Cloning and Characterization of a Novel GRP78-binding Protein in the Rat Brain*

The full-length cDNA clone of a novel GRP78-binding protein (GBP) was isolated from rat brain using PCR-selected cDNA subtraction. GBP was predominantly expressed in neuronal cells among various brain tissues. GBP mRNA was already detected in the E12 brain and then gradually increased to reach a peak within P0–2 weeks after birth. GBP expression in the brain decreased age-dependently to (cid:1) 30% of the postnatal level at 12 months. GBP encoded 1021 amino acids and was predicted to have two transmembrane regions and glutamic acid- and proline-rich regions. Because the sequence of GBP offered few clues to the possible function, we performed a GST-tagged GBP pull-down assay in PC12 lysates and identified GRP78, one of the heat shock proteins, as a counterpart. Observation of COS7 cells expressing green fluorescent protein- or Myc-tagged GBP showed that GBP was localized in the endoplasmic reticulum-Golgi domain where BODIPY 558/568 (4,4-difluro-5-(2-thienyl)-4-bora-3 (cid:1) ,4 (cid:1) -diaza- S -indacene)-labeled brefeldin A accumulated. To investigate a biological role for GBP, we established Neuro2a cells stably Northern various was purified, resolved by agarose gel electrophoresis under denaturing and blotted onto membranes. To investigate the expression of GBP mRNA in the various brain regions and in the whole brain at indicated ages, we purchased RNA-blotted membrane from Seegene Co. Ltd. An - 32 P-radiolabeled GBP cDNA fragment containing nucleotides 2972–3248 was hybridized in the Clon-tech hybridization solution. After washing, the membranes were ex- posed to a Fuji imaging plate and analyzed with BAS2000 (Fuji Film, Japan). In order to control for the amount of RNA present, the mem- branes were stripped and re-hybridized with a G3PDH cDNA probe. RT-PCR was also performed to estimate the mRNA levels of various tissues and cell lines as described previously (13). Briefly, 0.5 (cid:1) g of total RNA was converted into cDNA by reverse transcriptase, and specific DNAs were amplified by PCR using Taq polymerase. The PCR primers used in this study were as follows: GBP sense (cid:3) GTGCTTCAGAGA-3 (cid:3) ; GBP antisense -TCCAGGGGCCTG-GAATCCGG-3 (cid:3) ; TCCGAA-3 G3PDH

The central nervous system is established through various developmental steps, which include determination of cell fate to the neuronal lineage, proliferation, and differentiation of precursor cells, migration to defined regions, and formation of neuronal interactions. Most neurons derived at the embryonic stage construct a number of synapses during the early postnatal development of the brain, which accompanies programmed cell death of excessively produced neurons in order to form a proper neuronal network (1)(2)(3). A number of molecules such as humoral factors (4,5), receptors (6,7), adhesion molecules (8,9), and transcriptional factors (10,11) have been reported to participate in brain development at various stages. The molecular mechanisms for the formation of structural and functional nervous systems are complex, and our knowledge is limited. Therefore, we consider that many as yet unidentified genes would contribute to the regulation of brain development at the embryonic and postnatal stages.
In the present study, we describe the isolation and characterization of a novel gene enriched in the embryonic and postnatal stages of rat brain development. Our cloned gene encoded an ER 1 -Golgi localized protein that was associated with GRP78, one of the heat shock proteins. We suggest that our cloned GRP78-binding protein (GBP) plays a role in the functional development of neurons.

EXPERIMENTAL PROCEDURES
Molecular Cloning of Rat GBP-The expressions of mRNAs in the hypothalamus, including the suprachiasmatic nucleus, of postnatal rats (2 weeks after birth) and adult rats (10 weeks) were compared by PCR-selected cDNA subtraction according to the manufacturer's instructions (Clontech) (12).
