Enhancement of Transport-dependent Decarboxylation of Phosphatidylserine by S100B Protein in Permeabilized Chinese Hamster Ovary Cells*

Phosphatidylethanolamine synthesis through the phosphatidylserine (PtdSer) decarboxylation pathway requires PtdSer transport from the endoplasmic reticulum or mitochondrial-associated membrane to the mitochondrial inner membrane in mammalian cells. The transport-dependent PtdSer decarboxylation in permeabilized Chinese hamster ovary (CHO) cells was enhanced by cytosolic factors from bovine brain. A cytosolic protein factor exhibiting this enhancing activity was purified, and its amino acid sequence was partially determined. The sequence was identical to part of the amino acid sequence of an EF-hand type calcium-binding protein, S100B. A His6-tagged recombinant CHO S100B protein was able to remarkably enhance the transport-dependent PtdSer decarboxylation in permeabilized CHO cells. Under the standard assay conditions for PtdSer decarboxylase, the recombinant S100B protein did not stimulate PtdSer decarboxylase activity and exhibited no PtdSer decarboxylase activity. These results implicated the S100B protein in the transport of PtdSer to the mitochondrial inner membrane.

synthesis through the decarboxylation of nascent phosphatidylserine (PtdSer) requires both interorganelle and intramitochondrial transport of PtdSer, because PtdSer synthase is located on the endoplasmic reticulum and mitochondrialassociated membrane (MAM) (2)(3)(4), while PtdSer decarboxylase is located on the outer face of the mitochondrial inner membrane (5,6). Using the decarboxylation of nascent PtdSer as an index of PtdSer transport, Voelker (7)(8)(9)(10)(11) has extensively studied the processes of PtdSer transport from the ER or MAM to the mitochondrial inner membrane in intact cells and a permeabilized cell system and has established the following: 1) The PtdSer transport to mitochondria requires ATP (7)(8)(9). 2) The PtdSer transport to mitochondria can occur in the absence of its synthesis (9). 3) The general features of protein and PtdSer export of the ER are fundamentally different insofar as the export of proteins, but not that of PtdSer, requires cytosolic factors and guanine nucleotides (9). 4) Adriamycin, which is a potent inhibitor of the import of proteins into mitochondria, also inhibits the PtdSer transport between the outer and inner mitochondrial membranes (10), and 5). When permeabilized cells are disrupted by shearing, the PtdSer transport from the ER or MAM to mitochondria is restricted to the autologous organelle (11). In addition, studies by Vance and co-workers (12) have suggested that the PtdSer synthesized in the ER traverses the MAM en route to the mitochondria.
Although studies of the PtdSer transport from the ER or MAM to the mitochondrial inner membrane have revealed the general features of this transport process as described above, our knowledge about the genes or gene products specifically involved in this transport is very limited. Mitochondrial membrane protein(s) have been shown to be required for the transport of PtdSer from the MAM to the mitochondria (13), but such protein(s) remain to be identified. In this laboratory, a mammalian cell mutant defective in the intramitochondrial transport of PtdSer has been isolated (14); however, the mutant gene responsible for the defect has not yet been cloned. To further understand the mechanisms of PtdSer transport to mitochondria, we re-examined the effect of cytosol on the transport-dependent PtdSer decarboxylation in permeabilized CHO-K1 cells and found that cytosolic factors from bovine brain enhanced the transport-dependent PtdSer decarboxylation in permeabilized CHO-K1 cells in the presence of Ca 2ϩ . Here, we describe the purification and identification of one of the factors exhibiting this enhancing activity.

* This work was supported in part by the Human Sciences Basic Research Project and Integrated Study Projects on Drug Innovation
Science of the Japan Health Sciences Foundation, by Special Coordination Funds for Promoting Science and Technology from the Science and Technology Agency of Japan, by grants-in-aid for general scientific research from the Ministry of Education, Science and Culture of Japan, and by the ONO Medical Research Foundation. 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.
