150-kDa Oxygen-regulated Protein (ORP150) Suppresses Hypoxia-induced Apoptotic Cell Death*

To determine the contribution of 150-kDa oxygen-regulated protein (ORP150) to cellular processes underlying adaptation to hypoxia, a cell line stably transfected to overexpress ORP150 antisense RNA was created. In human embryonic kidney (HEK) cells stably overexpressing ORP150 antisense RNA, ORP150 antigen and transcripts were suppressed to low levels in normoxia and hypoxia, whereas wild-type cells showed induction of ORP150 with oxygen deprivation. Inhibition of ORP150 in antisense transfectants was selective, as hypoxia-mediated enhancement of glucose-regulated protein (GRP) 78 and GRP94 was maintained. However, antisense ORP150 transfectants displayed reduced viability when subjected to hypoxia, compared with wild-type and sense-transfected HEK cells. In contrast, diminished levels of ORP150 had no effect on cytotoxicity induced by other stimuli, including oxygen-free radicals and sodium arsenate. Although cellular ATP content was similar in hypoxia, compared with ORP150 antisense transfectants and wild-type HEK cells, suppression of ORP150 expression was associated with accelerated apoptosis. Hypoxia-mediated cell death in antisense HEK transfectants did not cause an increase in caspase activity or in cytoplasmic cytochromec antigen. A well recognized inducer of apoptosis in HEK cells, staurosporine, caused increased caspase activity and cytoplasmic cytochrome c levels in both wild-type and antisense cells. These data indicate that ORP150 has an important cytoprotective role in hypoxia-induced cellular perturbation and that ORP150-associated inhibition of apoptosis may involve mechanisms distinct from those triggered by other apoptotic stimuli.

Cellular adaptations to environmental oxygen deprivation constitute an important protective pathway for the host response to ischemia (1). One essential component of this response includes increased dependence on anaerobic metabolism as an energy source (2,3). Enhanced glycolysis is facilitated by hypoxia-mediated up-regulation of the noninsulindependent glucose transporter-1 and several enzymes involved in glycolytic metabolism by a mechanism that involves hypoxia-inducible factor-1. Another facet of the cellular biosynthetic response to hypoxia involves expression of diverse oxygen-regulated proteins (ORPs) 1 (4). These stress proteins participate in many aspects of cellular functions, and their characterization provides important clues as to means through which cells sustain hypoxic injury.
Among cell populations in the central nervous system, astrocytes are programmed to withstand hypoxic and ischemic stress, compared with more vulnerable neurons (5). Because of their strategic location, astrocytes are thought to have neurotrophic roles in response to cell stress, as observed in inflammation (6), trauma (7), and ischemic cerebrovascular diseases (8). For this reason, we have characterized proteins whose expression is modulated in astrocytes subjected to oxygen deprivation. We have previously identified several stress proteins in hypoxic cultured astrocytes; polypeptides with molecular masses corresponding to 94, 78, 33, and 28 kDa proved to be identical to GRP94, GRP78/Bip, heme oxygenase-1, and HSP28, respectively (2). A major hypoxia-inducible stress protein with a mass of 150 kDa was cloned and characterized as ORP150 (9, 10), a new member of heat shock protein family located in the endoplasmic reticulum. ORP150 is likely identical to a previously described polypeptide with a mass of 170 kDa, observed as a band with the latter mobility on SDSpolyacrylamide gel electrophoresis in hypoxic Chinese hamster ovary cells (11). The localization of ORP150, GRP94, and GRP78/Bip (12) to the endoplasmic reticulum suggests that hypoxia induces a stress response whose focal point may be in this organelle.
Because of the striking induction of ORP150 expression in cultured astrocytes (10) and mononuclear phagocytes (13) exposed to hypoxia, we speculated that this protein might have a cytoprotective role in response to cell stress associated with oxygen deprivation. To address this issue, we have established a cell line stably transfected to overexpress ORP150 antisense RNA in human embryonic kidney (HEK) cells. Antisense-transfected HEK cells display increased vulnerability to hypoxia, with loss of cell viability due to apoptosis, compared with wildtype and sense-transfected cultures. These data indicate that ORP150 has a pivotal role in cytoprotective cellular mechanisms triggered by oxygen deprivation.