Briefly, poly(A) ϩ RNA (0.5 g) was extracted from the brains at each age, converted to double-stranded cDNA (ds cDNA) by incubation with reverse transcriptase, 5Ј-AAGCAGTGGTAACAACGCAGAGTACGCG-GG-3Ј (10 M), and 5Ј-AAGCAGTGGTAACAACGCAGAGTACN Ϫ1 N-3Ј (10 M). The ds cDNA was digested to blunt-ended fragments by treatment with RsaI, and only ds cDNA from 2 week-rats was ligated with oligonucleotides: 5Ј-CTAATACGACTCACTATAGGGCTCGAGCGGCC-GCCCGGGCAGGT-3Ј (adapter1) or 5Ј-CTAATACGACTCACTATAGG-GCAGCGTGGTCGCGGCCGAGGT-3Ј (adapter2R). For the 1st hybridization, each adapter-ligated cDNA was respectively hybridized with an excessive amount of cDNA from adult rats. Following the 2nd hybridization by combination with each 1st hybridized solution, 2-week-specific cDNAs were amplified by PCR using the 1st primer, 5Ј-CTAATA-CGACTCACTATAGGGC-3Ј, 2nd primer1, 5Ј-TCGAGCGGCCGCCCG-GGCAGGT-3Ј, and 2nd primer2, 5Ј-AGCGTGGTCGCGGCCGAGGT-3Ј. The secondary PCR products were cloned using the T/A cloning system and sequenced on both strands using an ABI PRISM dye terminator cycle sequencing kit (PerkinElmer Life Sciences) with T7 and SP6 primers. To obtain the full-length cDNA of GBP, we produced 5Ј-RACE and 3Ј-RACE PCR products from rat forebrain RNA and sequenced them as described above.
Cell Lines, Cell Culture, and Transfection-Neuro2a and COS7 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, and PC12 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine serum and 5% calf serum. Transfection with each plasmid (2-6 g) was performed using LipofectAMINE Plus (Invitrogen) according to the man-ufacturer's instructions. Stable cell lines transfected with each plasmid were selected and maintained by G418 (Clontech) at concentrations of 0.5-2.0 mg/ml.
Northern Blotting and RT-PCR Analysis-Total RNA (5-10 g) from various rat tissues was purified, resolved by agarose gel electrophoresis under denaturing conditions, and blotted onto nylon membranes. To investigate the expression of GBP mRNA in the various brain regions and in the whole brain at indicated ages, we purchased RNA-blotted membrane from Seegene Co. Ltd. An ␣-32 P-radiolabeled rat GBP cDNA fragment containing nucleotides 2972-3248 was hybridized in the Clontech hybridization solution. After washing, the membranes were exposed to a Fuji imaging plate and analyzed with BAS2000 (Fuji Film, Japan). In order to control for the amount of RNA present, the membranes were stripped and re-hybridized with a G3PDH cDNA probe.
RT-PCR was also performed to estimate the mRNA levels of various tissues and cell lines as described previously (13). Briefly, 0.5 g of total RNA was converted into cDNA by reverse transcriptase, and specific DNAs were amplified by PCR using Taq polymerase. The PCR primers used in this study were as follows: GBP sense primer, 5Ј-GGCTCCCA-GTGCTTCAGAGA-3Ј; GBP antisense primer1, 5Ј-TCCAGGGGCCTG-GAATCCGG-3Ј; GBP antisense primer2, 5Ј-GATCCTCTCCATGTAGT-TCCGAA-3Ј; GBP antisense primer3, 5Ј-TGCAGTTCCACTACACAGG-C-3Ј; G3PDH sense primer, 5Ј-TCCACCACCCTGTTGCTGTA-3Ј; and G3PDH antisense primer, 5Ј-ACCACAGTCCATGCCATCAC-3Ј. The level of GBP mRNA in tissues and cell lines was detected using GBP sense primer and GBP antisense primer1 or -2, and T7 primer and GBP antisense primer3 were used to detect the level of Myc-tagged GBP mRNA in Neuro2a cells. The typical reaction conditions were 0.5 min at 95°C, 0.5 min at 60°C, and 1 min 72°C. The results shown represent 18 -30 cycles of amplification. After amplification, the cDNAs were separated by electrophoresis in 2.0% agarose gels and visualized using ethidium bromide.