Preparation of Permeabilized CHO-K1 Cells-The permeabilization of CHO-K1 cells was performed by the method of Voelker (9), with slight modifications. Cells were seeded at 1.1 ϫ 10 7 cells per 150-mm dish 20 -24 h prior to permeabilization. For permeabilization, the cells were washed three times with solution A (140 mM KCl, 10 mM NaCl, 2.5 mM MgCl 2 , 20 mM Hepes, pH 7.6, and 0.1 M CaCl 2 ), overlaid with 14 ml of solution A supplemented with 50 g/ml of saponin (Sigma) and then incubated at 37°C for 5 min. After this treatment, the medium was aspirated off, and the cells were placed on ice and harvested in 13 ml of solution A by scraping with a rubber policeman. The harvested cells were collected by centrifugation at 400 ϫ g for 5 min, washed twice by gentle suspension in 14 ml of solution A followed by centrifugation at 400 ϫ g for 5 min, and then gently suspended in solution A at ϳ2 ϫ 10 7 permeabilized cells per ml.
PtdSer and PtdEtn Synthesis in Permeabilized CHO-K1 Cells-Unless stated otherwise, the reactions for measuring PtdSer and PtdEtn synthesis were conducted in 100 l of solution A in the presence of permeabilized cells (ϳ5 ϫ 10 5 cells), 0.5 mM CaCl 2 , 2 mM ATP, 5 mM phosphocreatine, 1 unit of creatine phosphokinase, and 0.2 Ci of L-[U- 14 C]serine (160 Ci/mol) (Amersham Pharmacia Biotech). The reactions were performed in glass tubes at 37°C and terminated by the addition of 1.5 ml of MeOH/ CHCl 3 (2:1). The lipids were extracted from the reaction mixtures by the method of Bligh and Dyer (15) and separated on a thin layer chromatography plate (Silica Gel 60, Merck) with the solvent system of chloroform/methyl acetate/1-propanol/MeOH/ 0.25% KCl in water, 50:50:50:20:18. The radioactivity of the separated lipids was analyzed with a BAS2000 image analyzer (Fuji Film).
Purification Procedure-Fresh bovine brains were obtained from a local slaughterhouse. All procedures were carried out at 0 -4°C. To measure activity, aliquots of bovine brain cytosol and purification fractions were dialyzed against solution A. The brains were cut into small pieces and then homogenized with a Polytron homogenizer in five volumes of homogenizing buffer (50 mM Hepes/NaOH, pH 7.5, 0.25 M sucrose, and 5 mM EDTA). The homogenate was centrifuged at 1,000 ϫ g for 10 min and the supernatant was centrifuged at 18,000 ϫ g for 15 min. The resulting supernatant was centrifuged at 100,000 ϫ g for 1 h to obtain the soluble fraction (cytosol). The cytosol was dialyzed three times against 12 liters of 20 mM Hepes/NaOH (pH 7.5) and then loaded onto a ϳ70 ml Q-Sepharose HP (Amersham Pharmacia Biotech) column (2.6 cm i.d.) equilibrated with 20 mM Hepes/NaOH (pH 7.5). The column was washed with 250 ml of 20 mM Hepes/NaOH (pH 7.5) and then eluted with an 850-ml linear gradient, 0 -500 mM, of NaCl in 20 mM Hepes/NaOH (pH 7.5). The active fractions eluted with ϳ400 mM NaCl were pooled, adjusted to 2 M (NH 4 ) 2 SO 4 and 50 mM bis tris propane/ NaOH (pH 6.8) by adding solid (NH 4 ) 2 SO 4 and a 0.5 M solution of bis tris propane/NaOH (pH 6.8), and then centrifuged at 10,000 ϫ g for 30 min. The resulting supernatant was loaded onto an 8 ml-phenyl-Sepharose HP (Amersham Pharmacia Biotech) column (1 cm i.d.) equilibrated with 50 mM bis tris propane/NaOH (pH 6.8) containing 2 M (NH 4 ) 2 SO 4 . The column was washed with 20 ml of 50 mM bis tris propane/NaOH (pH 6.8) containing 2 M (NH 4 ) 2 SO 4 and then eluted with a 180-ml linear descending gradient, 2 to 0 M, of (NH 4 ) 2 SO 4 in 50 mM bis tris propane/ NaOH (pH 6.8). The active fractions were pooled, concentrated to 400 l with a collodion bag, loaded onto a Superdex 75 HR (10/30) column (Amersham Pharmacia Biotech) equilibrated with 25 mM Hepes/NaOH (pH 7.5) containing 150 mM NaCl, and then eluted with the same solution. The activity was eluted as a single peak corresponding to a molecular mass of ϳ20 kDa. The active fractions from the Superdex 75 HR column were pooled and directly loaded onto a Mono Q HR (5/5) column (Amersham Pharmacia Biotech) equilibrated with 20 mM Hepes/NaOH (pH 7.5) containing 150 mM NaCl. The column was washed with 3 ml of 20 mM Hepes/NaOH (pH 7.5) containing 150 mM NaCl, and eluted with 3 ml of 20 mM Hepes/NaOH (pH 7.5) containing 300 mM NaCl and a 15-ml linear gradient, 300 to 450 mM, of NaCl in 20 mM Hepes/NaOH (pH 7.5).
