Expression of angiogenic factors such as vascular endothelial growth factor (VEGF) under conditions of cell stress involves both transcriptional and translational events, as well as an important role for inducible endoplasmic reticulum (ER) chaperones. Coexpression of VEGF and 150-kDa oxygen-regulated protein (ORP), a novel ER chaperone, in human glioblastoma suggested a link between angiogenesis and ORP150. C6 glioma cells stably transfected with ORP150 antisense displayed selectively reduced ORP150 expression. Tumors raised after inoculation of immunocompromised mice with ORP150 antisense C6 glioma transfectants demonstrated an initial phase of growth comparable to wild-type C6 glioma cells which was followed by marked regression within 8 days. Decreased density of platelet/endothelial cell adhesion molecule 1-positive structures within the tumor bed was consistent with reduced angiogenesis in C6 gliomas expressing ORP150 antisense, compared with tumors derived from C6 cells overexpressing ORP150 sense or vector controls. In vitro, inhibition of ORP150 expression decreased release of VEGF into culture supernatants; in ORP150 antisense transfectants, VEGF accumulated intracellularly within the ER. These findings demonstrate a critical role for the inducible ER chaperone ORP150 in tumor-mediated angiogenesis via processing of VEGF, and, thus, highlight a new facet of angiogenic mechanisms amenable to therapeutic manipulation in tumors.

Formation of new blood vessels occurs throughout development and adult life, accompanying both physiological and pathophysiological situations (1). Among angiogenic factors, VEGF2 has come under intensive study because of its facile export from the cell, its unique targeting of the endothelium, and its association with a range of neoplastic and nonneoplastic disorders. For example, increased expression of VEGF in tumors has been shown to be critical for their growth, as blocking antibodies (2) or expression of dominant-negative VEGF receptor (3) suppressed tumor growth. Furthermore, up-regulation of VEGF in hypoxic tumor cells localizes its production to sites targeted for an angiogenic response (4).

Up-regulation of VEGF transcripts is often equated with cellular elaboration of the mature protein and induction of a robust angiogenic response. However, mechanisms underlying protein processing, also operating in an environment of cell stress (as in an hypoxic milieu), must be considered. In this context, the cellular response to oxygen deprivation is complex and involves, in addition to a switch to anaerobic metabolism, altered expression of a set of polypeptides termed ORPs (5). ORP150 was first identified and cloned from cultured astrocytes (6, 7) based on their ability to withstand and even produce neurotrophic factors in response to severe hypoxia (8). Expression of ORP150 in cultured human cells is essential for their survival under prolonged hypoxia (9). Subsequent studies have shown this polypeptide to be present in the ER and to share structural features with GRPs which function as chaperones, such as GRP78 and GRP94 (10). This suggests the possibility that ORP150 may be a component of the ER stress response to hypoxia potentially facilitating protein processing.

A link between tumor growth and ORP150 was suggested by our recent observation that ORP150 was up-regulated in human malignant tumors (11). Here, we demonstrate that increased levels of ORP150 promote VEGF processing with subsequent transport from ER to Golgi, followed by export out of the cell. In transplantable C6 glioma cells, enhanced expression of ORP150 promoted tumor growth (and was associated with increased VEGF antigen), whereas suppression of ORP150 caused regression of the neoplasm (and was associated with decreased VEGF antigen). Our results suggest an unexplored therapeutic target for modulation of angiogenesis, expression of the inducible ER chaperone ORP150.

Analysis of Human Glioblastoma and Rat C6 Glioma Tissue.

Human glioblastoma samples were obtained from surgical pathology specimens under the auspices of an approved Institutional Review Board protocol, and C6 gliomas were grown and harvested from nude mice as described previously (11). Adjacent sections were stained with either anti-human-ORP150 IgG (7), anti-VEGF IgG (Santa Cruz Biotechnology, Santa Cruz, CA), or anti-PECAM-1/CD31 IgG (Santa Cruz Biotechnology). Sites of primary antibody binding were visualized using an indirect immunoperoxidase technique with a commercially available kit (Vectastain avidin-biotin complex kit; Vector Laboratories, Burlingame, CA) as described previously (11).

Detection of Anti-ORP150 Autoantibody in Patient Sera.

Serum was obtained from six outpatients with a diagnosis of malignant tumor at the time of their regular clinic visit when blood was drawn for patient care purposes. Samples were generously provided by Dr. H. Tanigawa (Amagasaki Seikyo Hospital, Hyogo, Japan) with patient consent, along with samples from six age-matched controls, as described previously (12). The diagnoses in the six patients (tumor group) included: two patients with gastric cancer (65M and 71F), two patients with hepatocellular carcinoma (56M and 61M), one patient with adenocarcinoma of the gall bladder (67M), and one patient with thyroid cancer (48F).

To detect autoantibody to ORP150 in patient sera, an ELISA was developed as described previously (12). In brief, ORP150 was purified from 293 cells using ion exchange chromatography on fast protein liquid chromatography Mono Q, and ORP150 was absorbed onto microtiter wells of ELISA plates to which rabbit antihuman ORP150 IgG had already been bound. After incubation at 37°C for 1 h, plates were washed three times with PBS containing Tween 20 (0.05%) and incubated for 1 h at 37°C with the serum of patients at the indicated dilution in PBS containing Tween 20 (0.05%). Sites of primary antibody were visualized after addition of peroxidase-conjugated antihuman IgG (Boehringer Mannheim, Mannheim, Germany) and A495 was measured by Titertek (Dainippon Co., Tokyo, Japan). Anti-ORP150 antoantibody in patient serum was tittered by the diluting until A495 = 0.2 was obtained.

Cell Culture and Induction of Hypoxia.

C6 glioma cells were plated in DMEM containing FCS (10%) and penicillin/streptomycin (100 units/ml/100 μg/ml). When cultures achieved confluence, they were exposed to hypoxia using an incubator attached to a hypoxia chamber (Coy Laboratory Products, Ann Arbor, MI), which maintained a humidified atmosphere with low oxygen tension (8–10 Torr), as described previously (13). Oxygen tension in the medium was measured using a blood gas analyzer (ABL-2; Radiometer, Sweden).

Development of ORP150 Sense/Antisense Stably Transfected C6 Glioma Clones.

