High-performance affinity purification foridentification of 15-deoxy-D12,14-PGJ2

interacting factors using magnetic nanobeadsNaoya Maekawa,a Masaki Hiramoto,a,b Satoshi Sakamoto,c Motoki Azuma,c

Takumi Ito,c Marie Ikeda,a,d Michitaka Naitou,d Hukum P. Acharya,c

Yuichi Kobayashi,c Makoto Suematsu,e Hiroshi Handac and Takeshi Imaia,e,f*

ABSTRACT: Prostaglandin J2 (PGJ2) family have been reported to show various kinds of biological activities. Considerableprogress has been made toward understanding the mechanism of adipogenesis, however, the mechanisms of other actions ofPGJ2 family remain controversial. The 15-deoxy-D12,14-PGJ2 (15d-PGJ2) is one of the members of PGJ2 family, and is known as aligand for peroxisome proliferator-activated receptor g (PPARg), which promotes the expression of the crucial genes for adi-pogenesis. In this study, we found that 15d-PGJ2 did not stimulate PPARg-mediated gene expression in HEK293 cells whereas15d-PGJ2 transactivated PPARg-dependent transcription in other cell lines. Moreover, we confirmed that 15d-PGJ2 suppressedthe growth of HEK293 cells. These observations suggest that 15d-PGJ2 shows another biological activity e.g. growth inhibitionin HEK293 cells via unknown receptor for 15d-PGJ2.

The aim of this study is to develop and validate effective purification system for PGJ2 interacting factors (PGJIFs). We haverecently developed high performance magnetic nanobeads. In this study, we have newly developed 15d-PGJ2-immobilizedbeads by conjugating 15d-PGJ2 to the surface of these nanobeads. Firstly, we showed that PPARg specifically bound to 15d-PGJ2-immobilized beads. Secondly, we newly identified voltage dependent anionic channel 1 (VDAC1) as new PGJIF from crudeextracts of HEK293 cells using this affinity purification system. These data presented here demonstrate that 15d-PGJ2-immobilized beads are effective tool for purification of PGJIFs directly from crude cell extracts. Copyright © 2010 John Wiley &Sons, Ltd.

Keywords: prostaglandin J2; affinity purification; magnetic nanobeads; prostaglandin J2 interacting factors

IntroductionProstaglandins (PGs) are autacoids synthesized from 20-carbon-containing polyunsaturated fatty acids. They have been detectedin almost every tissue and body fluid. Although PGs are notstored in tissue or cells, their production may increase in responseto diverse stimuli (Smith, 1989, 1992). The cellular targets of PGsare mainly G-protein coupled receptors (GPCR): for example,PGD2 binds to DP1 and DP2 that are members of the GPCR(Almishri et al., 2005).

The PGJ2 family, including PGJ2, D12-PGJ2 and 15d-PGJ2, aremetabolites of PGD2, and have been reported to show variouskinds of biological activities (Scher and Pillinger, 2005; Uchidaand Shibata, 2008). Considerable progress has been madetowards understanding the mechanisms of action of PGJ2.Among PGJ2 derivatives, 15d-PGJ2 has highest biological activityand the highest affinity to PPARg (Kliewer et al., 1995). One of themost important findings about this compound is that 15d-PGJ2 isa ligand for the transcription factor, PPARg, which promotes adi-pocyte differentiation (Forman et al., 1995; Kliewer et al., 1995).PPARg is one member of the nuclear receptor superfamily, and itsartificial high-affinity ligand is thiazolidinedione, an anti-diabetesdrug (Lehmann et al., 1995; Heikkinen et al., 2007). However thephysiological role of this compound in vivo still remains asintriguing issue. Since PGJ2 exerts its biological effects at least inpart through a reaction with cellular proteins, the identification

* Correspondence to: T. Imai, Department of Aging Intervention, NationalCenter for Geriatrics and Gerontology, Obu, Aichi 474-8522, Japan. E-mail:timai@ncgg.go.jp

a Department of Aging Intervention, National Center for Geriatrics and Ger-ontology, Obu, Aichi 474-8522, Japan

b Department of Metabolic disorder, Research Institute, International MedicalCenter of Japan, Shinjuku, Tokyo 162-8655, Japan

c Department of Biomolecular Engineering, Tokyo Institute of Technology,Yokohama, Kanagawa 226-8501, Japan

d Department of Human Nutrition, Sugiyama Jyogakuen University, Nagoya,Aichi 464-8662, Japan

e Department of Biochemistry and Integrative Medical Biology, School ofMedicine, Keio University, Tokyo 160-8582, Japan

f Biofrontier Center, Tokyo Institute of Technology, Yokohama, Kanagawa 226-8501, Japan