Amino Acid Sequence Analysis-The purified protein was cleaved with BrCN (4 mg/ml in 70% formic acid) for 18 h at room temperature, and the resulting peptides were separated by reversed-phase HPLC on a Devolosil ODS column (0.3 ϫ 10 cm; Nomura Chem.) in 0.1% trifluoroacetic acid with an acetonitrile gradient. The amino acid sequences of peptides were determined with a Model 477 amino acid sequencer from Applied Biosystems.
Isolation of CHO S100B cDNA Clones-Poly (A) ϩ RNAs were prepared from CHO-K1 cells with a FastTrac 2.0 mRNA isolation kit (Invitrogen) and used for reverse transcription-PCR. First-strand cDNA was synthesized with a SMART™ rapid amplification of cDNA ends (RACE) cDNA amplification kit (CLONTECH) and PowerScript™ reverse transcriptase (CLONTECH), according to the manufacturer's in-structions. Using the resulting first-strand cDNA as a template, a partial CHO S100B cDNA was amplified by two-stage nested PCR with S100B-specific primers, which were designed based on human, rat, and mouse S100B cDNA sequences, and AmpliTaq Gold DNA polymerase (Perkin-Elmer), according to the manufacturer's instructions. The primers used for the first round of amplification were ATGTCTGAGCT-GGAGAAGG (sense) and TCACTCATGTTCAAAGAACTC (antisense). For the second round of amplification, ATGTCTGAGCTGGAGAAG-GCCA (sense) and TCACTCATGTTCAAAGAACTCATG (antisense) were used as primers. The resulting PCR product was cloned into a plasmid, pCR ® II-TOPO (Invitrogen) and then subjected to DNA sequencing. The sequence determined was used to design PCR primers for 5Ј-RACE. 5Ј-RACE was performed using a SMART RACE cDNA amplification kit with an antisense primer, CGAAGGCCATAAATTCCTG-GAAGTCACAC, and a nested antisense primer, TCTCCATCTC-CATCTTCATCCAG, according to the manufacturer's instructions. The resulting 5Ј-RACE product was cloned into pCR ® II-TOPO and then subjected to DNA sequencing. The sequence determined was used to design PCR primers for 3Ј-RACE. 3Ј-RACE was performed using a SMART™ RACE cDNA amplification kit with a sense primer, GGA-CACTGAAGTCAGAGAGGACACCAGC, and a nested sense primer, TGACCAGGAGCCTCCAGG, according to manufacturer's instructions. Because the nested sense primer used for 3Ј-RACE corresponded to the sequence upstream of an initiation codon, the resulting 3Ј-RACE product contained a full-length open reading frame encoding CHO S100B protein. The 3Ј-RACE product was cloned into pCR ® II-TOPO and then subjected to DNA sequencing. pCR ® II-TOPO carrying the 3Ј-RACE product was named pCR ® II-TOPO-cS100B.
Production of a His 6 -tagged Recombinant CHO S100B Protein in Escherichia coli-Using pCR ® II-TOPO-cS100B as a template, the coding region of CHO S100B cDNA was engineered by PCR so as to add a BamHI site immediately upstream of the initiation codon and to add a HindIII site immediately downstream of the termination codon. The PCR product was digested with BamHI and HindIII and then ligated with the pQE9 vector (Qiagen) digested with the same enzymes. The resulting plasmid was named pQE9-cS100B. pQE9-cS100B and an empty pQE9 vector were introduced into E. coli cells (M15 harboring plasmid pREP4; Qiagen). The resulting transformants, named M15/ pQE9-cS100B and M15/pQE9, respectively, were grown to a density of A 600 ϭ 1.0 in Luria-Bertani medium supplemented with 100 g/ml of ampicillin and 25 g/ml of kanamycin at 30°C, and then further cultivated for 2 h in the presence of 1 mM isopropyl-thio-␤-D-galactoside at 30°C. The cells were harvested and resuspended in solution A. After freezing and thawing, the cells were disrupted by sonication. The cell lysate was centrifuged twice at 100,000 ϫ g for 30 min, and the resulting supernatant was dialyzed against solution A and then subjected to biochemical characterization.