To construct the ORP150 antisense/sense vector, a fragment encoding the entire rat ORP150 cDNA was inserted into the vector pLNCX (Clontech Laboratories, Palo Alto, CA) in either antisense or sense orientation. The orientation of the inserted ORP150 cDNA fragment was confirmed by both restriction enzyme mapping and DNA sequencing. Transfection of sense/antisense ORP150 was performed according to the manufacturer’s protocol (Retro-X System; Clontech Laboratories). Selection and maintenance of neo-resistant transfectants were performed in the presence of G418 (1.0 mg/ml; Sigma, St. Louis, MO). After 14 days, single colonies were resuspended and grown in 96-well plates at a density of about one cell per well. Several cell lines were isolated, all of which were maintained in the presence of G418 (1.0 mg/ml). Cells were switched to medium free of G418 24 h prior to experiments. Assessment of cell viability used morphological criteria (continued adherence to the growth surface, maintenance of morphology) and release of LDH activity into the culture (Cacatua Chemical Co., Tokyo, Japan).

Northern and Western Blot Analysis.

About 20 μg of total RNA were extracted from C6 glioma cells (about 107 cells) exposed to either hypoxia or normoxia for 12 h, as described previously (6). The cDNA probe for Northern blot analysis of VEGF was generated by PCR (5′ primer, 5′-ACC ATGAAC TTT CTG-3′; 3′ primer, 5′-CCG CCT TGG CTT GTC ACA TCT GCA-3′), and the sequence of the probe was confirmed by DNA sequencing. For Western blotting, cultured cells (about 5 × 106cells) were exposed to hypoxia and lysed in PBS (200 μl) containing NP40 (1%), EDTA (5 mm), and PMSF (1 mm). Cell lysates were then subjected to SDS-PAGE/immunoblotting using either antihuman ORP150 antibody (concentration of antibody ×1000) (7) or anti-VEGF antibody (concentration of antibody ×400; Santa Cruz Biotechnology).

Growth of C6 Gliomas in Mice.

Experimental tumors were grown in mice as described elsewhere (14). In brief, confluent cultures of C6 glioma (about 3 × 107) cells were harvested with trypsin/EDTA (Life Technologies, Inc., Grand Island, NY) and resuspended in 200 μl of PBS, followed by s.c. injection into 8-week-old CD-1 nude mice (Charles River Breeding Laboratories, Wilmington, MA) with an 18-gauge needle in the flank. Mice injected with C6 glioma cells were monitored daily and tumor size was determined by caliper measurements. Tumor volume was calculated as length × width × height (mm3) based on measurements of three individuals unaware of the experimental protocol. Tumor experiments were performed a minimum of three times, and the experimental groups contained 10 animals each. In certain mice, 1 h before sacrifice, 1 mg of BrdUrd (Sigma) was injected by i.p. administration. Tumor tissue was retrieved, fixed in paraformaldehyde (4%), and sections were prepared. Sections were stained with H&E or were simultaneously immunostained for detection of BrdUrd (anti-BrdUrd antibody; Sigma) and for in situ apoptosis detection with a commercially available kit (Trevigen, Inc., MD). To assess vascular structures in tumor/wound tissue, immunostaining was performed with antimouse PECAM-1/CD31 antibody (Santa Cruz Biotechnology) according to the manufacturer’s protocol. Quantitative analysis was performed by scanning confocal microscopy and images were analyzed by NIH Image. Experiments were performed in triplicate by two individuals blinded to the protocol.

Determination of VEGF, bFGF, TGF-β1, and ORP150.

C6 glioma cells (5 × 105 cells) were either exposed to hypoxia/reoxygenation or maintained under normoxic conditions after placement in serum-free medium. Elaboration of VEGF antigen into culture supernatants was assessed by a commercially available kit (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. VEGF content of cell lysates, made by three freeze-thaw cycles in PBS (200 μl) containing EDTA (5 mm) and PMSF (1 mm), was determined by immunoblotting and densitometric analysis of the blots. For tissue levels of VEGF, samples (about 1 g) were homogenized using a Polytron homogenizer in 5 ml of ice-cold PBS containing EDTA (5 mm) and PMSF (1 mm). After centrifugation (5000 × g for 10 min at 4°C), supernatant was subjected to ELISA for VEGF, bFGF, and TGF-β1 antigens. Tissue homogenates were also used to determine the content of ORP150 antigen as described previously (15). In brief, homogenates (1 mg/ml; 0.1 ml) were coated on Nunc Maxi-Soap plates (Nunc, Roskilde, Denmark. Wells were incubated with anti-ORP150 IgG (30 μg/ml; 0.1 ml) for 2 h at 37°C, washed three times, and incubated with peroxidase-labeled goat antirabbit IgG (1:3000 dilution; 0.1 ml). Followed by the development with o-phenylenediamine dihydrochloride, ORP150 equivalent values were determined using a standard curve prepared with human recombinant ORP150 (7). In each case, protein content in the mixture was determined according to the method of Lowry et al.(16).

Subcellular Localization of VEGF.

Subcellular fractionation of C6 glioma cells was performed according to OptiPrep (Life Technologies, Inc.). In brief, C6 glioma cells (about 5 × 106 cells) were exposed to hypoxia for 24 h and homogenates were subjected to subcellular fractionation using a self-generated gradient. About 10 μg of protein from each sample were subjected to ELISA for VEGF. Fractions were also used for Western blot analysis with antimouse calnexin antibody (Transduction Laboratories, Lexington, KY) and antimouse TGN38 antibody (Transduction Laboratories) as markers for ER and Golgi apparatus, respectively. Quantitative measurement of VEFG content in ER was performed by subjecting the same lysate to ultracentrifugation (20% sucrose in PBS, 100,000 × g for 3 h at 4°C). The pellet was reconstituted in 200 μl of PBS, followed by the measurement of VEGF content by ELISA.

Immunoprecipitation.

Interaction of ORP150 and VEGF was assessed by immunoprecipitation using anti-ORP150 IgG and anti-VEGF IgG as described previously (6). In brief, about 5 × 106 C6 glioma cells were subjected to hypoxia and lysed in ice-cold PBS containing NP40 (1%), PMSF (1 mm), and EDTA (5 mm), followed by incubation with either rabbit anti-ORP150 IgG (3 μg/ml), nonimmune rabbit IgG (3 μg/ml), goat anti-VEGF IgG (20 μg/ml), or goat nonimmune IgG (20 μg/ml) overnight at 4°C. Mixtures were then incubated with protein G-Sepharose (2 mg/ml) for 2 h at 4°C, followed by the centrifugation at 2000 rpm for 10 min. Immunoprecipitates were then subjected to Western blotting with either anti-VEGF or anti-ORP150 IgG and a detection system as described above.

Statistical Analysis.