Abbreviations used: 15d-PGJ2, 15-deoxy-D12,14-PGJ2; DMEM, Dulbecco’smodified Eagle’s medium; DMSO, dimethyl sulfoxide; DP1, 2, prostanoidprostaglandin D2 receptor 1, 2; DTT, dithiothreitol; EDTA, ethylenediamine-tetraacetic acid; FBS, fetal bovine serum; GMA, glycidyl methacrylate; GPCR,G-protein coupled receptor; PGD2, prostaglandin D2; PGJ2, prostaglandin J2;PGJIF, prostaglandin J2 interacting factor; PMSF, phenylmethylsulfonyl fluo-ride; PPARg, peroxisome proliferator-activated receptor g, SDS-PAGE, sodiumdodecyl sulfate–polyacrylamide gel electrophoresis; VDAC1, voltage depen-dent anionic channel 1.

Research Article

Received 10 November 2009, Received 1 March 2010, Accepted 30 March 2010 Published online in Wiley Online Library: 04 June 2010

(wileyonlinelibrary.com) DOI 10.1002/bmc.1469

466

Biomed. Chromatogr. 2011; 25: 466–471Copyright © 2010 John Wiley & Sons, Ltd.

of target molecules of PGJ2 may facilitate the understanding ofthe diverse biological activities of PGJ2 in vivo.

On the other hand, to find cellular target proteins using affinitynanobeads technology, we have recently developed novelmagnetic nanobeads (FG beads, 200 nm in diameter) composedof magnetite particles/glycidyl methacrylate (GMA) styrenecopolymer/covered GMA (Nishio et al., 2008). These nanobeadshave several advantages over conventional affinity purificationsupports:(1) Their lack of pores results in efficient removal of residual pro-

teins during the washing steps, as well as easy access fortarget proteins to the fixed ligands.

(2) The small diameter of the beads (200 nm) provides a largesurface area (1 g of the beads have a surface area of 20 m2),giving the beads a high capacity to immobilize ligands.

(3) The chemical and physical stability of the beads permits cou-pling of ligands in the presence of a wide range of organicsolvents. As a result, FG-beads shows higher performancecompared with commercially available magnetic beads,adembeads (200 nm, ADEMTECH), nanomag-D (130 nm,micromod), and Dynabeads (2.8 mm, Invitrogen), in terms ofpurification efficiency of target proteins (Nishio et al., 2008).This enables the rapid and efficient purification of bindingproteins for capsaicin and L-arginine (Kuramori et al., 2009;Hiramoto et al., 2010).

In this study we apply this system for identification of PGJ2

target proteins. Firstly, we newly developed 15d-PGJ2

-immobilized nanobeads. Secondly, we showed that PPARg spe-cifically bound to 15d-PGJ2-immobilized beads to validate thissystem. Lastly, we successfully identified VDAC1 as a new PGJIFfrom HEK293 cell lysates. Here we show that 15d-PGJ2-immobilized beads are very effective for purification of PGJIFsdirectly from crude cell extracts.

Experimental

Materials

Highly efficient synthesis of 15d-PGJ2 was described previously (Acharyaand Kobayashi 2006). 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimidehydrochloride (EDC), triethylamine, N,N-dimethyl-4-aminopyridine(DMAP), N,N-dimethylformamide (DMF), acetic anhydride, dithiothreitoland iodoacetamide were purchased from Nacalai Tesque (Kyoto, Japan).Ethyleneglycol diglycidyl ether (EGDE) was purchased from Wako Chemi-cals (Osaka, Japan). Trypsin was obtained from Promega (Madison, WI,USA). HEK293 cells were obtained from American Type Culture Collection(Manassas, VA, USA).

15d-PGJ2-immobilized Beads

FG beads were prepared as previously described (Nishio et al., 2008).Epoxy groups on FG beads were aminolyzed by NH4OH and coupled toethyleneglycol diglycidyl ether (EGDE) to produce FGNEGDE beads.Epoxy groups on FGNEGDE beads were aminolyzed by NH4OH to produceFGNEGDEN beads. FGNEGDEN beads (5.0 mg) were incubated with5.0 mM 15d-PGJ2 in 500 mL of DMF containing EDC, triethylamine andDMAP at 25°C for 24 h. Unreacted amino groups on the surface of thebeads were masked with acetic anhydride in DMF containing triethy-lamine at 25°C for 24 h. 15d-PGJ2-immobilized beads were suspended indistilled water and stored at 4°C until use.