Purification of the His 6 -tagged CHO S100B Protein-After inducing the production of a His 6 -tagged CHO S100B protein as described above, M15/pQE9-cS100B cells were suspended in solution B (150 mM NaCl and 25 mM Hepes/NaOH, pH 7.5), subjected to freezing and thawing, and then disrupted by sonication. The lysate was clarified by centrifugation at 10,000 ϫ g for 30 min, and the supernatant was loaded onto a 1-ml HisTrap™ column (Amersham Pharmacia Biotech) equilibrated with solution C (500 mM NaCl and 25 mM Hepes/NaOH, pH 7.5) containing 10 mM imidazole. After washing with this solution, the His 6tagged CHO S100B protein was eluted with solution C containing 500 mM imidazole, concentrated with a collodion bag, and then loaded onto a Superdex 75 HR (10/30) column equilibrated with solution B. After elution with solution B, the purified His 6 -tagged CHO S100B protein was dialyzed against solution A and then subjected to biochemical characterization.
Preparation of a Membrane Fraction of CHO-K1 Cells-Exponentially growing CHO-K1 cells were washed with phosphate-buffered saline, suspended in 250 mM sucrose containing 10 mM Hepes/NaOH (pH 7.5) and 1 mM EDTA, and then homogenized with a Potter-Elvehjem Teflon homogenizer. The homogenate was centrifuged at 700 ϫ g for 5 min, followed by centrifugation of the supernatant at 100,000 ϫ g for 1 h. The resulting pellet was suspended in solution A and then centrifuged again at 100,000 ϫ g for 1 h. The pellet was resuspended in solution A and subjected to biochemical characterization.
Assay for PtdSer Decarboxylase Activity-Exponentially growing CHO-K1 cells were washed with phosphate-buffered saline, suspended in 250 mM sucrose containing 10 mM Hepes/NaOH (pH 7.5), disrupted by sonication, and then used as the enzyme source. The reaction mixture, with a final volume of 100 l, for measuring PtdSer decarboxylase activity comprised 0.03 Ci of [ 14 C]PtdSer (Amersham Pharmacia Bio-tech), 0.5 mM PtdSer from bovine brain (Sigma), 0.5 mg/ml of Triton X-100, 100 mM KH 2 PO 4 -K 2 HPO 4 buffer (pH 7.0), and 10 mM EDTA or 0.5 mM CaCl 2 . The reactions were performed in glass tubes at 37°C for 30 min and terminated by the addition of 1.5 ml of MeOH/CHCl 3 (2:1). The lipids were extracted from the reaction mixtures by the method of Bligh and Dyer (15) and separated on a thin layer chromatography plate (Silica Gel 60, Merck) using the solvent system described above. The radioactivity of the separated lipids was analyzed with a BAS2000 image analyzer (Fuji Film).
Other Methods-Proteins were measured with a BCA protein assay kit (Pierce), with bovine serum albumin as a standard. Tricine-SDS-PAGE was performed as described (16). Silver staining of proteins in polyacrylamide gels was performed with a silver staining kit (Amersham Pharmacia Biotech). DNA sequencing was performed by automated sequencing with an Applied Biosystem Prism 310 genetic analyzer and fluorescence-tagged dye terminator cycle sequencing.