Statistical analysis was performed by using either the nonpaired t test or ANOVA followed by multiple comparison analysis. Where indicated, data were analyzed by two-way ANOVA followed by the multiple contrast analysis.

Expression of ORP150 in Tumors and Generation of Autoantibodies in Patient Plasma.

In nontransformed cells, expression of ORP150, analogous to that of VEGF, occurs most often in response to environmental stress, such as oxygen deprivation. Tumors also display high levels of ORP150, as shown in a section of human glioblastoma (Fig. 1,A-I displays an H&E section of the tumor, and A-II and B-I show immunostaining for ORP150). Some ORP150 likely gains access to the extracellular space, resulting in the production of autoantibodies detectable in the sera of patients with a range of neoplasms (A455 nm = 0.53 ± 0.03, n = 6) compared with age-matched controls (A455 nm = 0.31 ± 0.03, n = 6; P < 0.05; Fig. 1,D). ORP150 antigen appears concentrated in tumor cells infiltrating the stroma, especially those closely associated with neovasculature (Fig. 1, compare B-I and B-III), and demonstrates a distribution overlapping, at least in part, that of VEGF (Fig. 1, B-II and C). Consistent with these data, increased levels of ORP150 have been observed in a range of human tumors, including breast and pancreatic carcinoma (11).

Rat C6 glioma cells provide an excellent system to study the relationship between ORP150 and VEGF, as shown by their coexpression in tumors raised in CD-1 nude mice (Fig. 2); A and B display sites of ORP150 and VEGF expression, respectively, and C demonstrates their association with vasculature, showing the condensation of ORP150 and VEGF signals just around the neovessel (PECAM-1-positive cells). Cultured C6 glioma cells expressed ORP150, and levels increased at both the mRNA (data not shown) and antigen levels (8-fold increase; Fig. 3, A and B) when cells were subjected to hypoxia, consistent with previous findings (6, 7, 9, 12). Expression of VEGF antigen in hypoxic C6 glioma cultures followed a similar time course (Fig. 3 C).

Characterization of Stable Sense/Antisense ORP150-transfected C6 Glioma Cells in Vitro and in Vivo.

To analyze the role of ORP150 in tumor growth and its possible relationship to posttranslational processing of VEGF, C6 glioma cells were stably transfected with vectors encoding antisense ORP150 (AS-ORP150/C6), sense ORP150 (S-ORP150/C6), or vector alone. Transfectants were cloned by limiting dilution, and typical results with two different clones of each type are shown: AS-ORP150/C6 transfectants displayed about 50-fold reduced levels of ORP150 mRNA (Fig. 4,A, Lanes 1 and 2), whereas S-ORP150/C6 displayed about 10-fold increased ORP150 mRNA (Fig. 4,A, Lanes 5 and 6) compared with vector-transfected controls or wild-type C6 glioma cells (Fig. 4,A, Lanes 3 and 4, respectively), as assessed by the densitometric analysis (Fig. 4,B). Immunoblotting of C6 glioma cell lysates displayed similar differences in ORP150 antigen (Fig. 4 C; D shows densitometric analysis). Functional studies described below were performed on two lines derived from each of the stable transfectants in parallel.

Basic cellular properties were maintained in each of the transfected cell lines. Proliferation of C6 glioma cells, either S-ORP150/C6, AS-ORP150/C6, or vector-only transfectants, was comparable to that of wild-type cultures (data not shown). When these cultures of stably transfected tumor cells were subjected to hypoxia, viability of each of the lines was well maintained based on lack of an increase in LDH (Fig. 4 E). Under the same conditions, studies to assess indices of apoptosis, DNA fragmentation, and caspase-3 activity showed no changes comparing different clones of ORP150-transfected C6 glioma cells and wild-type controls (data not shown).

Although each of the ORP150 transfectants appeared to have similar properties in vitro, following their implantation into CD1-nude mice, the resulting tumors differed markedly (Fig. 5,A). Whereas wild-type, vector control, and S-ORP150/C6 cells produced tumors which steadily increased in volume up to 16 days (longest time point examined), after day 4, tumors from AS-ORP150/C6 cells displayed a progressive decline in tumor volume. By day 16, there was a 2000-fold difference in tumor volume comparing tumors arising from AS-ORP150/C6 cells with the other tumors (P < 0.0001). To analyze differences in tumor growth, cellular proliferation versus apoptosis was analyzed in neoplasms arising from the glioma cell lines. The percentage of proliferating and apoptotic nuclei, in each group of tumors derived from the different C6 glioma transfectants, was equivalent on day 2 (Fig. 5, B-I and B-II). However, as tumors derived from the AS-ORP150/C6 cells began to regress on day 4, these tumors displayed an enhanced apoptotic rate (Fig. 5,B-II), considerably exceeding the decline in the fraction of proliferating cells (Fig. 5,B-I). To confirm a possible role for ORP150 in tumor growth, we extended our studies to include several other clones of C6 glioma transfectants. There was an apparent relationship between tumor volume (evaluated 10 days after inoculation of tumor cells) and the level of tissue ORP150 level (Fig. 5 C).

A strong factor influencing apoptosis in the tumor bed would be the effectiveness with which the different tumor lines elicited formation of neovasculature (data not shown). Within 4 days of implantation, S-ORP150/C6 cells displayed an approximately 2-fold increase and AS-ORP150/C6 cells showed an approximately 4-fold decrease in vascular structures based on PECAM-1 staining, compared with tumors derived from wild-type and vector-transfected C6 glioma cells (Fig. 6,A). More aggressive angiogenesis was evident at day 4 in tumors arising from the S-ORP150/C6 cells from the presence of many vascular structures observed in the tumor itself (Fig. 6,A). There were fewer vascular profiles in neoplasms arising from wild-type and vector-only C6 glioma (Fig. 6,A) transfectants; most were evident at the tumor margin. The fewest vascular profiles were observed in tumors arising from the antisense transfectants (Fig. 6,A). Further studies using several different clones of C6 glioma transfectants demonstrated a close relationship between angiogenesis in tumors and tissue content of ORP150 (Fig. 6 B). These data suggested that tumors from AS-ORP150/C6 cells might have a reduced capacity to induce angiogenesis versus more aggressive vascular in growth in tumors arising from S-ORP150/C6 cells.

Contribution of ORP150 to Posttranslational Processing of VEGF.