Preparation of Radiolabeled Recombinant PPARg andiNOS Protein

T7 promoter tagged DNA fragments of PPARg or iNOS cDNA were ampli-fied by PCR with two primers: T7 promoter fused to 5’ sequence of PPARg

cDNA and poly T fused to 3’ complimentary sequence. These DNA frag-ments were used to synthesize 35S-radiolabeled recombinant proteins ina coupled transcription/translation system according to the protocol ofmanufacture (Promega, WI, USA).

15d-PGJ2-binding Analysis

Control beads or 15d-PGJ2-immobilized beads (200 mg) were equilibratedwith binding buffer [20 mM HEPES–NaOH pH 7.9, 10% glycerol, 200 mM

KCl, 1 mM MgCl2, 0.2 mM CaCl2, 0.2 mM ethylenediaminetetraacetic acid(EDTA), 1 mM dithiothreitol (DTT) and 0.2 mM phenylmethylsulfonyl fluo-ride (PMSF)], and incubated with 200 mL of the radiolabeled PPARg at 4°Cfor 4 h using an RT-50 rotator (15 rpm, TAITEC, Saitama, Japan). Afterwashing with binding buffer, bound proteins were eluted by boiling for5 min with sodium dodecyl sulfate–polyacrylamide gel electrophoresis(SDS-PAGE) sample buffer. Eluates and inputs were subjected to SDS-PAGE. The gels were dried, and autoradiography was then performed tovisualize the radiolabeled proteins.

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

SDS-PAGE was used to analyze the protein samples from trypsin digestionanalysis and prostaglandin binding analysis, and to evaluate the affinitypurified PGJIFs as previously described (Laemmli, 1970). Samples werediluted in SDS-PAGE loading buffer (50 mM Tris–HCl pH 6.8, 2% SDS,100 mM b-mercaptoethanol, 10% glycerol, 0.01% bromophenolblue) anddenatured at 98°C for 5 min. Samples were applied in a 5–20% gradientprecast gel (Wako, Osaka, Japan), and run at 200V 60min using DPE-1020cassette electrophoresis Unit (Daiichi, Tokyo, Japan).

Luciferase Analysis

Luciferase reporter plasmid, pGL-ACOA-PPRE, which contains theluciferase gene under the control of peroxisome proliferator responseelement (PPRE) was constructed as follows. Complimentary oligonucle-otides corresponding to PPRE sequence derived from the promoter ofAcyl-CoA oxidase gene were synthesized: 5’-CGGGGACCAGGACAAAGGTCAGAGCTCGGGGACCAGGACAAAGGTCAGCTAGCGGGGACCAGGACAAAGGTCAC-3’ and 5’-TCGAGTGACCTTTGTCCTGGTCCCCGCTAGCTGACCTTTGTCCTGGTCCCCGAGCTCTGACCTTTGTCCTGGTCCCCGGTAC-3’. These twooligonucleotides were annealed and cloned into pGL4.10 (Promega)along with the thymidine kinase promoter derived from pGL4.74(Promega).

C2C12 cells (American Type Culture Collection, CRL-1772) and HEK293cells (American Type Culture Collection, CRL-1573) were grown in Dulbec-co’s modified Eagle’s medium (DMEM) supplemented with 10% fetalbovine serum (FBS). The 3T3-L1 cells (American Type Culture Collection,CL-173) were grown in DMEM supplemented with 10% bovine calf serum.These cells were seeded at 1 ¥ 105 cells per well in 24-well plate. After 24 h,the appropriate plasmids (pGL-ACOA-PPRE, 60 ng; pGL4.74, 6 ng;pcDNA3-FLAGhPg2, 210 ng; pCH-RXRa, 210 ng) were transfected to thesecells using Lipofectamine 2000 (Invitrogen). Twelve hours after transfec-tion, the medium was changed and the appropriate reagents (15d-PGJ2

or rosiglitazone) were added. Thirty hours after treatment, cell extractswere prepared and luciferase activities were determined using Dual-Gloluciferase assay system (Promega). Each transfection experiment was per-formed on triplicate cultures. The values are reported as means � SEM.Statistical significance (p < 0.005) was determined by unpaired Student’st-test (STATVIEW).