In Situ Hybridization-Antisense and sense RNA probes were prepared by in vitro transcription of an RT-PCR-amplified fragment of GBP cDNA (nucleotides 1122-3248) subcloned into pBSSK(ϩ) and synthesized from these cDNA templates with T3 or T7 RNA polymerase and a digoxigenin (DIG)-labeling mixture (Roche Molecular Biochemicals) (14). Postnatal (2-week) and adult (10-week) rat brains were perfused and fixed with 4% paraformaldehyde in 0.1 M phosphate buffer at 4°C, and coronal sections (30 m in thickness) were cut on a cryostat. The sections were treated with 0.1 mg/ml proteinase K (Sigma), 10 mM Tris-HCl buffer (pH 7.4), and 10 mM EDTA for 10 min at 37°C, 4% paraformaldehyde in 0.1 M phosphate buffer for 5 min, and 0.25% acetic anhydride in 0.1 M triethanolamine for 10 min. The sections were then incubated in hybridization buffer containing a DIG-UTP-labeled GBP riboprobe for 12 h at 60°C. Following the hybridization, the sections were washed sequentially in 2ϫ SSC containing 50% formamide, RNase solution, and 0.4ϫ SSC. Finally, they were processed for DIG-coloring steps using an anti-DIG antibody, 4-nitro blue tetrazolium chloride, and 5-bromo-4-chloro-3-indolyl phosphate as described previously (14).
Fluorescence Microscopy-COS7 cells were seeded on glass coverslips and transfected with the indicated plasmid as described above. For the detection of cells expressing Myc-tagged GBP, cells were fixed with PBS containing 4% formaldehyde for 15 min. After washing, the cells were permeabilized with PBS containing 0.2% Triton X-100 for 5 min and incubated with an anti-Myc antibody for 1 h. The cells were then incubated with a fluorescein isothiocyanate-conjugated anti-mouse IgG as the secondary antibody. The ER and Golgi apparatus were visualized by staining with BODIPY 558/568-brefeldin A for 30 min just before fixation with PBS containing 4% formaldehyde (16).
In all cases, the cells were mounted in Vectashield (Vector Laboratories) and observed by fluorescence microscopy (Olympus, Japan).
GST Pull-down Assay-PC12 cells in 10-cm dishes were preincubated with methionine-free medium for 60 min, and then 25 Ci of [ 35 S]methionine was added. After 6 h of incubation, the cells were lysed with lysis buffer (Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 1 mM PMSF, and protease inhibitors) and centrifuged at 15,000 rpm for 10 min. The collected supernatants were rotated with the C-terminal region of GBP-fused GST-or GST-immobilized beads overnight at 4°C (15). After extensive washing with lysis buffer, the proteins bound to each resin were eluted with 50 mM Tris-HCl (pH 8.0) containing 10 mM GSH and separated by 8% SDS-PAGE. After drying, the gels were exposed to a Fuji imaging plate and analyzed with BAS2000. To extract and analyze the proteins bound to GST-fused GBP, SDS-PAGE gels were stained with Sypro Ruby (Molecular Probes) or Coomassie Brilliant Blue, and the stained regions were digested in trypsin solution at 35°C for 20 h for analysis by liquid chromatography/ mass spectrometry/mass spectrometry (Aproscience, Japan).
Western Blotting and Co-immunoprecipitation-COS7 cells in 10-cm dishes were transfected with pcDNA3.1/GBP-myc.his (Invitrogen) and resuspended in lysis buffer (20 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, and protease inhibitors) containing 1% Triton-X100 or Nonidet P-40 for 20 min on ice. After centrifugation at 15,000 rpm, the detergent-soluble fraction was collected. The detergent-resistant fraction was resuspended and sonicated in an equal amount of lysis buffer. An aliquot of each sample was mixed with an equal amount of SDS sample buffer and boiled for 3 min. Western blotting was performed as described previously using the anti-Myc monoclonal antibody 9E10 (1:500) (Santa Cruz Biotechnology).