VEGF has been shown to have a central role in the growth of C6 gliomas. Inhibition of VEGF using antibodies (2) and introduction of dominant-negative VEGF receptors (3) both have been demonstrated to suppress growth of gliomas. A relationship between VEGF and ORP150 expression was shown by ELISA of tissue extracts from the experimental tumors harvested on day 4 (Fig. 7,A). Whereas tumors derived from S-ORP150/C6 cells showed increased levels of VEGF antigen, there was a marked reduction in VEGF antigen in AS-ORP150/C6 cells (Fig. 7,A) compared with controls. As these data were expressed as nanograms of VEGF per milligram of total protein in the tumor, this represents a very large difference in overall VEGF content comparing the very small tumors derived from AS-ORP150/C6 cells with the large S-ORP150/C6-derived gliomas. In contrast, Northern blot analysis of tumor extracts showed comparable induction of VEGF transcripts in AS-ORP150/C6-derived gliomas versus tumors arising from sense and vector transfectants and wild-type C6 cells (Fig. 7 B). Furthermore, bFGF and TGF-β1 antigens showed no significant difference in tissue extracts from the different C6 glioma-derived tumors (data not shown). Thus, the effect of ORP150 on VEGF expression was occurring at a point distal to synthesis/stabilization of mRNA for this angiogenic factor.

To investigate mechanisms underlying ORP150 regulation of VEGF expression, studies were performed in cell culture. Consistent with our in vivo results, both S-ORP150/C6 and AS-ORP150/C6 cells in culture showed a similar elevation of VEGF mRNA when subjected to hypoxia, compared with wild-type or vector-transfected controls (Fig. 8,B). However, there was a marked difference in elaboration of VEGF into culture supernatants in the hypoxic cells; AS-ORP150/C6 released almost no VEGF antigen whereas S-ORP150/C6 elaborated increased amounts compared with controls (Fig. 8,A). These differences were not associated with altered cell viability (there was no evidence of LDH release or DNA fragmentation in cultures subjected to hypoxia), but, rather, were due to intracellular retention of VEGF in AS-ORP150/C6 cells. Immunoblotting showed high levels of VEGF in lysates of AS-ORP150/C6 compared with S-ORP150/C6 and control cultures (Fig. 8, C and D show densitometric analysis). These data suggested a role for ORP150 in posttranslational processing of VEGF allowing for efficient export of the angiogenic factor.

We hypothesized that ORP150 might function as a chaperone for VEGF in the ER, leading us to perform immunoprecipitation studies on hypoxic C6 glioma cells to determine whether VEGF and ORP150 might interact. Lysates of hypoxic C6 glioma cells were immunoprecipitated with anti-ORP150 monoclonal antibody followed by SDS-PAGE/immunoblotting with antibody to VEGF. A prominent immunoreactive band was observed (Fig. 9,A). Complementary studies in which cell lysates were immunoprecipitated with antibody to VEGF displayed the presence of ORP150 in these precipitates (Fig. 9 B).

These data, indicating an enhanced association of VEGF with ORP150 in C6 glioma cells under hypoxic conditions, suggested a possible role for ORP150 in VEGF processing and led us to perform subcellular fractionation studies. Wild-type C6 cells subjected to hypoxia showed two peaks of VEGF immunoreactivity (Fig. 9,B), corresponding to fractions enriched for calnexin and TGN38, markers of ER (17), and Golgi (18), respectively. In contrast, AS-ORP150/C6 subjected to hypoxia displayed VEGF immunoreactivity corresponding to fractions enriched for the ER marker only (Fig. 9,C; note that the peak of VEGF antigen in fractions 9 and 10 is probably because of immature protein unable to enter the ER). Furthermore, by extending these studies to several clones, a positive relationship was shown among the VEGF elaboration into culture supernatants and the ability of ORP150 induction under maximal stimuli (hypoxic condition), whereas the opposite was true with respect to VEGF accumulation in the ER-rich fraction (Fig. 9 D). These data supported a role for ORP150 in posttranslational events involving processing VEGF essential for transit from ER to Golgi, requisite steps for eventual export from the cell.

One of the salient features of the cellular biosynthetic response to hypoxia is redirection of protein synthesis with production of ORPs, which overlap with GRPs (5, 7, 19, 20). Common denominators of this group of polypeptides include their placement in the heat shock protein family, the presence of ER retention sequences, and their properties as molecular chaperones in the ER to Golgi transfer of newly synthesized proteins (21). Although originally described in tumor cells, ORPs/GRPs are known to be induced by hypoxic stress in a wide range of normal and tumor tissues and cell lines. This view highlights the role of ER stress in the cellular response to oxygen deprivation and suggests the possibility that enhancing expression/function of chaperones might accelerate repair consequent to hypoxic tissue injury. Conversely, suppressing expression of such chaperones might diminish cellular recovery from environmental stress.

Elaboration of VEGF, a selective mitogen for vascular endothelial cells, represents another major response of mammalian cells to insufficient ambient oxygen concentrations (22). VEGF is present at the appropriate time and location for guiding an angiogenic response which promotes tumor survival as cellular proliferation increases local metabolic needs; i.e., the supply of nutrients from currently available vasculature has outstripped the available blood supply (23). VEGF is also a main target of cancer gene therapy because of its strong effect on angiogenesis and actually blocking antibodies or overexpressing a soluble VEGF receptor suppressed tumor growth (2, 3, 24). These considerations led us to examine the VEGF-mediated angiogenesis from a different cutting edge, the posttranslational processing of VEGF in the ER.

Since VEGF requires posttranslational processing at the level of glycosylation and disulfide bond formation (25, 26, 27), we considered the possibility that in the setting of cellular stress imposed by hypoxia, optimal function of molecular chaperones would be essential for secretion of mature VEGF. In view of the potentially overlapping properties of GRP78, GRP94, and ORP150 (as well as other chaperones), each of which is also induced by hypoxia in C6 glioma cells (data not shown), it was not clear if suppressing ORP150 by itself might have an effect on VEGF expression. However, C6 glioma cells overexpressing the antisense ORP150 construct secreted virtually no VEGF (Fig. 8 A), although expression of GRP78 and GFP94 were unaffected (data not shown). This is consistent with our previous observation that ORP150 binds to the ER form of the Mr 80,000 glycoprotein (glycoprotein 80), a major secretory protein in Madin-Darby canine kidney cells, and provides a more efficient pathway for protein transport because of its higher affinity for ATP than of GRP78 (28). In addition, we have found that ORP150 enhanced elaboration of brain-derived neutrophic factor and nerve growth factor in primary cultures of neurons exposed to hypoxia and in an ischemic brain (29). These data suggest that ORP150 functions as molecular chaperone under hypoxia to facilitate the protein transport/processing in ER, though not specific to VEGF. In tumor formation process, tumor cells produce and secrete various kinds of cytokines or growth factors (30). bFGF and TGF-β1 are known to be important factors secreted by C6 glioma cells (31, 32). However, their levels showed no significant change among AS-ORP150/C6, S-ORP150/C6, and control cultures. It is possible that the beneficial effect of ORP150 on posttranslational processing/export in hypoxia is limited to proteins processed in the ER.