MATERIALS AND METHODS
Construction of the ORP150 Antisense Vector-The vector pCAGGS, provided by Dr. Jun-ichi Miyazaki (14), contains a neomycin resistance * This work was supported by a Grant-in-aid for Scientific Research on Priority Areas 04268103 from the Ministry of Education, Science and Culture, Japan. 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.
Introduction of pCAGGS-Antisense/Sense ORP150 into HEK Cells and Selection of Stable Transfectants-Ten micrograms of vectors, with or without ORP150-antisense/sense, were transfected into HEK cells using the Tfx (Promega) Lipofectamine method. Selection and maintenance of the neo-resistant transfectants were performed in the presence of G418 (Sigma, 1.5 mg/ml). 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 are maintained in the presence of G418 (1.5 mg/ml). Cells were switched to the G418-free medium 24 h prior to experiments.
Induction of Hypoxia-Cells were plated at a density of about 5 ϫ 10 4 cells/cm 2 in Dulbecco's modified Eagle's medium containing fetal calf serum (10%) and penicillin/streptomycin (100 units/ml, 100 g/ml). Where indicated, HEK transfected with either antisense, sense, or vector-only construct were maintained in media containing G418 (800 g/ml). Prior to experiments (24 h), culture medium was changed to the same medium without G418 for both wild-type and sense/antisensetransfected HEK cells. When cultures achieved confluence (which took the same time for wild-type HEK cells, sense, antisense, or vector-only HEK transfectants), they were exposed to hypoxia using an incubator attached to an hypoxia chamber (Coy Laboratory Products, Ann Arbor, MI) which maintained a humidified atmosphere with low oxygen tension (8 -10 torr), as described previously (15). Oxygen tension in the medium was measured using a blood gas analyzer (ABL-2, Radiometer, Sweden).
Western Blot and Northern Blot Analyses-Cultured cells (about 5 ϫ 10 6 cells) were exposed to hypoxia, and, at the indicated time points, cells were washed three times with ice-cold phosphate-buffered saline (PBS) and lysed in the presence of PBS (200 l) containing Nonidet P-40 (1%), EDTA (5 mM), and phenylmethylsulfonyl fluoride (1 mM). After centrifugation (5000 ϫ g for 5 min at 4°C), protein concentration of each supernatant was determined by the Lowry method (16). Samples were applied to SDS-polyacrylamide gel electrophoresis (7.5%), transferred to polyvinylidene difluoride membranes, and visualized with polyclonal anti-human ORP150 IgG (0.15 mg/ml; purified by protein A column) raised to recombinant ORP150 polypeptide (residues 508 -999) (17). Where indicated, samples were also subjected to immunoblotting with anti-KDEL monoclonal antibody (Stressgen, Canada) according to the manufacturer's protocol.
Expression of ORP150 mRNA in HEK cells exposed to hypoxia employed a partial ORP150 cDNA corresponding to 151-570 base pairs (amino acids 51-190) for Northern analysis, as described (13). About 20 g of total RNA was extracted from HEK cells exposed to hypoxia or normoxia for the indicated times by the AGPC method; RNA was separated by electrophoresis on 1.0% agarose/formamide gels and transferred overnight onto Biodyne B paper (Pall BioSupport, New York). The membrane was prehybridized for 3 h at 42°C in hybridization buffer (0.9 M NaCl, 90 mM sodium citrate, pH 7.0) containing 5ϫ Denhardt's solution, SDS (0.5%), and heat-denatured salmon sperm DNA (100 g/ml). ORP150 cDNA was radiolabeled with [ 32 P]dCTP (NZ522, NEN Life Science Products) by the random hexamer procedure (18). After hybridization overnight at 42°C in hybridization buffer containing radiolabeled cDNA probe (5 ng/ml), filters were washed twice with 2ϫ SSC, 0.5% SDS and 0.2ϫ SSC, 0.5% SDS for 30 min at 52°C, exposed to x-ray film (Fuji Photo Film, Japan), and subjected to autoradiography. Blots were also hybridized with a radiolabeled probe for glyceraldehyde-3-phosphate dehydrogenase to serve as a control for RNA loading.