Bovine Brain Cytosol Enhances the Transport-dependent Decarboxylation of PtdSer in Permeabilized CHO-K1
Cells-We prepared saponin-permeabilized CHO-K1 cells according to the method of Voelker (9) and examined its ability to synthesize PtdSer and PtdEtn from serine precursor in the presence or absence of bovine brain cytosol. When the permeabilized cells were incubated with [ 14 C]serine at 37°C in solution A containing 0.2 mM ATP and 0.5 mM CaCl 2 , [ 14 C]PtdSer and [ 14 C]PtdEtn were produced in a time-dependent manner in the absence of bovine brain cytosol (Fig.  1), in agreement with Voelker's conclusion (9) that the synthesis of [ 14 C]PtdEtn in permeabilized cells, which is dependent on the transport of PtdSer from the ER or MAM to the mitochondrial inner membrane, does not require cytosol. The addition of bovine brain cytosol, however, remarkably enhanced the synthesis of [ 14 C]PtdEtn in permeabilized CHO-K1 cells, although the synthesis of [ 14 C]PtdSer in the presence of bovine brain cytosol was similar to that in its absence (Fig. 1). These results indicated that the bovine brain cytosol contained factor(s) that enhanced the transport-dependent decarboxylation of PtdSer in permeabilized CHO-K1 cells.
Purification of a Bovine Brain Cytosolic Factor That Enhances the Transport-dependent Decarboxylation of PtdSer-We fractionated bovine brain cytosol on a Q-Sepharose HP ion exchange column and subjected an aliquot of each fraction to assaying for transport-dependent PtdSer decarboxylation (TPSD)-enhancing activity. ]PtdSer) was used as an index of the TPSD-enhancing activity. The TPSD-enhancing activity was separated into two prominent peaks on Q-Sepharose HP column chromatography ( Fig. 2A). In this study, the TPSD-enhancing factor in the peak II fraction ( Fig.  2A) was further purified by successive chromatography on phenyl-Sepharose hydrophobic (Fig. 2B), Superdex 75 gel filtration (Fig. 2C), and Mono Q ion exchange (Fig. 2D) columns. The overall purification of the TPSD-enhancing activity from the cytosol (700 mg of proteins) was 300-fold, with a recovery yield of 20%, where a 1% increase in the ratio of [ 14 C]PtdEtn/([ 14 C]PtdEtn plus [ 14 C]PtdSer) was defined as 1 unit of activity. With the third gel filtration column, the TPSD-enhancing activity was eluted as a single peak corresponding to a molecular mass of ϳ20 kDa. When proteins in the fractions separated by the final Mono Q chromatography were analyzed by tricine-SDS-PAGE, an ϳ9 kDa protein was found to comigrate with the TPSD-enhancing activity (See Figs. 2D and 3.). These results suggested that a homo dimer of the purified ϳ9 kDa protein exhibited the TPSD-enhancing activity.
Identification of the Purified ϳ9-kDa Protein as a S100B Protein-To determine its amino acid sequence, the purified ϳ9 kDa protein was subjected to cyanogen bromide cleavage, and then the N-terminal amino acid sequences of the resulting two fragments were determined after purification of the fragments by reverse phase HPLC column chromatography. A search involving a protein sequence data bank indicated that the sequences determined, ITTAXHEFFEH and ETLDSDGDGEX-DFG, matched parts of the amino acid sequence of an EF-hand type calcium-binding protein, S100B. Thus, the purified ϳ9 kDa protein was identified as a S100B protein.
A His 6 -tagged Recombinant CHO S100B Protein Exhibits TPSD-enhancing Activity-We isolated CHO S100B cDNA clones. The nucleotide and predicted amino acid sequences of CHO S100B cDNA are shown in Fig. 4. A comparison of the predicted amino acid sequence of the CHO S100B protein with that of the S100B proteins of other species is also shown in Fig.  4. To confirm that the S100B protein exhibits the TPSD-enhancing activity, a His 6 -tagged recombinant CHO S100B (His-cS100B) protein was bacterially produced and subjected to assaying for the TPSD-enhancing activity. Fig. 5 shows the tricine-SDS-PAGE patterns of soluble proteins from E. coli transformants, designated as M15/pQE9 and M15/pQE9-cS100B, which harbored an empty vector plasmid and a His-cS100B protein-expressing plasmid, respectively. A protein of 12 kDa was produced in an expression plasmid-dependent manner (Fig. 5), indicating that the 12 kDa protein was a His-cS100B protein. From a lysate of M15/pQE9-cS100B cells, the His-cS100B protein was purified on HisTrap and Superdex 75 columns successively. The tricine-SDS-PAGE pattern of the purified His-cS100B protein is shown in Fig. 5. Although the purified His-cS100B protein had no significant effect on the synthesis of [ 14 C]PtdSer from [ 14 C]serine in permeabilized CHO-K1 cells (Fig. 6), the purified His-cS100B protein remarkably enhanced the synthesis of [ 14 C]PtdEtn from [ 14 C]serine (Fig. 6). This enhancement by the His-cS100B protein occurred in a saturable manner, ϳ3-fold enhancement being the maximal level (Fig. 6A). Like the purified His-cS100B protein, crude  (160 g of protein). The reactions were terminated at the indicated times by lipid extraction. PtdEtn and PtdSer were separated by thin layer chromatography, and the radioactivity associated with PtdEtn (A) or PtdSer (B) was analyzed with an image analyzer. The results are expressed as relative radioactivity; the radioactivity associated with PtdSer upon a 60-min incubation in the presence of cytosol was taken as 1000. Values are the averages for duplicate assays with variation of Ͻ15% between duplicates.