In our previous study, 293 cells overexpressing antisense ORP150 mRNA caused hypoxia-induced cell death (9), although C6 glioma cells overexpressing antisense ORP150 mRNA showed no significant changes in vitro comparing different clones of ORP150-transfected C6 glioma and wild-type cells. In vivo, tumors arising from the ORP150 antisense transfectants regressed in vivo, whereas tumors from wild-type C6 glioma cells continued to grow. However, until the neoplasm reached a size where angiogenesis was necessary to sustain further growth, the tumor cells transfected to overexpress the ORP150 antisense construct formed nodules that initially grew at a rate comparable to control tumor cells. These data suggest that the mechanism of cell death in tumors arising from the ORP150 antisense transfectants is mainly based on the inhibition of angiogenesis, which is caused by retention of VEGF in ER.

Taken together, these data highlight the role of ER chaperones in the cellular response to stress and suggest a new level of intervention, by modulating the expression of ORP150, especially in settings such as tumor growth where protein processing events in the ER may regulate angiogenesis.

Fig. 1.

Distribution of ORP150 and VEGF antigen in human tumors (A–C) and detection of human anti-ORP150 autoantibody in human sera from patients with tumors (D). Human glioblastoma from a 67-year-old male was subjected either to H&E staining (A-I) or immunohistochemical analysis using antibody to either ORP150 (A-II, B-I, and C-I), VEGF (B-II and C-II), or PECAM-1/CD31 (B-III). A-I and A-II are adjacent sections (original magnification, ×2), and B-I–III are adjacent sections (original magnification, ×100). C, sections were double stained with anti-ORP150 antibody (C-I; original magnification, ×200) and anti-VEGF antibody (C-II; original magnification, ×200). ORP150 and VEGF signals were digitally overlapped in C-III (original magnification, ×200). D, sera obtained from patients bearing malignant tumors (n = 6) were subjected to ELISA as described in the text. P < 0.05 by nonpaired t test compared with sera from age-matched control subjects (n = 6). Arrowheads in A-I, tumor.

Fig. 1.

Distribution of ORP150 and VEGF antigen in human tumors (A–C) and detection of human anti-ORP150 autoantibody in human sera from patients with tumors (D). Human glioblastoma from a 67-year-old male was subjected either to H&E staining (A-I) or immunohistochemical analysis using antibody to either ORP150 (A-II, B-I, and C-I), VEGF (B-II and C-II), or PECAM-1/CD31 (B-III). A-I and A-II are adjacent sections (original magnification, ×2), and B-I–III are adjacent sections (original magnification, ×100). C, sections were double stained with anti-ORP150 antibody (C-I; original magnification, ×200) and anti-VEGF antibody (C-II; original magnification, ×200). ORP150 and VEGF signals were digitally overlapped in C-III (original magnification, ×200). D, sera obtained from patients bearing malignant tumors (n = 6) were subjected to ELISA as described in the text. P < 0.05 by nonpaired t test compared with sera from age-matched control subjects (n = 6). Arrowheads in A-I, tumor.

Close modal
Fig. 2.

Expression of ORP150 and VEGF in C6 glioma in vivo. C6 gliomas grown in nude mice were subjected to immunohistochemical analysis using antibodies against either ORP150, VEGF, or PECAM-1. Tumor tissue was stained with antibody to ORP150 (A; original magnification, ×100), VEGF (B; original magnification, ×100), or PECAM-1 (C; original magnification, ×100).

Fig. 2.

Expression of ORP150 and VEGF in C6 glioma in vivo. C6 gliomas grown in nude mice were subjected to immunohistochemical analysis using antibodies against either ORP150, VEGF, or PECAM-1. Tumor tissue was stained with antibody to ORP150 (A; original magnification, ×100), VEGF (B; original magnification, ×100), or PECAM-1 (C; original magnification, ×100).

Close modal
Fig. 3.

Expression of ORP150 and VEGF in C6 glioma cells in vitro. Protein extracts from C6 cells were subjected to Western blotting using either antihuman ORP150 (A) or antimurine VEGF IgG (C). Densitometric analysis of multiple Western blots is shown in B, respectively (n = 6, mean ± SD is shown; ∗, P < 0.01 by nonpaired t test; and ∗∗, P < 0.01 by multiple comparison analysis, compared to normoxic level, □).

Fig. 3.

Expression of ORP150 and VEGF in C6 glioma cells in vitro. Protein extracts from C6 cells were subjected to Western blotting using either antihuman ORP150 (A) or antimurine VEGF IgG (C). Densitometric analysis of multiple Western blots is shown in B, respectively (n = 6, mean ± SD is shown; ∗, P < 0.01 by nonpaired t test; and ∗∗, P < 0.01 by multiple comparison analysis, compared to normoxic level, □).

Close modal
Fig. 4.

Establishment of ORP150 stable transfectants and their behavior in normoxic condition. Either total RNA (20 μg; A) or protein extract (10 μg; C) prepared from stably transfected clones of C6 glioma cells was analyzed by Northern (A) or Western (C) blot analysis: AS-ORP150/C6 (Lanes 1 and 2), nontransfected controls (Lane 3), vector alone (Lane 4), and S-ORP150/C6 (Lanes 5 and 6). Densitometric analysis of multiple Northern and Western blots from several clones is shown in B and D (n = 6, mean ± SD is shown; ∗∗, P < 0.01 by multiple comparison analysis, compared with wild-type). Migration of molecular weight markers (designated in thousands) is shown on the far right side of the gel (C). E, each type of transfectant and wild-type C6 glioma cells was exposed to hypoxia. Cell death was assessed by detecting LDH in the supernatant and is expressed as percentage of LDH activity present in cell lysates (n = 6, mean ± SD is shown).

Fig. 4.