For co-immunoprecipitation, COS7 cells in 10-cm dishes were transfected with pcDNA3.1/myc.his or pcDNA3.1/GBP-myc.his together with pflag-GRP78. 24 h after the transfection, the cells were dissolved in lysis buffer, and debris was discarded after centrifugation. Anti-Myc or anti-FLAG M2 monoclonal antibodies (Sigma) were added to the whole cell lysates and rotated overnight at 4°C. Then 20 l of protein G-Sepharose beads were added and rotated for a further 1 h. After washing the beads four times with 20 mM Tris-HCl (pH 8.0) containing 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 0.2% Nonidet P-40, and protease inhibitors, the beads were resuspended in SDS sample buffer, boiled, and analyzed by Western blotting. Each protein was detected with the anti-Myc antibody or anti-FLAG antibody, respectively.
Measurement of Cell Viability-Cell viability was determined by the MTT assay as described previously (17). Briefly, each Neuro2a cell line stably transfected with Myc-tagged GBP or empty vector (3 ϫ 10 4 cells/well) was cultured in the presence or absence of 10% fetal bovine serum for the indicated time, and the MTT solution was added to each well and incubated for 4 h. The cells were lysed with 0.04 M HCl in isopropyl alcohol, and the difference in absorbance between 595 and 655 nm was measured as an indicator of cell viability.

Molecular
Cloning of GBP-To identify novel genes in the hypothalamus during postnatal brain development, the mRNA expressions in the forebrains of postnatal day (P) 14, and adult rats at 10 weeks were compared by PCR-selected differential screening because retinal axons grow into the hypothalamus (retinohypothalamic tract) during the postnatal stage, and this reaches a peak at around 2 weeks (18,19). We screened the partial sequence of a novel gene, which was expressed at a higher level in postnatal brain compared with adult brain. To obtain the full-length cDNA of this gene, we produced 5Ј-RACE and 3Ј-RACE PCR products from the rat forebrain RNA, and fused them together. The nucleotides and deduced amino acid sequence of this rat cDNA are shown in Fig. 1. According to a search with BLAST, this sequence was almost 90% identical to a mouse one (GenBank TM accession number BC006896). A related sequence in human (GenBank TM accession number AK023577) was partial and about 80% identical to rat GBP (nucleotides 1077-3066). The cDNA encoded 1021 amino acids with a predicted molecular mass of 110 kDa. A motif search of this protein with PROSITE predicted two transmembrane regions (amino acid residues (aa) 14 -40 and 852-868), a prolinerich region (aa 739 -780), and a glutamic acid-rich region (aa 809 -831). In addition, several N-glycosylation, N-myristoylation, and phosphorylation sites were deduced (data not shown).
Enrichment of GBP mRNA in the Rat Brain-Northern blotting and RT-PCR analysis were performed to examine the distribution of GBP mRNA in tissues. As shown in Fig. 2A, enriched expression of GBP mRNA was observed in the brain compared with other tissues. The expression in the lung was moderate, and weak signals were detected in the heart, liver, kidney, spleen, testis, and muscle. Within the brain, Northern blotting of total RNA isolated from different brain regions showed that GBP was expressed at higher levels in the cerebral cortex, thalamus, and cerebellum and at lower levels in the pons, medulla oblongata, and spinal cord (Fig. 2B).
The temporal changes in the expression of GBP mRNA in the whole brain from the embryonic stage to the adult were exam-ined (Fig. 3). The expression of GBP was faintly detected at E12 and gradually increased until P0. After birth, the GBP level was almost maintained from E20.5 to 2 weeks, and then continuously declined until 12 months. Densitometric analysis of Northern blotting normalized by G3PDH mRNA intensity revealed that the expression of GBP in the 12-month-old rat brain had decreased by about 70% compared with that in postnatal brain.