soluble proteins from M15/pQE9-cS100B cells also remarkably enhanced the synthesis of [ 14 C]PtdEtn from [ 14 C]serine in permeabilized CHO-K1 cells (Fig. 6). In contrast, crude soluble proteins from M15/pQE9 cells, which carried a control empty vector, were incapable of significantly enhancing the synthesis of [ 14 C]PtdEtn from [ 14 C]serine (Fig. 6). These results indicated that the His-cS100B protein was able to enhance the transportdependent decarboxylation of PtdSer in permeabilized CHO-K1 cells.
Effects of Ca 2ϩ Concentration and ATP on the Enhancement of Transport-dependent Decarboxylation of PtdSer by the His-cS100B Protein-The assay for the TPSD-enhancing activity described above was performed in a solution containing ATP and 0.5 mM CaCl 2 . This CaCl 2 concentration is 10 3 -10 4 times higher than the physiological level. To determine whether or not the enhancement of the transport-dependent decarboxylation of PtdSer by the S100B protein  (Fig. 7). The addition of 2 mM ATP, 0.5 mM CaCl 2 , or both 2 mM ATP and 0.5 mM CaCl 2 to solution A led to efficient synthesis of [ 14 C]PtdSer and [ 14 C]PtdEtn (Fig. 7). These results were consistent with the previous conclusion (9) that the synthesis of PtdSer and PtdEtn from serine in permeabilized CHO-K1 cells require ATP or a high Ca 2ϩ level. The addition of the His-cS100B protein to solution A supplemented with 2 mM ATP, 0.5 mM CaCl 2 , or both 2 mM ATP and 0.5 mM CaCl 2 , respectively, led to ϳ2-, ϳ3-, and ϳ3-fold enhancement of the synthesis of [ 14 C]PtdEtn in permeabilized CHO-K1 cells (Fig. 7). In contrast to the [ 14 C]PtdEtn synthesis, the [ 14 C]PtdSer synthesis remained essentially unchanged upon the addition of the His-cS100B protein to these solutions. These results indicated that the enhancement of the transport-dependent decarboxylation of PtdSer by the His-cS100B protein occurred in a physiological salt solution supplemented with ATP and did not require ATP upon supplementation with a high level of Ca 2ϩ .