Establishment of ORP150 stable transfectants and their behavior in normoxic condition. Either total RNA (20 μg; A) or protein extract (10 μg; C) prepared from stably transfected clones of C6 glioma cells was analyzed by Northern (A) or Western (C) blot analysis: AS-ORP150/C6 (Lanes 1 and 2), nontransfected controls (Lane 3), vector alone (Lane 4), and S-ORP150/C6 (Lanes 5 and 6). Densitometric analysis of multiple Northern and Western blots from several clones is shown in B and D (n = 6, mean ± SD is shown; ∗∗, P < 0.01 by multiple comparison analysis, compared with wild-type). Migration of molecular weight markers (designated in thousands) is shown on the far right side of the gel (C). E, each type of transfectant and wild-type C6 glioma cells was exposed to hypoxia. Cell death was assessed by detecting LDH in the supernatant and is expressed as percentage of LDH activity present in cell lysates (n = 6, mean ± SD is shown).

Close modal
Fig. 5.

Characterization of ORP150 stably transfected C6 gliomas in vivo. A, tumor volume at the indicated days in AS-ORP150/C6 (filled column), wild-type (light gray column), vector-only (dark gray column), or S-ORP150/C6 (open column) stable transfectants (n = 4, mean ± SD is shown; ∗∗, P < 0.001 compared with wild-type by multiple contrast analysis). B, BrdUrd (labeling index, B-I) and terminal deoxynucleotidyl transferase-mediated nick end labeling assay (cell death index, B-II) were determined on days 2 and 4 after implantation of each of the stably transfected lines of C6 glioma cells as in A (n = 20, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type by multiple comparison analysis). C, experiments similar to those in A but using additional C6 glioma-stable transfectants to establish tumors: four antisense clones (•), two wild-type (⋄), two vector-alone clones (♦), and four sense clones (○). X axis, tissue content of ORP150 antigen assessed by ELISA as described in the text.

Fig. 5.

Characterization of ORP150 stably transfected C6 gliomas in vivo. A, tumor volume at the indicated days in AS-ORP150/C6 (filled column), wild-type (light gray column), vector-only (dark gray column), or S-ORP150/C6 (open column) stable transfectants (n = 4, mean ± SD is shown; ∗∗, P < 0.001 compared with wild-type by multiple contrast analysis). B, BrdUrd (labeling index, B-I) and terminal deoxynucleotidyl transferase-mediated nick end labeling assay (cell death index, B-II) were determined on days 2 and 4 after implantation of each of the stably transfected lines of C6 glioma cells as in A (n = 20, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type by multiple comparison analysis). C, experiments similar to those in A but using additional C6 glioma-stable transfectants to establish tumors: four antisense clones (•), two wild-type (⋄), two vector-alone clones (♦), and four sense clones (○). X axis, tissue content of ORP150 antigen assessed by ELISA as described in the text.

Close modal
Fig. 6.

Dependency of neovascularization on ORP150 content in C6 glioma cells. A, neovascularization was assessed in tumors derived from AS-ORP150/C6, wild-type, vector control, and S-ORP150/C6 transfectants harvested after 4 days of growth in nude mice. PECAM-1/CD31 immunostaining followed by quantitative analysis to determine percent area occupied by PECAM-1/CD31 immunoreactivity (n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type controls by multiple comparison analysis). B, plots area occupied by PECAM-1/CD31 immunoreactivity is plotted versus relative ORP150 level, determined as in Fig. 5,C. The symbols shown in B to depict C6 glioma clones are the same as in Fig. 5 C.

Fig. 6.

Dependency of neovascularization on ORP150 content in C6 glioma cells. A, neovascularization was assessed in tumors derived from AS-ORP150/C6, wild-type, vector control, and S-ORP150/C6 transfectants harvested after 4 days of growth in nude mice. PECAM-1/CD31 immunostaining followed by quantitative analysis to determine percent area occupied by PECAM-1/CD31 immunoreactivity (n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type controls by multiple comparison analysis). B, plots area occupied by PECAM-1/CD31 immunoreactivity is plotted versus relative ORP150 level, determined as in Fig. 5,C. The symbols shown in B to depict C6 glioma clones are the same as in Fig. 5 C.

Close modal
Fig. 7.

Characterization of VEGF expression in C6 glioma cells in vivo. A, VEGF ELISA was performed on protein extracts of tumors derived from AS-ORP150/C6 (filled column), wild type (light gray column), vector alone (dark gray column), and S-ORP150/C6 (open column; n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type by multiple comparison analysis). B, total RNA extracted from the indicated type of C6 glioma cell was subjected to Northern blotting with either radiolabeled VEGF cDNA probe (top) or β-actin probe (bottom).

Fig. 7.

Characterization of VEGF expression in C6 glioma cells in vivo. A, VEGF ELISA was performed on protein extracts of tumors derived from AS-ORP150/C6 (filled column), wild type (light gray column), vector alone (dark gray column), and S-ORP150/C6 (open column; n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild-type by multiple comparison analysis). B, total RNA extracted from the indicated type of C6 glioma cell was subjected to Northern blotting with either radiolabeled VEGF cDNA probe (top) or β-actin probe (bottom).

Close modal
Fig. 8.

Characterization of VEGF expression in C6 glioma cells in vitro. A, VEGF ELISA was performed on conditioned media from the indicated cultured C6 glioma-ORP150 transfectants exposed to hypoxia (24 h): AS-ORP150/C6 (filled column), wild type (light gray column), vector only (dark gray column), and S-ORP150/C6 (open column; n = 8, mean ± SD is shown; ∗∗, P < 0.005 by multiple contrast analysis compared with wild type). B, total RNA (20 μg/lane), prepared from the indicated type of cultured C6 glioma cell was exposed to hypoxia for 24 h and subjected to Northern blot analysis. C and D, lysates from C6 glioma cells subjected to hypoxia for up to 24 h were subjected to Western blot using anti-VEGF IgG (C). D, densitometric analysis of several experiments similar to that in C. Values are expressed as fold increase compared with wild-type C6 glioma cells (n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild type by multiple comparisons).

Fig. 8.

Characterization of VEGF expression in C6 glioma cells in vitro. A, VEGF ELISA was performed on conditioned media from the indicated cultured C6 glioma-ORP150 transfectants exposed to hypoxia (24 h): AS-ORP150/C6 (filled column), wild type (light gray column), vector only (dark gray column), and S-ORP150/C6 (open column; n = 8, mean ± SD is shown; ∗∗, P < 0.005 by multiple contrast analysis compared with wild type). B, total RNA (20 μg/lane), prepared from the indicated type of cultured C6 glioma cell was exposed to hypoxia for 24 h and subjected to Northern blot analysis. C and D, lysates from C6 glioma cells subjected to hypoxia for up to 24 h were subjected to Western blot using anti-VEGF IgG (C). D, densitometric analysis of several experiments similar to that in C. Values are expressed as fold increase compared with wild-type C6 glioma cells (n = 6, mean ± SD is shown; ∗∗, P < 0.01 compared with wild type by multiple comparisons).