Analysis of Cell Viability-To assess cell viability, wild-type HEK cells and stable transformant HEK cells transfected with either antisense, sense, or vector-only construct were either exposed to hypoxia or to sodium arsenate or hydrogen peroxide under normoxic conditions. Release of lactate dehydrogenase (LDH) activity into the culture medium was measured according to the manufacturer's protocol (Cacatua Chemical Co., Tokyo, Japan). Levels of cytoplasmic histone-associated DNA fragments were assayed using the Cell Death enzyme-linked immunosorbent assay (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer's manual. In brief, the cytosolic fraction (13,000 ϫ g supernatant) of approximately 500 cultured cells was used as antigen source in a sandwich enzyme-linked immunosorbent assay with primary anti-histone antibody coated to the microtiter plate and secondary anti-DNA antibody coupled to peroxidase. From the absorbance values, the fold increase of fragmentation was calculated according to Equation 1 (19) using either normoxic or non-treated counterparts as control cells.
Fold increase ϭ ͑absorbance hypoxic cells Ϫ absorbance blank ͒ ͑absorbance control cells Ϫ absorbance blank ͒ (Eq. 1) Where indicated, HEK cells were placed in hypoxia (as above), or in normoxia, in the presence of sodium arsenate (10 mM), or hydrogen peroxide (10 mM for 30 min) and subsequently incubated for 6 h prior to harvest. Cell proliferation was determined by seeding 2.5 ϫ 10 4 cells into 3.0-cm 2 well and beginning the incubation period at 37°C. At later time points, cells were collected by trypsinization, and cell number was determined with a Coulter counter as described (20). Mitochondrial function was also assessed by HEK cell incorporation of dimethylthiazol-diphenyltetrazolium (Wako Chemicals) as described (21). For this assay, cells were incubated for 3 h with dimethylthiazol-diphenyltetrazolium (500 g/ml), lysed with HCl (0.05 M)/isopropyl alcohol, and A 570 nm was determined. DNA Fragmentation-In addition to the enzyme-linked immunosorbent assay for cytoplasmic histone-associated DNA fragments, DNA fragmentation was directly evaluated by agarose gel electrophoresis (22). For isolation of DNA, about 10 5 of HEK cells were resuspended in 200 l of ice-cold PBS, containing 0.5 mg/ml proteinase K, 0.5 mg/ml RNase A, and 1% SDS, and incubated at 37°C for 30 min. After addition of 300 l of NaI solution (6 M NaI, 13 mM EDTA, 0.5% sodium N-lauroylsarcosine, 10 mg/ml glycogen, and 26 mM Tris-HCl, pH 8.0), tubes were incubated at 60°C for 15 min. DNA in the supernatant (fragmented DNA) was precipitated in isopropyl alcohol, air-dried, and analyzed on 2% agarose gels by ethidium bromide staining and ultraviolet transillumination.
Morphologic and Ultrastructural Analysis-Fluorescence microscopy utilized ORP150 antisense transfectants (10 6 cells) cultured under hypoxia for 18 h. Cells were washed twice with ice-cold PBS, stained with propidium iodide (5 g/ml), and analyzed by fluorescence microscopy with excitation at 360 nm. Quantitative assessment of propidium iodide-positive cells was performed by counting the number of fluorescent cells by an investigator without knowledge of the experimental protocol. Transfer of phosphatidylserine to the outer leaflet of the plasma membrane was quantitatively assessed by increased binding of annexin V (23) using a kit (Genzyme) according to the manufacturer's instructions. In brief, either wild-type or antisense-transfected HEK cells were exposed to hypoxia for the indicated periods, washed three times in PBS, and incubated with fluorescein isothiocyanate-labeled annexin V for 20 min at 37°C. After extensive washing in ice-cold PBS, cells were examined by fluorescence microscopy with excitation at 488 nm, and positive cells were quantitated by an individual unaware of the experimental protocol.
Electron microscopic analysis utilized either wild-type or ORP150 antisense transfectants exposed to hypoxia for the indicated periods and fixed in 6-well culture dishes in cold picric acid (0.2%), paraformaldehyde (4%), and glutaraldehyde (0.05%) in 0.1 M phosphate buffer, pH 7.4, for 2 h at 4°C. Cells were washed in phosphate buffer, postfixed for 1 h with 1% osmium tetroxide (OsO 4 ) in 0.1 M phosphate buffer, pH 7.4, at 4°C, and dehydrated. Ultrathin sections were stained with lead citrate and subjected to ultrastructural analysis.