Effects of the His-cS100B Protein on the Transport-dependent Decarboxylation of PtdSer by Isolated Membranes and on Ptd-
Ser Decarboxylase Activity-To examine whether or not the transport-dependent decarboxylation of PtdSer by isolated organelle membranes as well as that by permeabilized cells is enhanced by the His-cS100B protein, we prepared a membrane fraction containing ER membrane, MAM and mitochondria by centrifugation of a postnuclear supernatant of CHO-K1 cells at 100,000 ϫ g. When the isolated membranes were incubated at 37°C in solution A supplemented with [ 14 C]serine, ATP, and 0.5 mM CaCl 2 , [ 14 C]PtdSer and [ 14 C]PtdEtn were efficiently produced (Fig. 8), suggesting that the transport of nascent PtdSer to mitochondria had occurred. The addition of the His-cS100B protein to the incubation mixture led to 1.3-fold enhancement of [ 14 C]PtdEtn formation. However, this enhance- PtdEtn and PtdSer were separated by thin layer chromatography, and the radioactivity associated with PtdEtn (A) or PtdSer (B) was analyzed with an image analyzer. The results are expressed as relative radioactivity; the radioactivity associated with PtdSer upon 60 min incubation in the presence of 1 g of purified His-cS100B protein was taken as 1000. Values are the averages for duplicate assays with variation of Ͻ15% between duplicates. C and D, permeabilized CHO-K1 cells were labeled with [ 14 C]serine as described under "Experimental Procedures" in the presence (closed circles) or absence (open circles) of the purified His-cS100B protein (2.5 g). The reactions were terminated at the indicated times by lipid extraction. PtdEtn and PtdSer were separated by thin layer chromatography, and the radioactivity associated with PtdEtn (C) or PtdSer (D) was analyzed with an image analyzer. The results are expressed as relative radioactivity; the radioactivity associated with PtdSer upon 60 min incubation in the presence of purified His-cS100B was taken as 1000. Values are the averages for duplicate assays with variation of Ͻ15% between duplicates. ment was much lower than that of PtdEtn formation in permeabilized cells by the His-cS100B protein (Fig. 6). Therefore, the enhancement of transport-dependent decarboxylation of PtdSer by S100B protein in permeabilized cells seemed to be related to the preserved organelle structures in permeabilized cells and was probably not because of the stimulation of PtdSer decarboxylase.
Next, we examined the effect of the His-cS100B protein on the PtdSer decarboxylase activity in a homogenate of CHO-K1 cells, which was measured under the standard assay conditions. Under such conditions, the His-cS100B protein did not stimulate PtdSer decarboxylase activity and did not exhibit PtdSer decarboxylase activity (Table I). Because the S100B protein is a Ca 2ϩ -binding protein, we also examined the effect of the His-cS100B protein on PtdSer decarboxylase activity in the presence of Ca 2ϩ (0.5 mM). Ca 2ϩ itself did not affect PtdSer decarboxylase activity (Table I). Even in the presence of 0.5 mM Ca 2ϩ , the His-cS100B protein did not stimulate PtdSer decarboxylase activity and did not exhibit PtdSer decarboxylase activity. These results also supported the idea that the enhancement of the transport-dependent decarboxylation of PtdSer by the His-cS100B protein in permeabilized CHO-K1 cells was not because of the stimulation of PtdSer decarboxylase.

DISCUSSION
PtdEtn synthesis through the PtdSer decarboxylation pathway is required for maintenance of a normal PtdEtn level in mammalian cells (14,(17)(18)(19) and normal cell growth (20), and thus probably for the biogenesis of functional organelle membranes. The PtdEtn synthesis via the decarboxylation of nascent PtdSer requires the transport of PtdSer from the ER or MAM to the mitochondrial inner membrane, because of the difference in the subcellular localization of PtdSer synthase and PtdSer decarboxylase (2)(3)(4)(5)(6). Although enzymes involved in the PtdSer decarboxylation pathway, PtdSer synthase and PtdSer decarboxylase, have been identified through the isolation of cDNA clones (21)(22)(23), the mammalian genes and gene products involved in the transport of PtdSer from the ER or MAM to the mitochondrial inner membrane have not been identified so far. In this study, a Ca 2ϩ -binding protein, S100B, was shown to enhance the transport-dependent decarboxylation of PtdSer in permeabilized CHO-K1 cells. This enhancement could be possibly because of the enhancement of a certain process of PtdSer transport or to the stimulation of PtdSer decarboxylase activity. However, the latter possibility was unlikely, because the His-cS100B protein did not stimulate PtdSer decarboxylase activity under the standard assay conditions, irrespective of the presence or absence of Ca 2ϩ , and the enhancement of decarboxylation of PtdSer by the His-cS100B protein observed in isolated membranes was much lower than that in permeabilized cells. Thus, the S100B protein was implicated in the transport of PtdSer from ER or MAM to mitochondrial inner membrane.
The S100B protein is a member of a family of small (ϳ10 The results are expressed as relative radioactivity; the radioactivity associated with PtdSer upon incubation in solution A containing 2 mM ATP, an ATP regenerating system, and 0.5 mM CaCl 2 , and in the absence of the His-cS100B protein was taken as 1000. Values are the averages for duplicate assays with variation of Ͻ15% between duplicates.