Measurement of Cell Proliferation

HEK293 cells were seeded at 1 ¥ 105 cells per well in 24-well plate. Thesecells were grown in DMEM supplemented with 5% FBS with or withoutPGJ2, and cells were harvested and counted the number after 96 h. Eachexperiment was performed on triplicate cultures. The values are reported

467

Affinity purification for prostaglandin J2 interacting factors

Biomed. Chromatogr. 2011; 25: 466–471 wileyonlinelibrary.com/journal/bmcCopyright © 2010 John Wiley & Sons, Ltd.

as means � SEM. Statistical significance (p < 0.005) was determined byunpaired Student’s t-test (STATVIEW).

Affinity Purification of PGJIFs

Extracts of HEK293 cells were prepared as previously described (Dignamet al., 1983). Control beads or 15d-PGJ2-immobilized beads (200 mg) wereequilibrated with binding buffer (20 mM HEPES–NaOH pH 7.9, 10% glyc-erol, 200 mM KCl, 1 mM MgCl2, 0.2 mM CaCl2, 0.2 mM EDTA, 1 mM DTT and0.2 mM PMSF), and incubated with 200 mL of the cell extracts at 4°C for 4 husing RT-50 rotator (15 rpm, TAITEC, Saitama, Japan). After washing withbinding buffer, bound proteins were eluted with SDS-PAGE sample buffer.

Mass Spectrometry Analysis of PGJIFs

Affinity-purified PGJIFs were separated by SDS-PAGE and gels subjectedto silver staining as previously described (Shevchenko et al., 1996; Hira-moto et al., 2010). The specific protein bands were excised, reduced with10 mM DTT followed by alkylation with 55 mM iodoacetamide. Band sliceswere digested with trypsin (12 mg/mL) overnight and desalted withZipTip C18 (Millipore, Billerica, MA, USA). The extracted peptides werethen separated via nanoflow liquid chromatography (LC, Paradigm MS4,AMR, Tokyo, Japan) using a reverse-phase C18 column (Magic C18, AMR,Tokyo, Japan). The LC eluent was coupled to a micro-ionspray sourceattached to a LCQ Advantage MAX mass spectrometer (Thermo ElectronCorporation, Waltham, MA, USA). All MS/MS spectra were searched usingthe TurboSEQUEST algorithm within the BioWorks 3.2 software (ThermoElectron Corporation, Waltham, MA, USA).

Recombinant VDAC1 Protein Production

VDAC1 cDNA was amplified by RT-PCR with a pair of primers of 5’-CCGGGC GGA TCC ATG GCT GTG CCA CCC ACG TAT-3’ and 5’-CCC CCC CTC GAGTTA TGC TTG AAA TTC CAG CCT-3’, and inserted in pCMV-Tag 2B vector(Stratagene). The expression vector of pCMV-FVDAC1 or empty pCMVvector was transfected in HEK 293 cells using Lipofectamine 2000 (Invit-rogen) with standard protocol. Two days after transfection, HEK 293 cellswere harvested with binding buffer (20 mM HEPES–NaOH pH 7.9, 10%glycerol, 200 mM KCl, 1 mM MgCl2, 0.2 mM CaCl2, 0.2 mM EDTA, 1 mM DTTand 0.2 mM PMSF) with 1% of n-octylglucoside, centrifuged at 1300g for5 min and the supernatant was recovered. The supernatant was dialyzedagainst binding buffer for 4 h.

Results and Discussion

Preparation of 15d-PGJ2-immobilized Beads (Fig. 1)

To purify new target for 15d-PGJ2, we prepared 15d-PGJ2-immobilized beads. A schematic representation of the procedurefor conjugating 15d-PGJ2 to FG-beads is depicted in Fig. 1. Briefly,epoxy groups on FG beads were aminolyzed by NH4OH andcoupled to EGDE to produce FGNEGDE beads. EGDE, introducedas a spacer is important for reduction of steric hindrance. Epoxygroups on FGNEGDE beads were aminolyzed by NH4OH toproduce FGNEGDEN beads. 15d-PGJ2 was then conjugated toFGNEGDEN beads.