Measurement of Adenosine Nucleotide-The content of high energy adenosine metabolites was measured as described previously (3). In brief, either antisense transfectants or wild-type HEK cells (about 10 6 cells) plated on 9.6-cm 2 wells were exposed to hypoxia. At the indicated times, cultures were washed with ice-cold PBS, pelleted, and lysed in FIG. 2. Expression of ORP150 antigen (A), ORP150 transcripts (B), and GRP78 and GRP94 antigens (C) in cultured wild-type HEK cells and ORP150 antisense-transfected HEK cells exposed to hypoxia. A, protein extract from either wild-type HEK cells (20 g/lane; left panel) or ORP150 antisense transfectants (5 g/lane; right panel) exposed to hypoxia for the indicated times (0 means 0 h of hypoxia, as in the normoxic control) was subjected to Western blotting as described in the text. Note that a different amount of protein was loaded for the antisense transfectant. B, same amount of protein (20 g/lane) was loaded and subjected to Western blot using anti-KDEL monoclonal antibody (Stressgen) for GRP78 and GRP94 detection. C, total RNA (20 g/lane) prepared from either wild-type (wild) or antisense (anti)-transfected HEK cells was subjected to Northern analysis with radiolabeled human ORP150 cDNA probe. The migration of ribosomal RNA is indicated on the far right of the gel. GPDH, glyceraldehyde-3-phosphate dehydrogenase.

FIG. 3. Viability of wild-type HEK cells and ORP150 antisensetransfected HEK cells in hypoxia. Either antisense-transfected HEK cells (open bar), wild-type HEK cells (closed bar), sense-transfected HEK cells (hatched bar)
, or vector-only transfectants (dotted bar) were exposed to hypoxia up to 30 h. At the indicated times, fragmented DNA (A) and LDH activity (B) in the culture supernatant was measured as described in text. Note that all values of each group at time "0" in A are equal to 1.0 according to the formula described in text. Mean Ϯ S.D. is shown (n ϭ 6) for DNA fragmentation. Mean Ϯ S.D. is shown (n ϭ 24) for LDH release. 400 l of ice-cold perchloric acid (1 M). After centrifugation (4,000 ϫ g for 5 min), the supernatant was neutralized with KOH (2 M) and subjected to reversed phase high pressure liquid chromatography (C 18 column, Rainin). Pellets were stored for assay of protein content. The peak absorbance at 260 nm corresponded to ATP and was identified by its retention time on the column. The content of each adenosine metabolite (ATP, ADP, and AMP) was calculated using Chromatopack (Shimazu, Kyoto, Japan).

Measurement of Caspase-1-and -3-like Activity and Western Blotting for Cytosolic Cytochrome c and Caspase Substrates-Activity of
Caspase-1-and -3-like protease was assessed as described (22). In brief, either wild-type or ORP150 antisense-transfected HEK cells were exposed either to staurosporine (1 M) in normoxia or to hypoxia for the indicated times. At each time point, caspase-1-and -3-like activities were measured colorimetrically by cleavage of 7-amino-4-methylcoumarin-YVAD and 7-amino-4-methylcoumarin-DEVD, respectively. One unit was defined as the amount of enzyme required to release 0.22 nmol of AMC per min at 37°C. Immunoblotting of cytoplasmic cytochrome c was assessed by the Western blot as described (24).
To assess the possible contribution of caspase activation in the hypoxia-mediated cell death in antisense-transformant cells, polyclonal antibody for recombinant human caspase-9 was raised in rabbits. In brief, full-length caspase-9 cDNA was fused into glutathione S-transferase gene in plasmid pGEX-1T (Pharmacia Biotech, Uppsala, Sweden). Glutathione S-transferase-caspase-9 recombinant protein was expressed in Escherichia coli and purified by affinity chromatography to immunize rabbits.