Close modal
Fig. 9.

Binding of VEGF to ORP150 in C6 glioma cells and retention of VEGF antigen in ER in the antisense transformants. A, wild-type C6 glioma cells were placed in hypoxia for 24 h, lysates were immunoprecipitated with anti-ORP150 IgG (A-I) or anti-VEGF IgG (A-II), and were then subjected to SDS-PAGE/immunoblotting with anti-VEGF IgG (A-I) or anti-ORP150 IgG (A-II). Where indicated, the immunoprecipitating antibody was replaced with nonimmune IgG (Nonimmune). Western blot analysis was also performed in the cell lysate in C6 glioma cells exposed to hypoxia for 24 h using antibody to either VEGF (A-I; Western blot) or ORP150 (A-II; Western blot). B and C, lysates prepared from wild-type C6 glioma (B) cells and AS-ORP150/C6 stable transfectants (C) were exposed to hypoxia for 24 h and were then separated by OptiPrep, as described. Each fractionated sample was subjected to Western blotting with either anticalnexin IgG (top) or anti-TGN38 IgG (middle), along with the measurement of VEGF content by ELISA (bottom). D, studies were performed to assess VEGF antigen in culture supernatant using four antisense clones (•), two wild-type clones (⋄), two vector-alone clones (♦), and four sense clones (○). X axis, content ORP150 antigen determined by ELISA in cell extracts prepared after exposure of cells to hypoxia for 24 h, as described in the text.

Fig. 9.

Binding of VEGF to ORP150 in C6 glioma cells and retention of VEGF antigen in ER in the antisense transformants. A, wild-type C6 glioma cells were placed in hypoxia for 24 h, lysates were immunoprecipitated with anti-ORP150 IgG (A-I) or anti-VEGF IgG (A-II), and were then subjected to SDS-PAGE/immunoblotting with anti-VEGF IgG (A-I) or anti-ORP150 IgG (A-II). Where indicated, the immunoprecipitating antibody was replaced with nonimmune IgG (Nonimmune). Western blot analysis was also performed in the cell lysate in C6 glioma cells exposed to hypoxia for 24 h using antibody to either VEGF (A-I; Western blot) or ORP150 (A-II; Western blot). B and C, lysates prepared from wild-type C6 glioma (B) cells and AS-ORP150/C6 stable transfectants (C) were exposed to hypoxia for 24 h and were then separated by OptiPrep, as described. Each fractionated sample was subjected to Western blotting with either anticalnexin IgG (top) or anti-TGN38 IgG (middle), along with the measurement of VEGF content by ELISA (bottom). D, studies were performed to assess VEGF antigen in culture supernatant using four antisense clones (•), two wild-type clones (⋄), two vector-alone clones (♦), and four sense clones (○). X axis, content ORP150 antigen determined by ELISA in cell extracts prepared after exposure of cells to hypoxia for 24 h, as described in the text.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

2

The abbreviations used are: VEGF, vascular endothelial growth factor; ORP, oxygen-regulated protein; ORP150, Mr 150,000 ORP; PECAM-1, platelet/endothelial cell adhesion molecule 1; ER, endoplasmic reticulum; PMSF, phenylmethylsulfonyl fluoride; TGF, transforming growth factor; bFGF, basic fibroblast growth factor; LDH, lactate dehydrogenase; BrdUrd, bromodeoxyuridine; AS-ORP150/C6, antisense ORP150/C6; S-ORP150/C6, sense ORP150/C6.

