. However, HCVECs transfected with derlin-1shRNA1 and shRNA2 failed to form a network of tubular-like structures (Fig. 6). These results showed that derlin-1 silencing might repress the tube formation of HCVECs, indicating that derlin-1 may play an important role in angiogenesis. Discussion
Fig. 3. Detection of derlin-1 expression in HCVECs transfected with or without derlin-1 shRNA constructs. HCVEC was routinely cultured in the medium supplemented with ECGS and transfected with control vector; pSilencer-derlin-1shRNAN; pSilencer-derlin-1shRNA1; pSilencer-derlin1shRNA2. The expression of derlin-1 was checked by RT-PCR (A) and Western blot (B).
There has been considerable interest in identifying tumor endothelial markers, because they are tightly involved in tumor angiogenesis. Tumor angiogenesis is a complex process, which involves the interaction among cancer cells, endothelial cells, and the microenvironment [3,20]. Recent reports suggested that there was big difference between tumor vascular vessels and normal vessels [21]. Seventy-nine genes were found to be differentially expressed between colon cancer endothelial cells and
1276
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
Fig. 5. Effect of derlin-1 on cell cycles and apoptosis of HCVECs. HCVECs were transfected with control vector [(A) empty vector, (B) shRNAN] or derlin-1 shRNA constructs [(C) shRNA1, (D) shRNA2]. The effects of derlin-1 on cell cycles and apoptosis of HCVECs were analyzed by flow cytometry. The apoptotic peak was detected in the derlin-1 shRNA construct-transfected cells. Horizontal and vertical axes represent DNA content and cell number, respectively. The yellow peaks on the left of each graph correspond to G1, the broad yellow-green peaks in the middle correspond to S phase, and the green peaks on the right correspond to G2. The blue peaks in graphs (C) and (D) correspond to the apoptosis peak. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
Fig. 6. Derlin-1 knock-down inhibited the tube formation of HCVECs. HCVECs (A) transfected with derlin-1 shRNA constructs (C,D) or control vector (B) were plated in gelatin-coated dishes. After cells attached the plates, 1 · 106 platelets/well were added into each well, followed by incubation at 37 C overnight. The tube formation of endothelial cells was checked under phase contrast microscope.
normal colon endothelial cells [9]. Tumor vascular morphology and gene expression may also be affected by the host tumor cells [22]. Tumor angiogenesis-related genes
could be identified by treating HUVECs with cytokines or tumor cell culture medium. We undertook a different approach by treating HUVECs with hepatocellular carcinoma homogenate, VEGF, and b-FGF. Considering the important roles of VEGF and bFGF in angiogenesis [5], we assumed that HUVECs might be activated by tumor homogenate and VEGF. Indeed, HCC homogenate combined with VEGF, bFGF can not only upregulate VEGFR1, VEGFR2, and Ki67 expression and promote HUVEC growth, but also enhance cell motility and tube formation, suggesting that HCC homogenate and VEGF can effectively activate HUVECs. In consistency with others’ reports, our SSH study demonstrated that osteonectin, annexin A2, and MMP-1 were overexpressed in the activated HUVEC. In addition, we identify derlin-1 as a novel tumor endothelial antigen that may function downstream of VEGF or other angiogenic factors. First, derlin-1 is significantly upregulated by tumor homogenates, endothelial cell growth supplements, or VEGF. Second, derlin-1 knock-down results in HCVECs’ apoptosis and diminishes cellular proliferation, tube formation even in the presence of bFGF. Previously, it has been reported that derlin-1 is located on endoplasmic reticulum membrane [12]. Derlin-1 seems to be a multifunctional protein, which participates in the dislocation of
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
misfolded proteins from ER, mediates the retro-translocation of proteins from ER lumen into cytosol [12,13]. Derlin-1 reportedly carries four transmembrane domains, with both N-terminus and C-terminus within the cytosol. Derlin-1 depletion in Caenorhabditis elegans results in ER stress [23] and its expression is upregulated by inducers of ER stress in yeast [23] and C. elegans [9]. The overexpression of derlin-1 in tumor endothelial cells may indicate that tumor endothelium may be under greater ER stress than normal endothelium. It has been reported that VEGF upregulated the expression of proteins which participate in the process of degrading misfolded proteins in HUVEC [24]. Knock-down of derlin-1 may induce hepatocellular carcinoma endothelial cell apoptosis by triggering ER stress, due to the failure of processing and degrading its misfolded substrates. However, downregulation of derlin1 in colon cancer cell CL11187 and pancreatic cancer cell Suit II did not induce apoptosis (data not shown), suggesting derlin-1 is not essential for cancer cell viability, as in yeast [25–27], which may attribute to the compensation by other molecules. Alternatively, the residual amount of derlin-1 in derlin-1 knocked down cancer cells may be sufficient for the degradation of misfolded proteins under physiological condition. In healthy humans, VEGF promotes angiogenesis in embryonic development and is important in wound healing in adults [28]. It binds two VEGF receptors (VEGF receptor-1 and VEGF receptor-2), which are expressed on vascular endothelial cells [29]. VEGF is a key mediator of angiogenesis in cancer, in which it is upregulated by aberrantly expressed oncogenes, growth factors, and also stresses [30]. The production of VEGF and other growth factors by the tumor results in the ‘angiogenic switch’, where new vasculature is formed in and around the tumor, allowing it to grow exponentially [3]. We demonstrated here that VEGF upregulated derlin-1 expression in the endothelial cells. Taken together, these results suggest that derlin-1 may function as a downstream effector for VEGF to promote endothelial cell growth and tube formation. References [1] J. Folkman, Angiogenesis in cancer, vascular, rheumatoid and other disease, Nat. Med. 1 (1995) 27–31. [2] J. Folkman, What is the evidence that tumors are angiogenesis dependent? [editorial], J. Natl. Cancer Inst. 82 (1990) 4–6. [3] D. Hanahan, J. Folkman, Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis, Cell 86 (1996) 353–364. [4] B. StCroix, C. Rago, V. Velculescu, G. Traverso, K.E. Romans, E. Montgomery, et al., Gene expressed in human tumor endothelium, Science 289 (2000) 1197–1203. [5] G.D. Yancopoulos, S. Davis, N.W. Gale, J.S. Rudge, S.J. Wiegand, J. Holash, Vascular specific growth factors and blood vessel formation, Nature 407 (2000) 242–248. [6] J. Folkman, Tumor angiogenesis: therapeutic implications, N. Engl. J. Med. 285 (1971) 1182–1186. [7] J. Denekamp, Angiogenesis, neovascular proliferation and vascular pathophisiology as targets for cancer therapy, Br. J. Radiol. 66 (1993) 181–196.