Characterization of Stable Transfectants
Overexpressing Antisense/Sense ORP150 Transcripts-The ORP150 antisense/ sense construct was prepared by inserting a construct comprised of virtually the full coding sequence of human ORP150 and a portion of 3Ј-untranslated sequence (ϩ86/ϩ3093) into the EcoRI site of pCAGGS vector in either reverse or forward orientation (pCAGGS-ORP150 antisense or pCAGGS-ORP-150 sense, respectively; Fig. 1A). After transfection of pCAGGS-ORP150 antisense/sense into HEK cells, transfectants were selected and cloned in medium containing G418 and were designated ORP150 antisense or sense transfectants. Compared with wild-type HEK cells (Fig. 1B, lane 1) or cultures exposed to vector alone (Fig. 1B, lane 2), immunoblotting of lysates from two clones of ORP150 sense transfectants showed higher ORP150 antigen (Fig. 1B, lanes 3 and 4), whereas two clones of ORP150 antisense transformants displayed lower levels of ORP150 (Fig. 1B, lanes 5  and 6). Laser densitometry indicated a 10-fold increase in ORP150 expression in sense transfectants and about 40-fold suppression in antisense transfectants. For the experiments shown below, studies were performed in parallel with each of these cell lines, and representative results are shown.
Growth and Viability of ORP150 Sense/Antisense Transfectants-Proliferation of HEK cells, either ORP150 antisense/ sense/vector-only transfectants compared with wild-type cultures, was assessed by determining cell number (Fig. 1C) or by the dimethylthiazol-diphenyltetrazolium method. In each case, growth in cell culture over 5 days was comparable; there was no significant difference among wild-type, antisense, and sense transfectants to achieve confluence (data not shown), and no evidence of increased cell death was noted by uptake of trypan blue (data not shown).

Expression of ORP150 in HEK Cell and Effect of Hypoxia on Cell Viability in ORP150
Transfectants-ORP150 was first identified by its enhanced expression in hypoxic cultured astrocytes (10). Wild-type HEK cells subjected to oxygen deprivation displayed a time-dependent increase in ORP150 antigen first noted at 6 -12 h and sustained up to the longest time point, 24 h ( Fig. 2A, wild-type). In contrast, levels of ORP150 in the antisense HEK transfectants increased only slightly at 6 -15 h ( Fig. 2A, antisense). Consistent with suppression of ORP150 expression in the antisense transfectants, ORP150 transcripts were virtually undetectable in ORP150 antisense transfectants subjected to normoxia or hypoxia, compared with strong upregulation of ORP150 mRNA in wild-type HEK cultures (Fig.  2B). Although expression of ORP150 was suppressed in antisense transfectants, hypoxia-mediated increase in the levels of two glucose-regulated proteins, GRP78 and GRP94 (30), was similar in wild-type (Fig. 2C, wild-type) and ORP150 antisense (Fig. 2C, antisense) transfectants. Thus, ORP150 antisense transfectants showed a selective change in ORP150 antigen which did not extend to other genes previously shown to undergo increased expression in hypoxia.
In view of the association of increased ORP150 expression with environmental hypoxia, we assessed whether ORP150 contributed to the cellular adaptive response triggered by oxygen deprivation. ORP150 antisense transfectants displayed increased levels of cytoplasmic histone-associated DNA fragments within 12 h of hypoxia, which reached maximum within 18 h (Fig. 3A). Evidence of DNA fragmentation was followed by release of LDH activity in the culture supernatant (Fig. 3B). In contrast, both wild-type HEK cells and sense transfectants did not show loss of viability up to 30 h. Similar results were observed with vector alone HEK transfectants (Fig. 3, A and B).
Since ORP150 expression was associated with protection from hypoxia-induced cell death, the effect of other stimuli on ORP150 antisense transfectants was evaluated. Cell viability was comparable in wild-type and ORP150 antisense transfectants following incubation of cultures with sodium arsenate, a potent degrading agent for cellular protein, and hydrogen peroxide, a generator of oxygen free radicals (Fig. 4, A and B).

FIG. 4. Cell viability of Wild HEK cells and ORP150 antisense transformant HEK cells under chemical stresses (A and B). Either wild-type HEK cells (closed bar) or ORP150 antisense-transfectants HEK cells (open bar)
were cultured in the presence of either sodium arsenate (10 mM; A) or hydrogen peroxide (10 mM, B) up to 24 h. At the indicated times, LDH activity in each culture supernatant was measured, and cell viability was expressed as a percentage of maximal release. Mean Ϯ S.D. is shown (n ϭ 24). In each case, experiments were repeated a minimum of four times.