Evaluation of 15d-PGJ2-immobilized Beads (Fig. 2)

The 15d-PGJ2-immobilized beads were analyzed its bindingability to some PGJIFs such as PPARg and non-PGJIFs such as iNOS(Fig. 2). These radiolabeled proteins were mixed with 15d-PGJ2-immobilized beads (PG) or control beads (Co), and bound pro-teins were analyzed using SDS-PAGE. PPARg bound to 15d-PGJ2-immobilized beads and not to control beads. The recovery rate ofPPARg from 15d-PGJ2-immobilized beads was 0.3%, which is a

reasonable value for affinity purification because the interactionbetween PPARg and 15d-PGJ2 is transient and weak (Aldini et al.,2007; Uchida and Shibata, 2008). Recombinant iNOS proteinbound neither to 15d-PGJ2-immobilized beads nor to controlbeads. These results suggest that PGJIFs can be purified and non-PGJIFs cannot be purified in this nanobeads system.

Selection of Cell Lines for PPARg-independent 15d-PGJ2

Activity as Purification Materials (Fig. 3)

Bell-Parikh et al. reported a significant difference between PGJ2

concentration of PPARg transactivation and that of intracellularconcentration, suggesting the possibility of the existence of anew target of PGJ2 (Bell-Parikh et al., 2003; Powell, 2003). Alsoactin was cloned as new target of 15-deoxy-D12,14-PGJ2 (Aldiniet al., 2007).

To select the material for purification, we analyzed the effectsof 15d-PGJ2 on PPARg-mediated gene expression in variety of celllines. The adipogenic 3T3-L1 cells were transfected withluciferase reporter plasmid, which contains PPRE sequencesupstream of the luciferase reporter gene, and mammalianexpression plasmids for PPARg and RXRa. 15d-PGJ2 as well asrosiglitazone, which is a strong agonist for PPARg, stimulated theluciferase gene expression (Fig. 3). Moreover, in 3T3-L1 cells,PPARg-mediated gene expression was stimulated by 15d-PGJ2 orrosiglitazone without exogenously expressed PPARg and RXRa.Rosiglitazone or 15d-PGJ2 stimulates PPARg-mediated geneexpression in HeLa cells and C2C12 cells as well as in 3T3-L1 cells.In HEK293 cells, however, PPARg-mediated gene expression wasnot stimulated by 15d-PGJ2 or rosiglitazone without exogenousPPARg and RXRa (Fig. 3). The expression of PPARg is much less inHEK293 cells than that in 3T3-L1 cells (data not shown). Theseresults suggest that in HEK293 cells 15d-PGJ2 shows its biologicalactivity via another target protein.

15d-PGJ2-dependent Suppression of HEK 293 CellGrowth (Fig. 4)

Previous reports have indicated that PGJ2 family members havethe growth inhibition activity and apoptosis induction activity,although the mechanisms of the actions have been obscure(Ohno et al., 1988; Higashiyama et al., 1996). We examined theeffect of 15d-PGJ2 on the growth of HEK293 cells. The 15d-PGJ2

suppressed HEK293 cell growth in a dose-dependent manner(Fig. 4). Moreover, the suppression of the cell growth can bedetected in nanomolar concentrations of 15d-PGJ2, whereasstimulation of PPARg-mediated gene expression in 3T3-L1 cellscan only be detected in micromolar concentrations of 15d-PGJ2.The intracellular concentration of 15d-PGJ2 is normally nanomo-lar level. Taking these results together, HEK293 cells are suitablefor purification of new target for 15d-PGJ2 using nanobeadstechnology.

Purification and Identification of PGJIF from HEK293 CellsUsing 15d-PGJ2-immobilized Beads (Fig. 5)

Instead of recombinant proteins, crude extract of HEK293 cells, inwhich almost no endogenous PPARg-dependent activity wasobserved in Fig. 3, was mixed with 15d-PGJ2-immobilized beadsto purify unknown PGJIFs from HEK293 cells. The bound proteinswere separated by SDS-PAGE and visualized by silver staining.The patterns of bound proteins were almost same in the four

468

N. Maekawa et al.

Biomed. Chromatogr. 2011; 25: 466–471wileyonlinelibrary.com/journal/bmc Copyright © 2010 John Wiley & Sons, Ltd.

independent purifications and with other types of HEK293extract (cytosolic fraction or whole cell extract, data not shown).All of the bound proteins obtained from the four purificationswere put together and subjected to SDS-PAGE/silver staininganalysis (Fig. 5A). The polypeptide of ca 29 kDa was specificallybound to the 15d-PGJ2-immobilized beads, and was successfullyidentified as VDAC1 by mass spectrometry analysis (Fig. 5B).