1277
[8] O. Stoeltzing, W. Liu, N. Reinmuth, A. Parikh, S.A. Ahmad, Y.D. Jung, et al., Angiogenesis and antiangiotherapy of colon cancer liver metastasis, Ann. Surg. Oncol. 10 (2003) 722–733. [9] A. Ebwehard, S. Kahlert, V. Goede, B. Hemmerlein, K.H. Plate, H.G. Augustin, Heterogeneity of angiogenesis and blood vessel maturation in human tumors: implications for antiangiogenic tumor therapies, Cancer Res. 60 (2000) 1388–1393. [10] W.G. Roberts, J. Delaat, M. Nagane, S. Huang, W.K. Cavenee, G.E. Palade, Host microvasculature influence on tumor vascular morphology and endothelial gene expression, Am. J. Pathol. 153 (1998) 239–1248. [11] L. Diatchenko, YF. Lau, AP. Campbell, A. Chenchik, F. Moqadam, B. Huang, et al., Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries, Proc. Natl. Acad. Sci. USA 93 (1996) 6025–6030. [12] Y. Ye, Y. Shibata, C. Yun, D. Ron, T.A. Rapoport, A membrane protein complex mediates retro-translocation from the ER lumen into the cytosol, Nature 429 (2004) 841–847. [13] B.N. Lilley, H.L. Ploegh, A membrane protein required for dislocation of misfolded proteins from the ER, Nature 429 (2004) 834–840. [14] X. Zhong, Y.L. Ran, J.L. Lou, D. Hu, L. Yu, Y.S. Zhang, et al., Construction of human liver cancer vascular endothelium cDNA expression library and screening of the endothelium associated antigen genes, World J. Gastroenterol. 10 (2004) 1402–1408. [15] E.A. Jaffe, R.L. Nachman, C.G. Becker, et al., Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria, J. Clin. Invest. 52 (1973) 2745–2756. [16] J.L. Wang, Y.H. Liu, M.C. Lee, T.M. Nguyen, C. Lee, A. Kim, et al., Identification of tumor angiogenesis-related genes by subtractive hybridization, Microvas. Res. 59 (2000) 394–397. [17] S. Stie, G. Totzke, E. Grunewald, T. Neuhaus, S. Fronhorffs, A. Sachinidis, et al., Identification of syntenin and other TNF-inducible genes in human umbilical arterial endothelial cells by suppression subtractive hybridization, FEBS Lett. 467 (2000) 299–304. [18] K.E. Sawin, T.J. Mitchison, L.G. Wordeman, Evidence for kinesinrelated proteins in the mitotic apparatus using peptide antibodies, J. Cell Sci. 101 (1992) 303–313. [19] Y. Jiang, Y.E. Liu, I.D. Goldberg, Y.E. Shi, c Synuclein, a novel heat shock protein-associated chaperone, stimulates ligand-dependent estrogen receptor a signaling and mammary tumorigenesis, Cancer Res. 64 (2004) 4539–4546. [20] J.P. Girard, E.S. Baekkevold, T. Yamanaka, G. Haraldsen, P. Brandtzaeg, F. Amalri, Heterogeneity of endothelial cells: the specialized phenotype of human high endothelial venules characterized by suppression subtractive hybridization, Am. J. Pathol. 155 (1999) 2043–2055. [21] E. Ruoslahti, Specialization of tumor vasculature, Nat. Rev. Cancer 2 (2002) 83–90. [22] N.N. Khodarev, J. Yu, E. Labay, T. Darga, C.K. Brown, H.J. Mauceri, et al., Tumor-endothelium interactions in co-culture: coordinated changes of gene expression profiles and phenotypic properties of endothelial cells, J. Cell Sci. 116 (2003) 1013–1022. [23] K.J. Travers et al., Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation, Cell 101 (2000) 249–258. [24] Z. Pawlowska, P. Baranska, H. Jerczynska, W. Koziolkiewicz, C.S. Cierniewski, Heat shock proteins and other components of cellular machinery for protein synthesis are up-regulated in vascular endothelial cell growth factor-activated human endothelial cells, Proteomics 5 (2005) 1217–1227. [25] M. Knop, A. Finger, T. Braun, K. Hellmuth, D.H. Wolf, Der1, a novel protein specifically required for endoplasmic reticulum degradation in yeast, EMBO J. 15 (1996) 753–763. [26] R.G. Gardner, G.M. Swarbrick, N.W. Bays, S.R. Cronin, S. Wilhovsky, L. Sleelig, C. Kim, R.Y. Hampton, Endoplasmic
1278
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
reticulum degradation requires lumen to cytosol signaling. Transmembrane control of Hrd1p by Hrd3p, J. Cell Biol. 151 (2000) 69–82. [27] Y. Oda, K. Okada, H. Yoshida, R. Kaufman, K. Nagata, K. Mori, Derlin-2 and Derlin-3 are regulated by the mammalian unfolded response and are required for ER-associated degradation, J. Cell Biol. 172 (2006) 383–393.