FIG. 8. Effect of the His-cS100B protein on the synthesis of PtdEtn and PtdSer by a isolated membrane fraction of CHO-K1 cells.
A membrane fraction of CHO-K1 cells was prepared as described under "Experimental Procedures." The isolated membranes (28 g of protein) were incubated in 100 l of solution A containing [ 14 C]serine (0.2 Ci, 160 Ci/mol), 2 mM ATP, an ATP regenerating system, and 0.5 mM CaCl 2 , in the presence (closed circles) or absence (open circles) of 4 g of the His-cS100B protein. The reactions were terminated at the indicated times by lipid extraction. PtdEtn and PtdSer were separated by thin layer chromatography, and the radioactivity associated with PtdEtn (A) or PtdSer (B) was analyzed with an image analyzer. The results are expressed as relative radioactivity; the radioactivity associated with PtdSer upon incubation for 60 min in the absence of the His-cS100B protein was taken as 1000. Values are the averages for duplicate assays with variation of Ͻ15% between duplicates. kDa) Ca 2ϩ -binding proteins of the EF-hand type known as the S100 family, which comprises 19 members (24 -26). The S100B protein exists as a homodimer or a heterodimer with the S100A1 protein (27,28), which is another S100 family member and is related most closely to the S100B protein. The S100B protein has been shown to interact with various proteins, and has been implicated in the Ca 2ϩ -dependent regulation of a variety of intracellular activities such as protein phosphorylation, enzyme activities, cell proliferation, cytoskelton assembly, neurite outgrowth, and intracellular Ca 2ϩ homeostasis (24 -26). Interestingly, Ca 2ϩ binding to the S100B protein dimer induces a conformational change, which results in exposure of the binding surface of an individual monomer, with which target proteins of the S100B protein are believed to interact (26). Thus, the S100B protein dimer appears to expose two binding surfaces and to be capable of cross-bridging two proteins, in a Ca 2ϩ -dependent manner. Inconsistent with our finding that cytosol enhances the transport-dependent decarboxylation of PtdSer in permeabilized cells, Voelker (9) has reported that cytosol slightly inhibits it. This inconsistency may be attributed to the differences in the assay conditions. In the experiment performed by Voelker, the effect of cytosol was examined in the presence of EGTA, which chelates Ca 2ϩ and arrests PtdSer synthesis. On the other hand, our assay was performed in a solution containing Ca 2ϩ and in the presence of PtdSer synthesis. Thus, the enhancement of the transport-dependent decarboxylation of PtdSer by cytosol might require free Ca 2ϩ and/or the presence of PtdSer synthesis.
Bovine brain cytosol appears to contain several factors that enhance the transport-dependent decarboxylation of PtdSer in permeabilized cells, because the enhancing activity has been separated into several peaks on Q-Sepharose ion exchange column chromatography. We have also tried to purify factors other than the S100B protein and found that, in addition to the S100B protein, at least two factors, which are larger than 200 kDa and smaller than 10 kDa, respectively, can enhance the transport-dependent decarboxylation of PtdSer in permeabilized cells. 2 It is therefore likely that several cytosolic factors are involved in the transport or decarboxylation of PtdSer.
Studies by Voelker (9,11) have suggested that a diffusible PtdSer transport intermediate, such as a soluble protein or a transport vesicle carrying PtdSer, is not involved in the import of PtdSer into mitochondria. Importantly, studies involving CHO-K1 cells disrupted by saponin-permeabilization followed by shearing have demonstrated that PtdSer produced in one population of disrupted cells cannot be transported to the mitochondria of a second population of disrupted cells (11). Therefore, PtdSer transport from the ER or MAM to mitochondria is physically restricted. This restriction implies that PtdSer is imported into mitochondria via a tightly associated contact site between ER or MAM and mitochondria. Then, how can the cytosolic S100B protein enhance the PtdSer transport to the inner mitochondrial membrane? One possible explanation is that the S100B protein regulates a putative PtdSer transport machinery existing in the ER membrane, MAM, or outer mitochondrial membrane. Another possible explanation is that the S100B protein stabilizes the contact site or increases the number of contact sites between the ER or MAM and mitochondria through crossbridging membrane proteins of these organelles. Identification of S100B-interacting membrane protein(s) that exist in these organelles might provide new insights into the transport of PtdSer from the ER or MAM to mitochondria.