www.elsevier.com/locate/yabio

Analytical Biochemistry 335 (2004) 253–259

ANALYTICAL

BIOCHEMISTRY

A yeast two-hybrid technology-based system for the discoveryof PPARc agonist and antagonist

Qing Chen, Jing Chen, Tao Sun, Jianhua Shen, Xu Shen*, Hualiang Jiang*

Drug Discovery Design Center, State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica,

Shanghai Institutes for Biological Sciences, Graduate School of the Chinese Academy of Sciences, Chinese Academy of Sciences,

555 Zuchongzhi Road, Shanghai 201203, China

Received 10 July 2004

Abstract

Peroxisome proliferator-activated receptor c (PPARc) is an important therapeutic drug target against several diseases such asdiabetes, inflammation, dyslipidemia, hypertension, and cancer. Ligand binding to PPARc is responsible for controlling the biolog-ical functions, and developing new technology to measure ligand–PPARc binding is significant for both the function study of thereceptor and ligand discovery. In this study, we exploited an efficient approach for the discovery of PPARc agonist and antagonistvia a yeast two-hybrid system based on the fact that PPARc interacts with the coactivator CBP (CREP-binding protein) ligand-de-pendently. We employed the MEL1 reporter gene instead of the traditionally used LacZ gene to evaluate the protein–protein inter-actions by conducting a convenient a-galactosidase assay in the yeast strain AH109 with genes of PPARc-LBD (ligand-bindingdomain) and CBP N terminus introduced. With this built screening platform, the EC50 values of the PPARc agonists rosiglitazone,troglitazone, pioglitazone, indomethacin, 15-deoxy-D12,14-prostaglandin J2 (15d-PGJ2), and GI262570 were investigated, and thequantitatively antagonistic effect by IC50 of the PPARc typical antagonist GW9662 on the rosiglitazone agonistic activity was fullyexamined. The reliability of this presented system evaluated by the comparable agreement of EC50 and IC50 values for the test com-pounds with the reported ones indicated that this yeast two-hybrid-based approach is powerful for PPARc agonist and antagonistscreening. In addition, because this screening system is designed for use in a microtiter plate format where numerous chemicals couldbe readily screened, it is hoped that this yeast two-hybrid screening approach may be adaptable for high-throughput settings.� 2004 Elsevier Inc. All rights reserved.

Keywords: Peroxisome proliferator-activated receptorc (PPARc); Yeast two-hybrid; CREP-binding protein (CBP); Agonist; Antagonist;a-Galactosidase

The peroxisome proliferator-activated receptor(PPAR)1 belongs to the nuclear receptor superfamilyand contains three subtypes: PPARa, PPARc, andPPARd (also called PPARb) [1]. PPARc functions in

0003-2697/$ - see front matter � 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ab.2004.09.004

* Corresponding authors. Fax: +86 21 5080 7088.E-mail addresses: xshen@mail.shcnc.ac.cn (X. Shen), hljiang@mail.shcnc1 Abbreviations used: PPAR, peroxisome proliferator-activated receptor; C

13-HODE, 9-hydroxyoctadecadienoic acid; 15d-PGJ2, 15-deoxy-D12,14-prinflammatory drug; CARLA, coactivator-dependent receptor ligand assay;HTS, high-throughput screen; LBD, ligand-binding domain; DMSO, dimetDNA-binding domain; AD, activation domain; mCBP, mouse CREP-bindingmedium lacking Leu and Trp; HDL, high-density lipoprotein.

regulating genes involved in lipogenesis, adipocyte dif-ferentiation, cell proliferation, inflammatory processes,and modulation of insulin sensitivity as a ligand-induc-ible transcription factor and has been developed as an

.ac.cn (H. Jiang).BP, CREP-binding protein; 9-HODE, 9-hydroxyoctadecadienoic acid;ostaglandin J2; TZD, thiazolidinedione; NSAID, nonsteroidal anti-SPA, scintillation proximity assay; SPR, surface plasmon resonance;hyl sulfoxide; PNP-a-Gal,p-nitrophenyl a-DD-galactopyranoside; DBD,protein; SD-T, SD minimal medium lacking Trp; SD-LT, SD minimal

254 Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259

important therapeutic target in treating variousmetabolic disorders such as diabetes, inflammation, dysl-ipidemia, hypertension, and cancer [1,2]. The ligand-dependent recruitment of coactivators, such as CBP(CREP-binding protein), SRC-1, and TIF2, is regardedas a critical step for PPARc-mediated ligand-dependenttranscriptional activation [3,4]. Ligandbinding toPPARcis responsible for controlling the biological functions, anddiscovering new ligands that may modulate PPARc�sfunction is now a main focus in the pharmaceuticalindustry.