[28] N. Ferrara, VEGF and the quest for tumor angiogenesis factors, Nat. Rev. Cancer 2 (2002) 795–803. [29] N. Ferrara, H.P. Gerber, J. LeCouter, The biology of VEGF and its receptors, Nat. Med. 9 (2003) 669–676. [30] Q.T. Le, N.N. Denko, A.J. Giaccia, Hypoxic gene expression and metastasis, Cancer Metastasis Rev. 23 (2004) 293–310.
Identification of derlin-1 as a novel growth factor-responsive endothelial antigen by suppression subtractive hybridization q Yuliang Ran a, Yangfu Jiang b, Xing Zhong a, Zhuan Zhou a, Haiyan Liu a, Hai Hu a, Jin-Ning Lou c, Zhihua Yang a,* a
b
Department of Cellular and Molecular Biology, Cancer Institute (Hospital), Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, PR China Division of Molecular Oncology, State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, Sichuan 610041, PR China c Institute of Clinical Medical Sciences, China–Japan Friendship Hospital, Beijing 100029, PR China Received 20 July 2006 Available online 7 August 2006
Abstract Endothelial cells play an important regulatory role in embryonic development, reproductive functions, tumor growth and progression. In the present study, the suppression subtractive hybridization (SSH) method was employed to identify differentially expressed genes between non-stimulated endothelial cells and activated endothelial cells. Following mRNA isolation of non-stimulated and hepatocellular carcinoma homogenate-stimulated cells, cDNAs of both populations were prepared and subtracted by suppressive PCR. Sequencing of the enriched cDNAs identified a couple of genes differentially expressed, including derlin-1. Derlin-1 was significantly up-regulated by tumor homogenates, VEGF, and endothelial growth supplements in a dose-dependent manner. Knock-down of derlin-1 triggered endothelial cell apoptosis, inhibited endothelial cell proliferation, and blocked the formation of a network of tubular-like structures. Our data reveal that derlin-1 is a novel growth factor-responsive endothelial antigen that promotes endothelial cell survival and growth. 2006 Elsevier Inc. All rights reserved. Keywords: Angiogenesis; Derlin-1; Cancer
Angiogenesis is an important biological event for a variety of physiological and pathological processes, including embryonic development, reproductive functions, tumorigenesis, and other proliferative processes [1]. Tumor angiogenesis is essential for tumor growth and metastasis, which is promoted by cytokines or chemokines that have mitogenic or chemotactic effects on vascular endothelial cells [2,3]. Over the last decade, much progress has been made in the identification of the regulators of vasculogenesis or q Contract grant sponsor: National Natural Science Foundation of China (Key Program) (Contract Grant No. 30230150) and National High Technology Research and Development Program of China (863 Program) (Contract Grant No. 2004AA221140). * Corresponding author. E-mail address: [email protected] (Z. Yang).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.07.186
angiogenesis [4]. VEGF and Tie 2 ligands play a central role in endothelial proliferation and in the assembly of vessel wall [5]. The dependence of tumor growth and metastasis on angiogenesis has led to the concept of antiangiogenic therapy for malignant tumors [6]. Inhibition of tumor angiogenesis as an anticancer strategy has generated much excitement among cancer researchers and clinicians [7,8]. However, realization of the full potential of antiangiogenic approaches will require a better understanding of the molecular differences between normal and tumor vessels in a variety of normal and abnormal circumstances. Many evidence indicated that there was big difference in gene expression profile between normal vascular endothelial cells and tumor endothelial cells, as well as the significant difference in vascular morphology and function [9,10].