To elucidate the expression pattern of GBP mRNA in embryonic, postnatal, and adult rat brain brains, we performed in situ hybridization histochemistry. The expression of GBP mRNA was observed predominantly in the gray matter, and the appearance and distribution of the GBP-expressing cells suggested that GBP mRNA was expressed in neurons but not in glial cells. GBP-expressing cells were detected in various regions of the rat forebrain, including the hypothalamus, and cells strongly expressing GBP were observed in the thalamus, cerebral cortex, amygdala, and cerebellum (Fig. 4). In agreement with the results of the Northern blotting, the expression of GBP mRNA in aged rat brain was ubiquitously decreased in comparison with that in postnatal developing brain.
Immunostaining of COS7 Cells Transfected with GBP-Next, we prepared Myc-and GFP-tagged GBP constructs, and we investigated the expression and intracellular localization in intact cells by immunoblotting and fluorescence microscopy. As shown in Fig. 5, Myc-tagged GBP in COS7 cells was almost solubilized by 1% Triton X-100 or Nonidet P-40, and bands corresponding to a 170kDa protein were detected by immunoblotting. COS7 cells transfected with Myc-tagged GBP or GFPtagged GBP showed similar localization of GBP (Fig. 5, A and  C). Furthermore, we confirmed that GFP-tagged GBP in COS7 cells almost overlapped with the fluorescent probe (BODIPY)labeled brefeldin A which specifically stains the ER and Golgi apparatus (Fig. 5C). To characterize further the localization of GBP, we constructed three deletion mutants, consisting of the N-terminal region (aa 1-411), the middle region (aa 375-817), and the C-terminal region (aa 666 -1021) and established stable cell lines expressing full-length GBP or a mutant GBP. As shown in Fig. 6, only the C-terminal region of GBP among the three mutants had a unique cellular localization, and its pattern was similar to that of full-length GBP. Association Analysis between GBP and GRP78 -To investigate the roles of our cloned gene, we used affinity methods to identify its binding proteins. An affinity matrix was prepared by immobilization of a GST-GBP (aa 606 -913) fusion protein on glutathione-Sepharose beads, because the roles of this region were somewhat deduced by the observation of the COS7 cells transfected with the GFP-tagged GBP (aa 666 -1021) plasmid as described above. A band corresponding to an ϳ45-kDa protein was mainly detected by Coomassie Brilliant Blue staining, although this cDNA encoded 248 amino acids with a predicted molecular mass of about 54 kDa (Fig. 7A). We considered that the GST-fused C-terminal region of GBP in Escherichia coli would be digested during the extraction and immobilization to the GSH-Sepharose beads. Incubation of [ 35 S]methionine-labeled PC12 cell lysates with the immobilized GST-GBP-(606 -1022) fusion protein resulted in the binding of several proteins (Fig. 7B). Among these proteins, we identified two proteins with molecular masses of about 70 kDa from non-labeled PC12 cell lysates (see "Experimental Procedures"). Liquid chromatography/mass spectrometry/mass spectrometry analysis of the fragments of the lower and upper bands and comparison with the sequences in the GenBank TM Data Bank revealed almost complete identity to rat Hsc73 and GRP78, respectively.
To investigate the association with Hsc73 or GRP78 in intact cells, COS7 cells were co-transfected with Myc-tagged GBP and FLAG-tagged Hcs73 or GRP78. After 24 h of transfection, the cells were lysed and anti-Myc antibody immunoprecipitates were subjected to SDS-PAGE, transferred to polyvinylidene difluoride membrane, and immunoblotted with an anti-FLAG antibody. Under these conditions, FLAG-tagged GRP78 was identified in the anti-Myc antibody immunoprecipitates (Fig.  7C). Therefore we named this novel gene, GRP78-binding protein (GBP). On the other hand, interaction with Myc-tagged GBP and FLAG-tagged Hsc73 was very weak (data not shown). No signals were detected in immunoprecipitates without the anti-Myc antibody or the lysates from empty vector-transfected cells.