The recombinant VDAC1 was prepared in order to confirm the15d-PGJ2 binding activity of this protein. This recombinantprotein was mixed with 15d-PGJ2-immobilized beads (PGJ2) orcontrol beads (Cont), and bound proteins were analyzed usingSDS-PAGE (Fig. 5C). VDAC1 bound to 15d-PGJ2-immobilizedbeads and not to control beads. The excess amount of 15d-PGJ2

inhibited the binding between VDAC1 and 15d-PGJ2-immobilized beads (compare lane 5 with lane 6 in Fig. 5C), indi-cating that VDAC1 specifically bound to 15d-PGJ2. The recoveryrates of VDAC1 from 15d-PGJ2-immobilized beads were ca 1.0 %.This recovery rate suggested that the binding activity of VDAC1was higher than that of PPARg, indicating that 15d-PGJ2 affinity toVDAC1 is higher than that to PPARg, and that VDAC1 specificallybound 15d-PGJ2. The new type of target of 15d-PGJ2 was effec-tively purified and identified using newly developed 15d-PGJ2-immobilized beads.

Previous reports have indicated that 15d-PGJ2 promotes cyto-chrome c release (Landar et al., 2006). It is well-known that cyto-chrome c is released from mitochondria through VDAC (Shimizu

Figure 1. Preparation of 15d-PGJ2-immobilized beads. Epoxy groups on FG beads wereaminolyzed by NH4OH (FGN beads) and coupled to EGDE to produce FGNEGDE beads.Epoxy groups on FGNEGDE beads were aminolyzed by NH4OH to produce FGNEGDENbeads. FGNEGDEN beads were then coupled with carboxyl groups of 15d-PGJ2 in DMFcontaining EDC, triethylamine and DMAP.

469

Affinity purification for prostaglandin J2 interacting factors

Biomed. Chromatogr. 2011; 25: 466–471 wileyonlinelibrary.com/journal/bmcCopyright © 2010 John Wiley & Sons, Ltd.

et al., 1999; Tsujimoto and Shimizu, 2000), and promotes the apo-ptosis pathway. These observations make us imagine that 15d-PGJ2 binds and opens VDAC1, promotes cytochrome c release,and finally induces growth inhibition or apoptosis.

ConclusionsPGJ2 family members are known to be natural ligands for PPARgas well as powerful inhibitors of cell proliferation and viral repli-

cation. In addition, these compounds have multiple functionsthat remain incompletely understood, such as pro- and anti-apoptotic effects, cell differentiation-inducing effects, and inhibi-tory effects on inflammatory processes, whether they depend onPPARg or not.

In order to purify, identify and study PGJIFs, we have devel-oped a new effective affinity purification system using 15d-PGJ2-immobilized beads. In this study, we showed that this system isvery effective for purification of PGJIFs directly from crude cellextracts. We conclude that 15d-PGJ2-immobilized beads are apowerful tool for purification of PGJIFs.

In general, several kinds of column chromatography are nec-essary to purify several different proteins from crude cell extracts.In addition, magnetic carriers are favorable for automated high-throughput screening systems using a permanent magnet.Therefore, 15d-PGJ2-immobilized beads developed in this studyare useful for efficient purification of PGJIFs from various types ofcells, from widely different conditions, or in the presence of avariety of reagents. We expect that 15d-PGJ2-immobilized beadswill contribute to the purification of PGJIFs from different kinds oftissues or cells. The identification and characterization of thesePGJIFs should reveal the mechanism of various biological func-tions of PGJ2.

AcknowledgementsThis work was supported by a Grant-in-Aid from the Ministry ofEducation, Culture, Sports, Science and Technology (MEXT18659493), the Ministry of Health, Labor and Welfare (MHLW17C-2, H21-186 and H22-010) and Seichokagakukyokai, Kanzawa,Suzuken, Mishimakaiun, Hokto, Nakatomi Foundations to T. I.

Figure 2. Evaluation of 15d-PGJ2-immobilized beads. In vitro radiola-beled PPARg and iNOS proteins were mixed with 15d-PGJ2-immobilizedbeads (PG) or control beads (Co), and bound proteins were recovered.One percent of input (In), eluate fraction from control beads (Co) andeluate fraction from 15d-PGJ2-immobilized beads (PG) were analyzedusing SDS-PAGE (5–20% gradient gel). The PGJ2-binding PPARg is indi-cated with open circle.