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
The formation of malignant cells that lack the ability to switch on angiogenesis rarely leads to clinically detectable cancers. Experimental evidence to support the concept of the ‘angiogenic switch’ already exists [3]. Accumulation of genetic alterations in tumor cells might turn on the expression or release of angiogenic factors, which may stimulate endothelial cell proliferation, migration, and tube formation. Suppression subtractive hybridization methodology has been successfully used to identify the differentially expressed genes between two reciprocally subtracted libraries [11]. We took advantage of this approach to identify the differentially expressed genes between quiescent and tumor homogenate-activated endothelial cells. Here, we report on the identification of a novel regulator of angiogenesis, derlin-1, by suppression subtractive hybridization. Derlin-1 reportedly participated in the dislocation of misfolded proteins from ER, mediates the retro-translocation of proteins from ER lumen into cytosol [12,13]. We demonstrated here that derlin-1 was regulated by VEGF, and promoted endothelial cell survival. Materials and methods Cell culture. The human vascular endothelial cells derived from hepatocellular carcinoma (HCVEC) were cultured as described previously [14]. The dishes for HCVEC culture were coated with 2% of gelatin. The coated dishes were incubated at 37 C for 15–20 min before cell plating. The HCVEC was maintained in DMEM containing 20% of fetal bovine serum, 100 lg/ml of endothelial cell growth supplements (ECGS), 2 mmol/L of L-glutamine, 100 lg/ml of sodium heparsulfate, 40 lU/ml of insulin, 100 U/ml of penicillin, and 100 lg/ml of streptomycin. Human umbilical vascular endothelial cell (HUVEC) was isolated from fresh neonatal umbillium by a recommended procedure [15]. Activation of HUVEC and suppression subtractive hybridization (SSH). To prepare the homogenate of human hepatocellular carcinoma, fresh tumor tissue was washed with sterilized PBS, cut into small pieces, and homogenized in DMEM. The homogenate was filtered and stored at 80 C. To activate HUVECs, HUVECs were treated with bFGF, VEGF, and the homogenate of HCC. The mRNA of activated HUVECs was isolated as described. The cDNA library for activated HUVECs was constructed. Briefly, the first and second strand of cDNA were synthesized by THERMOScript RT-PCR system, cDNA was ligated into Uni-ZAP XR vector, packaged by Gigapack III gold packaging extract. The titer of primary phage cDNA library was amplified. The differentially expressed gene fragments between HUVECs and activated HUVECs were identified by suppressive subtractive hybridization (SSH) as described [11,16,17]. Briefly, the SSH was performed between unstimulated HUVEC (driver) and HUVEC stimulated with HCC homogenate, bFGF, and VEGF (tester). For preparation of polyadenylated RNA, total RNA was extracted with the TRIzol reagent and purified with the Poly(A)Quick kit (Stratagene, Heidelberg, Germany). The further steps were performed according to the PCR-Select1cDNA Substraction kit (Clontech, Palo Alto, CA, USA). It is primarily based on a technique called suppression PCR and combines normalization and subtraction in a single procedure. The differentially expressed gene fragments were amplified by PCR and labeled with 32p-dATP and 32p-dCTP by Primer-a-Gene Labeling system (Promega). The labeled gene fragments were used as probes to screen the established HUVECs’ cDNA library. Preparation of polyclonal antibodies. Anti-derlin-1 antisera were generated by immunizing rabbits with peptides coupled to keyhole-limpet haemocyanin through an added Cys residue. The derlin-1 sequence used was (C)RHNWGQGFRLGDQ. The titer for anti-derlin-1 antisera was
1273
more than 1 · 109. Antibodies specific to the C-terminus of human derlin1 were affinity-purified with Sepharose 4B which was conjugated with the C-terminus peptide of derlin-1 as described [18]. The affinity-purified antiderlin-1 polyclonal antibody was used for immunohistochemical and Western blot analysis. Western blot analysis. Cells were washed twice with phosphate-buffered saline and lysed with cold RIPA lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40, 1% Triton X-100, 0.1% SDS, 1% sodium deoxycholate, 1 mM EDTA, 50 mM NaF, 10 mM sodium pyrophosphate, and 0.5 mM DTT) containing protease inhibitors (phenylmethylsulfonyl fluoride (PMSF) 1 mmol/L and leupeptin 0.1 g/L). Cell lysates were collected from culture plates using a rubber policeman, and protein collected by centrifugation. Protein concentrations were determined by BCA protein assay (Pierce, Rockford, IL). Aliquots of 20 lg proteins were boiled in 2· loading buffer (0.1 M Tris–Cl, pH 6.8, 4% SDS, 0.2% bromophenyl blue, and 20% glycerol) for 10 min, then loaded into 10% Tris–HCl– polyacrylamide gels (Bio-Rad, Hercules, CA), and transferred electrophoretically to PVDF membrane (Pierce). Membranes were incubated with primary antibodies against derlin-1 and appropriate HRP-secondary antibodies. Membranes were additionally probed with an antibody against actin (Sigma) to ensure equal loading of protein between samples. Detection was performed with enhanced chemiluminescence reagent (Pierce). Cell proliferation assay. For HCC homogenates assay, fresh HCC tissues were homogenized in DMEM and desterilized. HUVECs were seeded in 12-well plate at 15,000 cells/well. The cells were treated with or without HCC homogenates, VEGF, and bFGF. Cell number was counted for 6 days. For the in vitro cell proliferation assays, exponentially growing cells were seeded in quadruplicate at 1000 cells/well into 96-well plate. Cell numbers were determined every other day for 7 days by MTT assay. Briefly, 20 lL of MTT (5 g/L) was added into each well and cultured for another 4 h, the supernatant was discarded, then 100 lL DMSO was added. When the crystals were dissolved, the optical absorbent density (A) values of the slides were read on the minireader II at the