Hypoxia-induced Cell Death in ORP150 Antisense Transfectants: Evidence of Apoptosis-Several lines of evidence demonstrated that accelerated cell death in ORP150 antisense transfectants was due to apoptosis. First, addition of cycloheximide suppressed cell death in ORP150 antisense transfectants (Fig.  5, A and B). In the presence of cycloheximide (2 /ml), overall protein synthesis was suppressed by Ͼ90%, as assessed by the incorporation of [ 3 H]leucine to trichloroacetic acid-precipitable material. When cycloheximide (2 g/ml) was added from 13 to 16 h after the start of hypoxia, levels of cytoplasmic histoneassociated DNA fragments (Fig. 5A) and LDH release (Fig. 5B) remained at basal levels, whereas addition of cycloheximide at 18 -20 h had no protective effect. Lower levels of cycloheximide (0.5 g/ml), which showed a lesser effect on overall protein FIG. 6. Morphology of ORP150 antisense HEK cell transfected following exposure to hypoxia. A, either wild-type or ORP150 antisense transfectants (10 5 cells) were maintained in hypoxia for up to 30 h. At the indicated times, cells were harvested, and DNA was extracted and subjected to agarose gel (2%) electrophoresis. DNA was visualized by ethidium bromide staining and ultraviolet light transillumination. B, either wild-type (B-I) or antisense-transfected HEK cells (B-II) were exposed to hypoxia for 18 h, fixed, and stained with propidium iodide (magnification ϫ 100 for I and II; ϫ 400 for III). C, either wild-type (C-I) or antisense-transfected HEK cells (C-II, -III, and -IV) were exposed to hypoxia for up to 20 h, fixed, and studied by transmission electron microscopy. Micrographs were obtained from cells exposed to hypoxia for 20 h (I; wild type) or 16  synthesis (about 50% suppression), added 14 h into the period of hypoxia, had a diminished protective effect (Fig. 5, C and D).
Several additional lines of evidence supported the conclusion that ORP150 antisense transfectants were more susceptible to hypoxic stress. First, DNA prepared from ORP150 antisense transfectants showed laddering on agarose gels after 15 h of exposure to hypoxia (Fig. 6A, Antisense). In contrast, wild-type HEK cell controls showed little change in the pattern of DNA migration up to the longest time point, 30 h (Fig. 6A, wild type). Second, morphologic study of ORP150 antisense transfectants revealed features consistent with apoptotic cell death (Fig. 6B). Fluorescence microscopy 18 h following exposure of cells to hypoxia, using propidium iodide to visualize DNA, displayed clumped nuclear chromatin and formation of "apoptotic bodies" in ORP150 antisense transfectants (Fig. 6B, II-III) compared with uniform faint DNA staining in wild-type controls (Fig. 6B,  I). Third, electron microscopy of ORP150 antisense transfectants subjected to hypoxia for 14 -18 h showed fragmented nuclei with condensed chromatin (Fig. 6C, II-III) and formation of apoptotic bodies (Fig. 6C, IV). Wild-type control cells harvested even after 20 h of the onset of hypoxia displayed normal nuclear morphology (Fig. 6C, I). Furthermore, transfer of phosphatidylserine to the outer leaflet of the plasma membrane was shown by the increased binding of annexin V to antisense-transfected HEK cells 12-16 h after the onset of hypoxia. In contrast, there was no increase in annexin V binding observed to wild-type HEK cells (Fig. 6D). Enhanced binding of annexin V to antisense-transfected HEK cells preceded the increase in propidium iodide-positive cells (Fig. 6E). The percentage of propidium iodide-positive cells, representing those undergoing the final stages of the apoptotic process, reached about 40% at maximum. Taken together, these data are consistent with apoptosis as the mechanism of cell death which is accelerated in ORP150 antisense transfectants.