Figure 3. Rosiglitazone and 15d-PGJ2 simulate PPARg-dependent tran-scription in 3T3-L1, HeLa and C2C12 cells, but not in HEK 293 cells. 3T3-L1cells were transfected with PPRE-dependent luciferase reporter plasmidwith mammalian expression plasmids for PPARg and RXRa (columns 1–3),or without expression plasmid (columns 4–6). HeLa cells (columns 7–9),C2C12 cells (columns 10–12) or HEK293 cells (columns 13–15) were trans-fected with PPRE-dependent luciferase reporter plasmid without expres-sion plasmid. Twelve hours after transfection, the medium was changedand 10 mM of rosiglitazone (columns 2, 5, 8, 11 and 14) or 15d-PGJ2

(columns 3, 6, 9, 12 and 15) was added. Cell extracts were prepared 30 hlater and luciferase activities were measured. Relative luciferase activitiesare presented. *,**Statistically significant with its control: *p < 0.05; **p <0.005. #Statistically not significant with its control.

Figure 4. PGJ2 inhibits HEK293 cell growth. HEK293 cells were grownwith 10 nM (column 2), 100 nM (column 3) or 1 mM (column 4) of PGJ2 orwithout PGJ2 (column 1). Average cell numbers are presented. *Statisti-cally significant with its control (without PGJ2, column 1); p < 0.05.

470

N. Maekawa et al.

Biomed. Chromatogr. 2011; 25: 466–471wileyonlinelibrary.com/journal/bmc Copyright © 2010 John Wiley & Sons, Ltd.

ReferencesAcharya HP and Kobayashi Y. Highly efficient total synthesis of D12-PGJ2,

15-deoxy-D12,14-PGJ2, and their analogues. Tetrahedron 2006; 62:3329–3343.

Aldini G, Carini M, Vistoli G, Shibata T, Kusano Y, Gamberoni L, Dalle-Donne I, Milzani A and Uchida K. Identification of actin as a 15-deoxy-D12,14-prostaglandin J2 target in neuroblastoma cells: massspectrometric, computational, and functional approaches to investi-gate the effect on cytoskeletal derangement Biochemistry 2007; 46:2707–2718.

Almishri W, Cossette C, Rokach J, Martin JG, Hamid Q and Powell WS.Effects of prostaglandin D2, 15-deoxy-D12,14-PGJ2, and selective DP1and DP2 receptor agonists on pulmonary infiltration of eosinophils inbrown Norway rats. Journal of Pharmacology and ExperimentalTherapy 2005; 313: 64–69.

Bell-Parikh LC, Ide T, Lawson JA, McNamara P, Reilly M and FitzGerald GA.Biosynthesis of 15-deoxy-D12,14-PGJ2 and the ligation of PPARg Journalof Clinical Investigations 2003; 112: 945–955.

Dignam JD, Lebovitz RM and Roeder RG. Accurate transcription initiationby RNA polymerase II in a soluble extract from isolated mammaliannuclei. Nucleic Acids Research 1983; 11: 1475–1489.

Forman BM, Tontonoz P, Chen J, Brun RP, Spiegelman BM and Evans RM.15-Deoxy-D12,19-PGJ2 is a ligand for the adipocyte determinationfactor PPARg. Cell 1995; 83: 803–812.

Heikkinen S, Auwerx J and Argmann CA. PPARg in human andmouse physiology Biochimica et Biophysica Acta 2007; 1771: 999–1013.

Higashiyama K, Niiya K, Ozawa T, Hayakawa Y, Fujimaki M and SakuragawaN. Induction of c-fos protooncogene transcription and apoptosis bydelta 12-prostaglandin J2 in human Pl-21 myeloid leukemia and RC-K8pre-B lymphoma cells. Prostaglandin 1996; 52: 143–156.

Hiramoto M, Maekawa N, Kuge T, Ayabe F, Watanabe A, Masaike Y,Hatakeyama M, Handa H and Imai T. High-performance affinity chro-matography method for identification of L-arginine interactingfactors using magnetic nanobeads Biomedical Chromatography 2010;24: 606–612.

Kliewer SA, Lenhard JM, Willson TM, Patal I, Morris DC and Lehmann JM. Aprostaglandin J2 metabolite binds peroxisome proliferator-activatedreceptor gamma and promotes adipocyte differentiation. Cell 1995;83: 813–819.

Kuramori C, Azuma M, Kume K, Kaneko Y, Inoue A, Yamaguchi Y, Kabe Y,Hosoya T, Kizaki M, Suematsu M and Handa H. Capsaicin binds to

prohibitin 2 and displaces it from the mitochondria to the nucleus.Biochemical and Biophysical Research Communications 2009; 379:519–525.