wavelength of 490 nm. Each assay was repeated three times. Construction of shRNA expression vector. Two shRNAs targeting derlin-1 and one negative control shRNA were synthesized. shRNAs were cloned into pSilencer downstream of human U6 promotor. The expression cassets are as follows: Derlin-1shRNA1, 5 0 -GATCCCAGAGACATGA TTGTATCATTCAAGAGATGATACAATCATGTCTCTGTTTTTTG GAAA-3 0 ; derlin-1shRNA2, 5 0 -AGCTTTTCCAAAAAACACGATTTA AGGCCTGCTATCTCTTGAATAGCAGGCCTTAAATCGTGG-3 0 ; negative control shRNA, 5 0 -AGCTTTTCCAAAAAAAGGCTTAGGA ATCATACTATCTCTTGAATAGTATGATTCCTAAGCCTG-3 0 . Gene transfections. The expression vectors for derlin-1 shRNA and negative control shRNA were transfected into human HCVECs as described [19]. Briefly, Subconfluent proliferating cells in 12-well plate were incubated with 2 lg of expression vectors in 1 ml of serum-free medium containing LipoFECTAMINE for 5 h. Culture was washed to remove the excess medium to allow the expression of shRNAs. Transfected cells were selected with G418 for 2 weeks. Isolated clones were picked up. The expression of derlin-1 was checked by RT-PCR and flow cytometry. RT-PCR. Total RNA from HCVECs was isolated using Trizol reagent. Approximately 4 lg of total RNA was subjected to reverse transcription by M-MLV, followed by semiquantitative PCR analysis. The primer sequence for derlin-1 is as follows: sense, 5 0 -TCGGCAAAC TCGGCCTCATC-3 0 ; antisense, 5 0 -GAATGGCGGAGGCGGGAGA-3 0 . Gapdh was also amplified as internal control. Flow cytometry. Replicate cultures (n = 5) of 1 · 106 cells were plated in cell culture wells. The cells were harvested, washed with PBS, and fixed in 70% ethanol for 30 min at 4 C, then treated with 50 lg/ml RNase A (Sigma), stained with 50 lg/ml of propidium iodide for 20 min at 4 C without light, and analyzed by flow cytometry for DNA synthesis and cell cycle status. Tube formation assay. To prepare the gelatin-coated plates, 96-well plates were coated with DMEM supplemented with 3% of gelatin, incubated at 4 C overnight. 1 · 104 cells/well were plated in gelatin-coated
1274
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
plates. After cells attached the plates, 1 · 106 platelets/well were added into each well, followed by incubation at 37 C overnight. The tube formation of endothelial cells was checked under phase contrast microscope.
Results Stimulation of HUVEC proliferation by HCC homogenates To validate the optimal stimuli to mimic the in vivo activation of tumor endothelial cells, we investigated the effects of HCC homogenates, VEGF, and bFGF on HUVEC proliferation. Fresh HCC tissues were homogenized in DMEM and desterilized. HUVECs were seeded in 12-well plate at 15,000 cells/well. We found that HUVEC growth was significantly stimulated by HCC homogenates with a protein concentration between 250 ng/ml and 500 ng/ml. ELISA revealed that the concentration of VEGF and bFGF in 250 ng/ml HCC homogenates were 53.29 pg/ml and 11.43 pg/ml, respectively. To evaluate the effect of HCC homogenates on endothelial cells, We treated HUVEC with 250 ng/ml HCC homogenates, 53.29 pg/ml VEGF, 11.43 pg/ml bFGF, and a combination of 53.29 pg/ml VEGF and 11.43 pg/ml bFGF. The dosage of VEGF and bFGF was based on the amount of VEGF and bFGF in 250 ng/ml HCC homogenates, as detected by ELISA. Cell growth curve demonstrated that HCC homogenates stimulated HUVEC growth more profoundly than equivalent VEGF, whereas bFGF failed to stimulate HUVEC growth at this dosage (Fig. 1). These results indicated that tumor homogenates may serve as a desirable stimuli to activate endothelial cells.
Fig. 1. Stimulation of HUVEC growth by HCC homogenate. HUVECs were seeded in 12-well plate at 15,000 cells/well. The cells were treated with 250 ng/ml HCC homogenates, 53.29 pg/ml VEGF, 11.43 pg/ml bFGF, and a combination of 53.29 pg/ml VEGF and 11.43 pg/ml bFGF, respectively. The dosage of VEGF and bFGF was based on the amount of VEGF and bFGF in 250 ng/ml HCC homogenates, as detected by ELISA. Cell number was counted for 6 days.
Upregulation of derlin-1 expression in proliferating endothelial cells In an effort to identify novel angiogenesis-related genes in HCC, the mRNA of HUVEC stimulated with HCC homogenate (tester) and non-stimulated HUVEC (driver) was isolated. We performed SSH on HUVECs exposed to HCC homogenate and isolated approximately 35 differentially expressed clones between HUVECs and activated HUVECs, including some known angiogenesis-related genes, such as osteonectin, fibronectin 1, annexin A2, and MMP-1. Interestingly, we also isolated some novel angiogenesis-related genes, one of which was identified to be MGC3067/derlin-1, which was located in 8q24.13 and contained 7 exons. Derlin-1 expression was up-regulated in the HCC homogenate-activated human umbilical vascular endothelial cells (HUVECs), compared with unactivated HUVECs (Fig. 2A). Next, we tested whether derlin-1 expression could be regulated by endothelial cell growth supplements, which contain growth factors, hormones, trace elements, and other proteins. Consistent with HCC homogenate’s upregulation of derlin-1 expression in HUVEC, the endothelial cell growth supplements (ECGS) upregulated derlin-1 expression in HUVEC in a dose-dependent manner (Fig. 2B). To test whether derlin-1 expression is regulated by the key angiogenic factor VEGF, which are frequently overexpressed in tumor cells, we treated the
Fig. 2. Upregulation of derlin-1 expression in endothelial cells. (A) Induction of derlin-1 expression in endothelial cells by HCC homogenates. HUVEC was cultured in the absence of ECGS and bFGF, and treated with different doses of HCC homogenates. Cell lysates were subjected to Western blot. (B) Stimulation of derlin-1 expression by ECGS. HCVEC was cultured in the absence of bFGF, and treated with different doses of ECGS. Cell lysates were subjected to Western blot. (C) Upregulation of derlin-1 expression by VEGF. HCVEC was cultured in the absence of ECGS, and treated with different doses of VEGF. Cell lysates were subjected to Western blot.