Overexpression of GBP Attenuates Serum Starvation-induced Cell Death-In order to investigate a biological role for GBP, we established four Neuro2a cell lines stably expressing Myctagged GBP or an empty vector. Two overexpressed cell lines markedly up-regulated the level of GBP mRNA, although Neuro2a cells did endogenously express some GBP (Fig. 8A). The band of Myc-tagged GBP in Neuro2a cells corresponded to about 135 kDa by immunoblotting, which was different from that in the COS7 cell lysates (Fig. 8B). Overexpression of GBP did not change the features or growth rate of Neuro2a cells. However, overexpression of GBP attenuated the time-dependent reduction in cell viability that was caused by serum deprivation (Fig. 8C). After 48 h of serum starvation, both Neuro2a cell lines overexpressing GBP were more resistant to the cell death induced by serum starvation (Fig. 8D).

DISCUSSION
In the present study, we isolated and characterized a novel GRP78-binding protein, GBP, in the rat brain. GBP was predicted to have two transmembrane regions, a proline-rich region and a glutamic acid-rich region, but the sequence offered few clues to the possible function. In order to identify regions likely to be functionally important, we observed the intracellular localization of full-length and deletion mutants of this gene in COS7 cells. We then found that GRP78, one of the heat shock proteins, was a GBP-binding protein by the GST pull-down assay method using the identified region of GBP. Our results suggest that the C-terminal region of GBP would have some function in the localization to the ER-Golgi region and interaction with GRP78, but the precise functional domain in GBP remains to be determined. The bands of GBP corresponding to about 135-and 170-kDa proteins in Neuro2a and COS7 lysates, respectively, were detected by immunoblotting although GBP encoded 1021 amino acids with a deduced molecular mass of 110 kDa. We consider that some sites in GBP would be individually modified by as yet unknown and different mechanisms in Neuro2a and COS7 cells because a motif search of GBP by PROSITE suggested that GBP contains several N-glycosylation, N-myristoylation, and phosphorylation sites.
We demonstrated that GBP mRNA was predominantly expressed in the neurons, but not glial cells, of the rat brain, by Northern blotting, and in situ hybridization. GBP-expressing cells were widely detected in various regions of the rat brain such as the cortex, thalamus/hypothalamus, amygdala, and cerebellum. Interestingly, in the cerebellum, granule cells but not Purkinje cells expressed GBP mRNA (data not shown). GBP mRNA was already detected in the E12 brain, and it gradually increased to reach a peak within P0 -2 weeks after birth. These observations indicate that GBP would play a role in the development of the embryonic and postnatal brain and the function of neuronal cells. Furthermore, the age-dependent decrease in GBP expression implies that GBP might affect neuronal susceptibility to various extra-and intracellular stresses to cause several neuronal disorders.
GRP78 was originally identified because its expression is significantly increased in glucose-deprived cells. Similar to other heat shock proteins, such as Hsp40, Hsp70, and GRP98, GRP78 has been shown to be induced by a variety of stimuli, including an increase in intracellular calcium, oxidative stress, and ER stress (20 -22). GRP78 has been reported to exist as an ER lumen protein and to act as a molecular chaperon that regulates protein folding and translocation into the ER and protein secretion, and ER disruption occurred when mutated GRP78 was overexpressed (23)(24)(25)(26). The localization of GBP in the ER-Golgi apparatus and its association with GRP78 in our experiments is consistent with the intracellular localization of GRP78. In the central nervous system, the levels of GRP78 are markedly induced in response to ischemia, axotomy, and kain- ate-induced seizures (27,28). Down-regulation of the GRP78 level by treatment with GRP78 antisense caused sensitivity to a variety insults such as glutamate, Fe 2ϩ , and amyloid ␤-peptide (29). In particular, GRP78 has been suggested to suppress oxyradical accumulation and to stabilize mitochondrial functions in PC12 cells (29), but the precise mechanisms remain to be determined. Very recently, GRP78 has been reported to associate with ATF6 and caspase-7/12, respectively (22,30), and suggested to regulate stress-related gene expression and apoptosis. Both ATF6 and caspases are anchored in the ER by binding to GRP78 under resting conditions, but some ER stresses cause cleavage of both proteins and translocation to the Golgi-nucleus and cytosol, respectively. In our preliminary experiments, the amount of FLAG-tagged GRP78 co-immunoprecipitated with Myc-tagged GBP was hardly affected by treatment with thapsigargin, which causes ER stress by inhibiting ER Ca 2ϩ -ATPase, suggesting that the association of GRP78 and GBP is constitutive. On the other hand, the expression of GBP mRNA in neuronal cell lines was hardly modulated by several stimuli that cause growth arrest and/or apoptosis (data not shown). The significance of GBP in vivo under pathological conditions remains to be determined although GBP might affect ER-Golgi-localized proteins, such as ATF-6 and caspase-7/12, by forming a complex with GRP78.