Laemmli UK. Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 1970; 227: 680–685.

Landar A, Shiva S, Levonen AL, Oh JY, Zaragoza C, Johnson MS and Darley-Usmar VM. Induction of the permeability transition and cytochrome crelease by 15-deoxy-Delta12,14-prostaglandin J2 in mitochondria.Biochemistry Journal 2006; 394: 185–195.

Lehmann JM, Moore LB, Smith-Oliver TA, Wilkison WO, Willson TM andKliewer SA. An antidiabetic thiazolidinedione is a high affinity ligandfor peroxisome proliferator-activated receptor gamma (PPARgamma). Journal of Biological Chemistry 1995; 270: 12953–12956.

Nishio K, Masaike Y, Ikeda M, Narimatsu H, Gokon N, Tsubouchi S,Hatakeyama M, Sakamoto S, Hanyu N, Sandhu A, Kawaguchi H, Abe Mand Handa H. Development of novel magnetic nano-carriers for high-performance affinity purification. Colloids and Surfaces B, Biointerfaces2008; 64: 162–169.

Ohno K, Sakai T, Fukushima M, Narumiya S and Fujiwara M. Site andmechanism of growth inhibition by prostaglandins. IV. Effect of cyclo-pentenone prostaglandins on cell cycle progression of G1-enrichedHeLa S3 cells. Journal of Pharmacology and Experimental Therapy 1988;245: 294–298.

Powell WS. 15-deoxy-D12,14-PGJ2: endogenous PPARg ligand or minoreicosanoid degradation product? Journal of Clinical Investigations2003; 112: 828–830.

Scher JU and Pillinger MH. 15d-PGJ2: the anti-inflammatory prostaglan-din? Clinical Immunology 2005; 114: 100–109.

Shevchenko A, Wilm M, Vorm O and Mann M. Mass spectrometricsequencing of proteins silver-stained polyacrylamide gels. AnalyticalChemistry 1996; 68: 850–858.

Shimizu S, Narita M and Tsujimoto Y. Bcl-2 family proteins regulate therelease of apoptogenic cytochrome c by the mitochondrial channelVDAC. Nature 1999; 399: 483–487.

Smith WL. The eicosanoids and their biochemical mechanisms of action.Biochemistry Journal 1989; 259: 315–324.

Smith WL. Prostanoid biosynthesis and mechanisms of action. AmericanJournal of Physiology 1992; 263: F181–191.

Tsujimoto Y and Shimizu S. VDAC regulation by the Bcl-2 family of pro-teins. Cell Death and Differentiation 2000; 7: 1174–1181.

Uchida K and Shibata T. 5-Deoxy-delta(12,14)-prostaglandin J2: an elec-trophilic trigger of cellular responses. Chemical Research and Toxicol-ogy 2008; 21: 138–144.

Figure 5. Purification and identification of PGJIF from HEK293 cell extracts by 15d-PGJ2-immobilized beads. (A)Purification of PGJIFs. HEK 293 cell extracts were mixed with 15d-PGJ2-immobilized beads (PG, lane 3) or controlbeads (Co, lane 2), and bound proteins were separated by SDS-PAGE (5–20% gradient gel) and visualized by silverstaining. (B) Identification of PGJIFs. Four polypeptides were identified by ion-spray mass spectrometry. Identi-fied amino acid sequences are indicated. (C) Evaluation of 15d-PGJ2 binding activity of newly identified PGJIF.Recombinant VDAC1 protein with FLAG tag was mixed with 15d-PGJ2-immobilized beads (PGJ2) or control beads(Cont) in the binding buffer without (lanes 1–5) or with (lane 6) 15d-PGJ2, and bound proteins were recovered.Ten percent of input (In), eluate fraction from control beads (lanes 2 and 3) or 15d-PGJ2-immobilized beads (lanes4 and 5), without FLAG-VDAC1 protein (lanes 2 and 4) or with FLAG-VDAC1 protein (lanes 3, 5 and 6) wereanalyzed using SDS-PAGE (5–20% gradient gel), and visualized with western blotting with anti-FLAG antibody.The PGJ2-binding VDAC1 is indicated with open circles.

471

Affinity purification for prostaglandin J2 interacting factors

Biomed. Chromatogr. 2011; 25: 466–471 wileyonlinelibrary.com/journal/bmcCopyright © 2010 John Wiley & Sons, Ltd.