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
1275
quiescent endothelial cells with different doses of recombinant VEGF. Western blot analysis demonstrated that derlin-1 expression was significantly upregulated by VEGF in a dose-dependent manner (Fig. 2C). Derlin-1 knock-down leads to HCVEC apoptosis To determine the functions of derlin-1 in proliferating endothelial cells, we constructed three eukaryotic expression vectors, two of which could express shRNA targeting derlin-1 (derlin-1shRNA1 and derlin-1shRNA2). Another one served as a negative control, which was called derlin1shRNAN. Derlin-1 was overexpressed in a human tumor endothelial cell line that was derived from hepatocellular carcinoma (HCVEC), as detected by RT-PCR and Western blot (Fig. 3). We then transfected HCVEC with derlin1shRNA1, derlin-1shRNA2, derlin-1shRNAN, and pSilencer (empty vector), respectively. The expression of derlin-1 in these transfectants was checked by RT-PCR and Western blot analysis. Both derlin-1shRNA1 and derlin-1shRNA2 could dramatically inhibit the expression of derlin-1, whereas derlin-1shRNAN and pSilencer could not (Fig. 3). There was dramatic change in the morphology of HCVECs transfected with derlin-1shRNA1 and derlin1shRNA2, as compared with HCVECs transfected with derlin-1shRNAN and pSilencer. Derlin-1 knock-down resulted in vacuolar degeneration in HCVECs. There were obviously different morphological characteristics in the cells overexpressing derlin-1 shRNA compared with the control cells when studied by a transmission electron microscope. Control cells had intact membrane, organelles, and normal nuclear morphology. Apoptotic cells in the derlin-1 knock-down group showed homogeneous chromatin condensation within the nucleus (data not shown). Derlin-1 shRNA also significantly inhibited HCVEC proliferation, as analyzed by MTT assay (Fig. 4).
Fig. 4. Effect of derlin-1 knock-down on cellular proliferation. Derlin-1 silencing inhibited HCVEC proliferation. Absorbance at 490 nm showed a significant decrease in growth of derlin-1shRNA-transfected cells as compared with parental and vector control cells (P < 0.01).
To detect the effect of derlin-1 on cell cycles and apoptosis of HCVECs, we picked up two clones transfected with pSilencer, two clones transfected with derlin-1shRNAN, three clones with derlin-1shRNA1, and three clones with derlin-1shRNA2 to analyze cell cycle and apoptosis by flow cytometry (Fig. 5). No apoptotic peak was detected in both empty vector- and shRNAN-transfected clones. However, the apoptotic peaks were detected in cells transfected with derlin-1shRNA1 and derlin-1shRNA2. The average apoptotic rate in cells transfected with derlin-1shRNA1 and derlin-1shRNA2 was 32.8% and 18.2%, respectively. No significant change in G1, S, and G2 phase was detected. These data indicated that derlin-1 might promote HCVEC growth and prevent cell apoptosis. Derlin-1 knock-down inhibits the tube formation of HCVECs We previously found that HCVECs are capable of forming tubular network. To determine the effect of derlin-1 on the assembly of vascular endothelial cells, we investigated the tube formation of HCVECs in gelatincoated dishes. Both HCVECs transfected with empty vector and derlin-1shRNAN could form typical tubular structure. However, HCVECs transfected with derlin-1shRNA1 and shRNA2 failed to form a network of tubular-like structures (Fig. 6). These results showed that derlin-1 silencing might repress the tube formation of HCVECs, indicating that derlin-1 may play an important role in angiogenesis. Discussion
Fig. 3. Detection of derlin-1 expression in HCVECs transfected with or without derlin-1 shRNA constructs. HCVEC was routinely cultured in the medium supplemented with ECGS and transfected with control vector; pSilencer-derlin-1shRNAN; pSilencer-derlin-1shRNA1; pSilencer-derlin1shRNA2. The expression of derlin-1 was checked by RT-PCR (A) and Western blot (B).
There has been considerable interest in identifying tumor endothelial markers, because they are tightly involved in tumor angiogenesis. Tumor angiogenesis is a complex process, which involves the interaction among cancer cells, endothelial cells, and the microenvironment [3,20]. Recent reports suggested that there was big difference between tumor vascular vessels and normal vessels [21]. Seventy-nine genes were found to be differentially expressed between colon cancer endothelial cells and
1276
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
Fig. 5. Effect of derlin-1 on cell cycles and apoptosis of HCVECs. HCVECs were transfected with control vector [(A) empty vector, (B) shRNAN] or derlin-1 shRNA constructs [(C) shRNA1, (D) shRNA2]. The effects of derlin-1 on cell cycles and apoptosis of HCVECs were analyzed by flow cytometry. The apoptotic peak was detected in the derlin-1 shRNA construct-transfected cells. Horizontal and vertical axes represent DNA content and cell number, respectively. The yellow peaks on the left of each graph correspond to G1, the broad yellow-green peaks in the middle correspond to S phase, and the green peaks on the right correspond to G2. The blue peaks in graphs (C) and (D) correspond to the apoptosis peak. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
Fig. 6. Derlin-1 knock-down inhibited the tube formation of HCVECs. HCVECs (A) transfected with derlin-1 shRNA constructs (C,D) or control vector (B) were plated in gelatin-coated dishes. After cells attached the plates, 1 · 106 platelets/well were added into each well, followed by incubation at 37 C overnight. The tube formation of endothelial cells was checked under phase contrast microscope.