Effect of Hypoxia on Energy Metabolism in ORP150 Antisense-transfected HEK Cells-Depletion of high energy phosphate compounds, such as ATP, has been suggested as a contributory factor for hypoxia-induced cell death. ATP content was well maintained in wild-type and ORP150 antisense transfectants during the first 12 h of hypoxia and slowly decreased, in parallel, over 16 -20 h (Fig. 7). In contrast, once the apoptosis was triggered in ORP150 antisense transfectants, by 24 h (see Fig. 3A), ATP levels became undetectable in the nonviable cells (Fig. 7). These data suggest that depletion of high energy metabolites is not the trigger for accelerated cell death in ORP150 antisense transfectants.
Caspase Activity in Hypoxic Wild-type HEK Cells and ORP150 Antisense Transfectants-To characterize further hypoxia-mediated cell death in antisense transfectants, apoptosis was induced in both wild-type HEK cells and antisense ORP150 transfectants by exposure to staurosporine (1 M; see Ref. 29). Staurosporine increased levels of cytoplasmic DNA fragments (Fig. 8A), as well as release of LDH activity into culture supernatants (Fig. 8B) in both wild-type and antisensetransfected HEK cells. This was preceded by an increase in caspase-like-1 and -3 activities in ORP150 antisense transfectants (Fig. 9, A and C). In contrast, exposure to hypoxia caused no apparent increase in these caspase activities (Fig. 9, B and  D). Furthermore, hypoxic HEK antisense transfectants did not display increased cytosolic cytochrome c antigen (Fig. 9E), although apoptosis was proceeding based on evidence of DNA fragmentation and morphologic criteria (Fig. 6). However, staurosporine-treated HEK cells, both wild-type and ORP150 antisense transfectants, showed an increase in cytoplasmic cytochrome c (Fig. 9F).

FIG. 7. ATP levels in wild-type and antisense-transfected HEK cells exposed to hypoxia. Either antisense transfectants (open bars)
or wild-type (closed bars) HEK cells were exposed to hypoxia (0 -36 h), and ATP content was measured by reversed phase high pressure liquid chromatography as described in text. Mean Ϯ S.D. is shown (n ϭ 6). ND denotes the time point at which ATP content could not be determined due to the cell death.
ptosis, no such activation was observed in the hypoxia-mediated cell death in ORP150 antisense-transfected HEK cells ( Fig. 10 A-C and E). Activation of caspase-8 was not observed in antisense-transfected HEK cells, either cultures exposed to hypoxia or those treated with staurosporine (Fig. 10D). The activated form of this caspase was not detected even after the prolonged exposure of the membrane to the film. These data suggest that hypoxia-mediated cell death in the antisense ORP150 transfectants involved, at least in part, distinct pathways from those observed in mitochondria-initiated cell death following exposure of HEK cells to staurosporine (31). DISCUSSION Expression of ORP150, a stress protein originally purified and cloned from cultured rat astrocytes subjected to hypoxia, has been observed in several types of cells subjected to oxygen deprivation (9,10,13). Factors responsible for ORP150 expression are also present in atherosclerotic lesions and in breast cancers. In atherosclerosis, mononuclear phagocytes infiltrating vascular lesions stained for ORP150 and lipid loading increased ORP150 levels in cultured monocyte-derived cells (13). In breast cancer, tumor cells at the rapidly growing periphery of lesions displayed high levels of ORP150, regardless of estrogen receptor status of the patients, suggesting a role of this stress protein for cancer cells to invade the less vascular area (17). Constitutive ORP150 expression is noted in tissues with a well developed endoplasmic reticulum required to synthesize large amounts of secretory proteins (e.g. liver and pancreas) (10). This is similar to what has been reported for in GRP78 and GRP94 (32) and is consistent with a possible role for ORP150 FIG. 10. Activation of caspases in response to staurosporine or hypoxia in antisense ORP150-transfected HEK cells. Antisense ORP150-transfected HEK cells (about 10 6 cells) were either exposed to hypoxia or treated with staurosporine for the indicated period, and cells were then lysed in PBS containing Nonidet P-40 (2%), SDS (0.5%), and deoxycholic acid (0.5%). Protein extracts (10 g/lane) were applied to SDS-polyacrylamide gel electrophoresis, followed by the Western blotting using either goat anti-poly(ADP-ribose) polymerase (A), goat anticaspase-2 antibody (B), goat anti-caspase-7 antibody (C), mouse anticaspase-8 antibody (D), or rabbit anti-caspase-9 antibody (E). Proform and processed forms of these enzymes are indicated by open and closed arrowheads, respectively. The migration of simultaneously run molecular weight markers is indicated on the left side of the gel. as a molecular chaperone and participant in protein folding.