Furthermore, we found that stable overexpression of GBP in Neuro2a cells suppressed serum deprivation-induced cell death, which is consistent with the anti-apoptotic feature of GRP78. Serum deprivation would disturb a variety of cellular signaling pathways, such as mitogen-activated protein kinases and p53 cascades, to induce apoptosis. Because ATF-6 and caspase-12 are reported to be predominantly activated by ER stress reagents, we consider that GBP would attenuate other serum deprivation-induced cascades among several death signaling pathways in Neuro2a cells.
In conclusion, we cloned a novel GRP78-binding protein (GBP), which is predominantly expressed in the rat brain and possesses an anti-apoptotic property, although it is not clear how overexpression of GBP in Neuro2a cells maintained cell viability after serum deprivation. Apoptotic cell death is accompanied by proliferation and differentiation of neurons and the formation of synapses at embryonic and postnatal stages. Neuronal dysfunction during aging and neuronal diseases is also associated with apoptotic cell death. Thus, further studies on the functions of GBP, including GRP78, would provide new FIG. 7. Analysis of the association between GBP and GRP78. A, the GST-fused C-terminal region of GBP expressed in E. coli was extracted by GSH-conjugated Sepharose beads, resolved by SDS-PAGE, and stained with Coomassie Brilliant Blue. B, the GST-fused C-terminal region of GBP or GST immobilized beads was incubated with [ 35 S]methionine-labeled PC12 lysates as described under "Experimental Procedures." After washing and elution, the proteins bound to each resin were resolved by SDS-PAGE and visualized with BAS2000. C, COS7 cells were co-transfected with different combinations of expression constructs as indicated. Co-immunoprecipitation was performed using the anti-Myc monoclonal antibody as described under "Experimental Procedures." Immunoprecipitates (IP) were analyzed using the anti-Myc or anti-FLAG antibodies, respectively. IB, immunoblot; Ab, antibody.

FIG. 8. Overexpression of GBP attenuates serum starvationinduced cell death in Neuro2a cells.
A, Neuro2a cells stably expressing Myc-tagged GBP (GBP-1 and -2) or empty vector (Con-1 and -2) were established by selection with G418, and the level of Myc-tagged GBP (myc-GBP) and total GBP (GBP) mRNA in each clone was analyzed by RT-PCR. B, the expression of Myc-tagged GBP in Neuro2a cells was detected by immunoblotting. C, Neuro2a cells stably expressing Myc-tagged GBP (triangles) or empty vector (circles) were cultured with (filled symbols) or without 10% serum (open symbols) for the indicated times, and then the cell viability was measured using the MTT assay as described under "Experimental Procedures." Each value represents the percentage of the control value at day 0 (D0). D, 2 days after serum starvation, the cell viability of each clone was measured by the MTT assay. Each value represents the mean Ϯ S.D. of four independent cultures and is expressed as the percentage of the cell viability of each clone cultured in the presence of serum for 2 days. *, the difference from cells expressing the empty vectors (Con-1 and -2) was statistically significant by analysis of variance (p Ͻ 0.01, (Fisher's protected least significant difference)).