normal colon endothelial cells [9]. Tumor vascular morphology and gene expression may also be affected by the host tumor cells [22]. Tumor angiogenesis-related genes
could be identified by treating HUVECs with cytokines or tumor cell culture medium. We undertook a different approach by treating HUVECs with hepatocellular carcinoma homogenate, VEGF, and b-FGF. Considering the important roles of VEGF and bFGF in angiogenesis [5], we assumed that HUVECs might be activated by tumor homogenate and VEGF. Indeed, HCC homogenate combined with VEGF, bFGF can not only upregulate VEGFR1, VEGFR2, and Ki67 expression and promote HUVEC growth, but also enhance cell motility and tube formation, suggesting that HCC homogenate and VEGF can effectively activate HUVECs. In consistency with others’ reports, our SSH study demonstrated that osteonectin, annexin A2, and MMP-1 were overexpressed in the activated HUVEC. In addition, we identify derlin-1 as a novel tumor endothelial antigen that may function downstream of VEGF or other angiogenic factors. First, derlin-1 is significantly upregulated by tumor homogenates, endothelial cell growth supplements, or VEGF. Second, derlin-1 knock-down results in HCVECs’ apoptosis and diminishes cellular proliferation, tube formation even in the presence of bFGF. Previously, it has been reported that derlin-1 is located on endoplasmic reticulum membrane [12]. Derlin-1 seems to be a multifunctional protein, which participates in the dislocation of
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
misfolded proteins from ER, mediates the retro-translocation of proteins from ER lumen into cytosol [12,13]. Derlin-1 reportedly carries four transmembrane domains, with both N-terminus and C-terminus within the cytosol. Derlin-1 depletion in Caenorhabditis elegans results in ER stress [23] and its expression is upregulated by inducers of ER stress in yeast [23] and C. elegans [9]. The overexpression of derlin-1 in tumor endothelial cells may indicate that tumor endothelium may be under greater ER stress than normal endothelium. It has been reported that VEGF upregulated the expression of proteins which participate in the process of degrading misfolded proteins in HUVEC [24]. Knock-down of derlin-1 may induce hepatocellular carcinoma endothelial cell apoptosis by triggering ER stress, due to the failure of processing and degrading its misfolded substrates. However, downregulation of derlin1 in colon cancer cell CL11187 and pancreatic cancer cell Suit II did not induce apoptosis (data not shown), suggesting derlin-1 is not essential for cancer cell viability, as in yeast [25–27], which may attribute to the compensation by other molecules. Alternatively, the residual amount of derlin-1 in derlin-1 knocked down cancer cells may be sufficient for the degradation of misfolded proteins under physiological condition. In healthy humans, VEGF promotes angiogenesis in embryonic development and is important in wound healing in adults [28]. It binds two VEGF receptors (VEGF receptor-1 and VEGF receptor-2), which are expressed on vascular endothelial cells [29]. VEGF is a key mediator of angiogenesis in cancer, in which it is upregulated by aberrantly expressed oncogenes, growth factors, and also stresses [30]. The production of VEGF and other growth factors by the tumor results in the ‘angiogenic switch’, where new vasculature is formed in and around the tumor, allowing it to grow exponentially [3]. We demonstrated here that VEGF upregulated derlin-1 expression in the endothelial cells. Taken together, these results suggest that derlin-1 may function as a downstream effector for VEGF to promote endothelial cell growth and tube formation. References [1] J. Folkman, Angiogenesis in cancer, vascular, rheumatoid and other disease, Nat. Med. 1 (1995) 27–31. [2] J. Folkman, What is the evidence that tumors are angiogenesis dependent? [editorial], J. Natl. Cancer Inst. 82 (1990) 4–6. [3] D. Hanahan, J. Folkman, Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis, Cell 86 (1996) 353–364. [4] B. StCroix, C. Rago, V. Velculescu, G. Traverso, K.E. Romans, E. Montgomery, et al., Gene expressed in human tumor endothelium, Science 289 (2000) 1197–1203. [5] G.D. Yancopoulos, S. Davis, N.W. Gale, J.S. Rudge, S.J. Wiegand, J. Holash, Vascular specific growth factors and blood vessel formation, Nature 407 (2000) 242–248. [6] J. Folkman, Tumor angiogenesis: therapeutic implications, N. Engl. J. Med. 285 (1971) 1182–1186. [7] J. Denekamp, Angiogenesis, neovascular proliferation and vascular pathophisiology as targets for cancer therapy, Br. J. Radiol. 66 (1993) 181–196.