The localization of ORP150 in the endoplasmic reticulum, along with other polypeptides whose expression is increased in hypoxia, such as GRP94, and GRP78/Bip, suggests that hypoxia may be a situation characterized by an endoplasmic reticulum stress response (33). Consistent with an important role for GRP78, the most abundant member of this family expressed constitutively in the endoplasmic reticulum, homologous recombination experiments in yeast to delete the GRP78 gene (34) as well as manipulation of GRP78 levels (30) have shown this factor to have a central role in cellular homeostasis under physiologic conditions (i.e. in normoxia without other metabolic stress) and in the cellular response to environmental stress. The results of our studies with HEK cells stably transfected with an antisense ORP150 construct, the latter resulting in suppression of ORP150 expression, provide insights into the contribution of this polypeptide to cellular properties. Whereas antisense ORP150-transfected HEK cells proliferate and maintain their viability under normoxic conditions, following exposure to hypoxia, cell death increases compared with control cells (wild-type and ORP150 sense-transfected HEK cells). Furthermore, although diminished expression of ORP150 potentiated hypoxia-induced cytotoxicity, antisense ORP150-transfected HEK cells did not show increased vulnerability to other stresses such as sodium arsenate and hydrogen peroxide. Suppression of ORP150 in antisense HEK transfectants appeared to be selective, as hypoxia-mediated up-regulation of GRP78 and GRP94 was still observed.
When HEK cells with suppressed ORP150 expression were exposed to hypoxia, they demonstrated apoptosis based on a number of criteria as follows: (a) transfer of phosphatidylserine to the outer leaflet of the plasma membrane; (b) nuclear condensation and nuclear disorganization (the latter including disruption of nucleoli and margination of chromatin in discrete masses along the inner side of the nuclear membrane); (c) DNA fragmentation; and (d) inhibition of cell death by the addition of cycloheximide (although the specificity of using cycloheximide to identify apoptosis is limited, see Ref. 35). However, events leading to the final common pathway of DNA fragmentation in ORP150 antisense-transfected HEK cells subjected to hypoxia differ from those in normoxic HEK cells incubated with staurosporine (25). Although release of mitochondrial cytochrome c antigen into the cytosol increased caspase-like proteinase activity, and the cleavage of major caspases are central events in apoptosis associated with mitochondrial dysfunction (36 -38), neither was observed in hypoxic antisense ORP150 HEK transfectants undergoing apoptosis. These data suggest that hypoxia-mediated cell death in the ORP150 antisense-transfected HEK cells may involve distinct pathway(s) from those previously associated with mitochondria-initiated cell death.
Retention of immature/unfolded protein in the endoplasmic reticulum can cause apoptosis in several cell lines. The accumulation of dengue virus protein in the endoplasmic reticulum, for example, results in cell death in a mouse neuronal cell line (39). Furthermore, in HEK cells, accumulation of immature protein in the endoplasmic reticulum has been shown to trigger cell death by activation of the transcriptional factor NF-B (33). This led us to consider whether the apoptosis-like cell death observed in the ORP150 antisense HEK transfectants exposed to hypoxia might be associated with NF-B activation. However, no activation of NF-B was noted in hypoxic HEK cells (either wild-type or antisense ORP150 transfectants), and addition of the inhibitor pyrrolidine dithiocarbamate (40) was without effect in our experimental system. In contrast, pilot data concerning free Ca 2ϩ in HEK cells indicate impaired buffering of endoplasmic reticulum stores in the antisense ORP150 transfectants (data not shown), suggesting potential dysfunction of the endoplasmic reticulum in the ORP150 antisensetransfected HEK cultures.
Low ambient oxygen concentration forces cells to express a set of stress proteins, thereby facilitating vital functions of cellular organelles. We have demonstrated that one of those stress proteins, ORP150, is likely to subserve a cytoprotective role at the level of the endoplasmic reticulum; ORP150 enhances cellular ability to sustain oxygen deprivation. Analysis of functional properties of ORP150 is likely to add to our understanding of how the cellular response to hypoxia buttresses intracellular mechanisms enhancing survival due to environmental stress.