1
Alvarez Arroyo M. V., Caramelo C., Angeles Castilla M., Gonzalez Pacheco F. R., Martin O., Arias J. Role of vascular endothelial growth factor in the response to vessel injury.
Kidney Int. Suppl.
,
68
:
S7
-S9,  
1998
.
2
Ferrara N., Carver-Moore K., Chen H., Dowd M., Lu L., O’Shea K. S., Powell-Braxton L., Hillan K. J., Moore M. W. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene.
Nature (Lond.)
,
380
:
439
-442,  
1996
.
3
Machein M. R., Risau W., Plate K. H. Anti-angiogenetic gene therapy in a rat glioma model using a dominant-negative vascular endothelial growth factor receptor 2.
Hum. Gene Ther.
,
10
:
1117
-1128,  
1999
.
4
Larcher F., Robles A. I., Duran H., Murillas R., Quintanilla M., Cano A., Conti C. J., Jorcano J. L. Up-regulation of vascular endothelial growth factor/vascular permeability factor in mouse skin carcinogenesis correlates with malignant progression state and activated H-ras expression levels.
Cancer Res.
,
56
:
5391
-5396,  
1996
.
5
Heacock C. S., Sutherland R. M. Induction characteristics of oxygen regulated protein.
Int. J. Radiat. Oncol. Biol. Phys.
,
12
:
1287
-1290,  
1986
.
6
Kuwabara K., Matsumoto M., Ikeda J., Hori O., Ogawa S., Maeda Y., Kitagawa K., Imuta N., Kinoshita K., Stern D., Yanagi H., Kamada T. Purification and characterization of a novel stress protein, the 150-kDa oxygen-regulated protein (ORP150), from cultured rat astrocytes and its expression in ischemic mouse brain.
J. Biol. Chem.
,
279
:
5025
-5032,  
1996
.
7
Ikeda J., Kaneda S., Kuwabara K., Ogawa S., Kobayashi T., Matsumoto M., Yura T., Yanagi H. Cloning and expression of cDNA encoding the human 150 kDa oxygen-regulated protein, ORP150.
Biochem. Biophys. Res. Commun.
,
230
:
94
-99,  
1997
.
8
Maeda Y., Matsumoto M., Ohtsuki T., Kuwabara K., Ogawa S., Hori O., Shui D. Y., Konoshita T., Kamada T., Stern D. Hypoxia-reoxygneation mediated induction of interleukin-6 in cultured rat astrocytes and expression in ischemic gerbil brain: a paracrine mechanism enhancing neuron-survival.
J. Exp. Med.
,
180
:
2297
-2308,  
1994
.
9
Ozawa K., Kuwabara K., Tamatani M., Yakatsuji K., Tsumakoto Y., Kaneda S., Yanagi H., Stern D., Ogawa S., Tohyama M. 150-kDa oxygen-regulated protein (ORP150) suppresses hypoxia-induced apoptotic cell death.
J. Biol. Chem.
,
274
:
6397
-6404,  
1999
.
10
Wynn R. M., Davie J. R., Cox R. P., Chuang D. T. Molecular chaperones: heat-shock proteins, foldases, and matchmakers.
J. Lab. Clin. Med.
,
124
:
31
-36,  
1994
.
11
Tsukamoto Y., Hirota S., Kuwabara K., Kawano K., Yoshikawa K., Ozawa K., Kobayashi T., Yanagi H., Kitamura Y., Tohyama M., Ogawa S. Expression of 150 kDa oxygen-regulated protein (ORP150), a new member of the HSP70 family, in human breast cancers.
Lab. Investig.
,
78
:
699
-706,  
1998
.
12
Tsukamoto Y., Kuwabara K., Hirota S., Ikeda J., Stern D., Yanagi H., Matsumoto M., Ogawa S., Kitamura Y. The 150 kDa oxygen-regulated protein (ORP150) is expressed in human atherosclerotic plaques and allows mononuclear phagocytes to withstand cellular stress on exposure to hypoxi and modified LDL.
J. Clin. Investig.
,
98
:
1930
-1941,  
1996
.
13
Ogawa S., Gerlach H., Esposito C., Macaulay A. P., Brett J., Stern D. Hypoxia modulates the barrier and coagulant function of cultured bovine endothelium.
J. Clin. Investig.
,
85
:
1090
-1098,  
1990
.
14
Saleh M., Stacker S. A., Wilks A. F. Inhibition of growth of C6 glioma cells in vivo by expression of antisense vascular endothelial growth factor sequence.
Cancer Res.
,
56
:
393
-401,  
1996
.
15
Brett J., Schmidt A. M., Yan S. D., Zou Y. S., Weidman E., Pinsky D., Nowygrod R., Neeper M., Przysiecki C., Shaw A., Migheli A., Stern D. Survey of the distribution of a newly characterized receptor for advanced glycation end products in tissues.
Am. J. Pathol.
,
143
:
1699
-1712,  
1993
.
16
Lowry O., Rosenbrough N. J., Farr L. A., Randall R. J. Protein measurement with the Folin phenol reagent.
J. Biol. Chem.
,
193
:
265
-275,  
1951
.
17
Bergeron J. J., Brenner M. B., Thomas D. Y., Williams D. B. Calnexin: a membrane-bound chaperone of the endoplasmic reticulum.
Trends Biochem. Sci.
,
19
:
124
-128,  
1994
.
18
Banting G., Ponnambalam S. TGN38 and its orthologues: roles in post-TGN vesicle formation and maintenance of TGN morphology.
Biochim. Biophys. Acta
,
1355
:
209
-217,  
1997
.
19
Roll D. E., Murphy B. J., Laderoute K. R., Sutherland R. M., Smith H. C. Oxygen-regulated 80 kDa protein and glucose regulated 78kDa protein are identical.
Mol. Cell. Biochem.
,
103
:
141
-148,  
1991
.
20
Lee A. S. Mammalian stress response: induction of the glucose-regulated protein family.
Curr. Opin. Cell Biol.
,
4
:
267
-273,  
1992
.
21
Herrmann J. M., Malkus P., Schekman R. Out of the ER: outfitters, escorts, and guides.
Trends Cell Biol.
,
9
:
5
-7,  
1999
.
22
Dvorak H. F., Sioussat T. M., Brown L. F., Berse B., Nagy J. A., Sotrel A., Manseau E. J., Van de Water L., Senger D. R. Distribution of vascular permeability factor (vascular endothelial growth factor) in tumors: concentration in tumor blood vessels.
J. Exp. Med.
,
174
:
1275
-1278,  
1991
.
23
Minchenko A., Bauer T., Salceda S., Caro J. Hypoxic stimulation of vascular endothelial growth factor expression in vitro and in vivo.
Lab. Investig.
,
71
:
374
-379,  
1994
.
24
Kong H. L., Hecht D., Song W., Kovesdi I., Hackett N. R., Yayon A., Crystal R. G. Regional suppression of tumor growth by in vivo transfer of a cDNA encoding a secreted form of the extracellular domain of the flt-1 vascular endothelial growth factor receptor.
Hum. Gene Ther.
,
9
:
823
-831,  
1998
.
25
Claffey K. P., Senger D. R., Spiegelman B. M. Structural requirements for dimerization, glycosylation, secretion, and biological function of VPF/VEGF.
Biochim. Biophys. Acta
,
1246
:
1
-9,  
1995
.
26
Keck R. G., Berleau L., Harris R., Keyt B. A. Disulfide structure of the heparin binding domain in vascular endothelial growth factor: characterization of post-translational modifications in VEGF.
Arch. Biochem. Biophys.
,
344
:
103
-113,  
1997
.
27
Walter D. H., Hink U., Asahara T., Van Belle E., Horowitz J., Tsurumi Y., Vandlen R., Heinsohn H., Keyt B., Ferrara N., Symes J. F., Isner J. M. The in vivo bioactivity of vascular endothelial growth factor/vascular permeability factor is independent of N-linked glycosylation.
Lab. Investig.
,
74
:
546
-556,  
1996
.
28
Bando Y., Ogawa S., Yamauchi A., Kuwabara K., Ozawa K., Hori O., Yanagi H., Tamatani M., Tohyama M. 150-kDa oxygen-regulated protein (ORP150) functions as a novel molecular chaperone in MDCK cells.
Am. J. Physiol.
,
278
:
C1172
-C1182,  
2000
.
29
Tamatani M., Matsuyama T., Yamaguchi A., Mitsuda N., Tsukamoto Y., Taniguchi M., Che Y. H., Ozawa K., Hori O., Nishimura H., Yamashita A., Okabe M., Yanagi H., Stern D. M., Ogawa S., Tohyama M. ORP150 protects against hypoxia/ischemia-induced neuronal death.
Nat. Med.
,
7
:
317
-323,  
2001
.
30
Lazar-Molnar E., Hegyesi H., Toth S., Falus A. Autocrine and paracrine regulation by cytokines and growth factors in melanoma.
Cytokine
,
12
:
547
-554,  
2000
.
31
Okumura N., Takimoto K., Okada M., Nakagawa H. C6 glioma cells produce basic fibroblast growth factor that can stimulate their own proliferation.
J. Biochem. (Tokyo)
,
106
:
904
-909,  
1989
.
32
Liau L. M., Fakhrai H., Black K. L. Prolonged survival of rats with intracranial C6 gliomas by treatment with TGF-β antisense gene.
Neurol. Res.
,
20
:
742
-747,  
1998
.