1277
[8] O. Stoeltzing, W. Liu, N. Reinmuth, A. Parikh, S.A. Ahmad, Y.D. Jung, et al., Angiogenesis and antiangiotherapy of colon cancer liver metastasis, Ann. Surg. Oncol. 10 (2003) 722–733. [9] A. Ebwehard, S. Kahlert, V. Goede, B. Hemmerlein, K.H. Plate, H.G. Augustin, Heterogeneity of angiogenesis and blood vessel maturation in human tumors: implications for antiangiogenic tumor therapies, Cancer Res. 60 (2000) 1388–1393. [10] W.G. Roberts, J. Delaat, M. Nagane, S. Huang, W.K. Cavenee, G.E. Palade, Host microvasculature influence on tumor vascular morphology and endothelial gene expression, Am. J. Pathol. 153 (1998) 239–1248. [11] L. Diatchenko, YF. Lau, AP. Campbell, A. Chenchik, F. Moqadam, B. Huang, et al., Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries, Proc. Natl. Acad. Sci. USA 93 (1996) 6025–6030. [12] Y. Ye, Y. Shibata, C. Yun, D. Ron, T.A. Rapoport, A membrane protein complex mediates retro-translocation from the ER lumen into the cytosol, Nature 429 (2004) 841–847. [13] B.N. Lilley, H.L. Ploegh, A membrane protein required for dislocation of misfolded proteins from the ER, Nature 429 (2004) 834–840. [14] X. Zhong, Y.L. Ran, J.L. Lou, D. Hu, L. Yu, Y.S. Zhang, et al., Construction of human liver cancer vascular endothelium cDNA expression library and screening of the endothelium associated antigen genes, World J. Gastroenterol. 10 (2004) 1402–1408. [15] E.A. Jaffe, R.L. Nachman, C.G. Becker, et al., Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria, J. Clin. Invest. 52 (1973) 2745–2756. [16] J.L. Wang, Y.H. Liu, M.C. Lee, T.M. Nguyen, C. Lee, A. Kim, et al., Identification of tumor angiogenesis-related genes by subtractive hybridization, Microvas. Res. 59 (2000) 394–397. [17] S. Stie, G. Totzke, E. Grunewald, T. Neuhaus, S. Fronhorffs, A. Sachinidis, et al., Identification of syntenin and other TNF-inducible genes in human umbilical arterial endothelial cells by suppression subtractive hybridization, FEBS Lett. 467 (2000) 299–304. [18] K.E. Sawin, T.J. Mitchison, L.G. Wordeman, Evidence for kinesinrelated proteins in the mitotic apparatus using peptide antibodies, J. Cell Sci. 101 (1992) 303–313. [19] Y. Jiang, Y.E. Liu, I.D. Goldberg, Y.E. Shi, c Synuclein, a novel heat shock protein-associated chaperone, stimulates ligand-dependent estrogen receptor a signaling and mammary tumorigenesis, Cancer Res. 64 (2004) 4539–4546. [20] J.P. Girard, E.S. Baekkevold, T. Yamanaka, G. Haraldsen, P. Brandtzaeg, F. Amalri, Heterogeneity of endothelial cells: the specialized phenotype of human high endothelial venules characterized by suppression subtractive hybridization, Am. J. Pathol. 155 (1999) 2043–2055. [21] E. Ruoslahti, Specialization of tumor vasculature, Nat. Rev. Cancer 2 (2002) 83–90. [22] N.N. Khodarev, J. Yu, E. Labay, T. Darga, C.K. Brown, H.J. Mauceri, et al., Tumor-endothelium interactions in co-culture: coordinated changes of gene expression profiles and phenotypic properties of endothelial cells, J. Cell Sci. 116 (2003) 1013–1022. [23] K.J. Travers et al., Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation, Cell 101 (2000) 249–258. [24] Z. Pawlowska, P. Baranska, H. Jerczynska, W. Koziolkiewicz, C.S. Cierniewski, Heat shock proteins and other components of cellular machinery for protein synthesis are up-regulated in vascular endothelial cell growth factor-activated human endothelial cells, Proteomics 5 (2005) 1217–1227. [25] M. Knop, A. Finger, T. Braun, K. Hellmuth, D.H. Wolf, Der1, a novel protein specifically required for endoplasmic reticulum degradation in yeast, EMBO J. 15 (1996) 753–763. [26] R.G. Gardner, G.M. Swarbrick, N.W. Bays, S.R. Cronin, S. Wilhovsky, L. Sleelig, C. Kim, R.Y. Hampton, Endoplasmic
1278
Y. Ran et al. / Biochemical and Biophysical Research Communications 348 (2006) 1272–1278
reticulum degradation requires lumen to cytosol signaling. Transmembrane control of Hrd1p by Hrd3p, J. Cell Biol. 151 (2000) 69–82. [27] Y. Oda, K. Okada, H. Yoshida, R. Kaufman, K. Nagata, K. Mori, Derlin-2 and Derlin-3 are regulated by the mammalian unfolded response and are required for ER-associated degradation, J. Cell Biol. 172 (2006) 383–393.
[28] N. Ferrara, VEGF and the quest for tumor angiogenesis factors, Nat. Rev. Cancer 2 (2002) 795–803. [29] N. Ferrara, H.P. Gerber, J. LeCouter, The biology of VEGF and its receptors, Nat. Med. 9 (2003) 669–676. [30] Q.T. Le, N.N. Denko, A.J. Giaccia, Hypoxic gene expression and metastasis, Cancer Metastasis Rev. 23 (2004) 293–310.