Up to now, a variety of natural and synthetic PPARcagonists have been reported. Natural ligands, such as9- or 13-hydroxyoctadecadienoic acid (9-HODE or 13-HODE, respectively) and 15-deoxy-D12,14-prostaglan-din J2 (15d-PGJ2), have been demonstrated to increasePPARc-mediated transactivation [5,6]. Of the syntheticagonists, thiazolidinediones (TZDs) are clinically usedto reduce insulin resistance and hyperglycemia in type2 diabetes [7,8] and to stimulate adipocyte differentiation[9]. Tyrosine-based PPARc agonists, such as GI262570,GW1929, and GW7845, are synthetic nonthiazolidined-ione agonists [10–12], whereas indomethacin is anotherkind of non-TZD synthetic agonist known as a nonste-roidal anti-inflammatory drug (NSAID) [13].

However, although clinical benefits have beenachieved by the use of PPARc agonists in the treatmentof type 2 diabetes, they are often accompanied by unde-sired side effects such as weight gain and edema [14]. It isreported that high-fat diet-induced or aging-inducedadipocyte hypertrophy, obesity, and insulin resistancecan be well avoided in a heterozygous PPARc-deficientmice model [15,16]. Moreover, moderately reducing thetranscriptional activity of PPARc via Pro12Ala poly-morphism in human PPARc2 also exerted resistance ef-fects against type 2 diabetes and obesity [17–19]. All ofthese findings have provided a new hint for developingPPARc antagonist or partial agonist to discover novelPPARc-modulating drugs in anti-diabetic drug design,which might retain efficacious insulin-sensitizing proper-ties while minimizing potential adverse side effects.

In fact, developing new technology to discoverPPARc ligand has been significant to both the receptor�sfunction study and drug lead compound discovery.Now, numerous types of cell-free assays have beendeveloped for PPARc ligand screening, including com-petition radioreceptor assay [20], coactivator-dependentreceptor ligand assay (CARLA) [21], scintillation prox-imity assay (SPA) [22], and surface plasmon resonance(SPR) biosensor-based assay [23]. Despite the high-throughput attributes from these cell-free assays, ligandscreening is not performed in a cellular context and com-pounds are mostly selected according to their bindingaffinity to the receptor with unclear potency on thetransactivation activity. In this respect, however, cellularscreens employing mammalian cells, such as transactiva-

tion assay [24] and chimeric receptor transactivation as-say [7], evidently have provided a more physiologicallyrelevant model system. Nevertheless, it is known thatthe manipulation of mammalian cells is generally costly,time-consuming, and often difficult to apply to auto-mated systems used for high-throughput screen (HTS).In comparison, the yeast cells may present a more con-venient alternative for ligand screening, in a heterolo-gous yet eukaryotic environment, with easy geneticmodification. Recently, Taniguchi and Mizukami [25]developed a yeast-based method for PPARc agonistscreening that involved the comparatively laboriousb-galactosidase assay using lacZ as the two-hybrid re-porter gene, and little was described concerning thescreening details. In this work, we successfully devel-oped an approach for screening both PPARc agonistand antagonist based on a yeast two-hybrid system. Inthis system, we employed theMEL1 gene (the GAL genefamily member) as the two-hybrid reporter gene; assuch, the ligand-dependent interactions betweenPPARc-LBD (ligand-binding domain) and CBP couldbe quantitatively evaluated by the convenient a-galacto-sidase assay. In addition, the EC50 values of the typicalPPARc agonists and the IC50 value of the antagonistGW9662 in inhibiting the agonistic activity of rosiglitaz-one against PPARc were fully determined by using theconstructed yeast two-hybrid system. The reliability ofthis presented system evaluated by the comparableagreement of EC50 and IC50 values for the test com-pounds with the reported ones indicated that this yeasttwo-hybrid-based approach is powerful for PPARcagonist and antagonist screening.

Materials and methods

Reagents

Indomethacin was purchased from Calbiochem, 15d-PGJ2 was purchased from Biomol, and GW9662 waspurchased from Merck. Rosiglitazone, pioglitazone,and troglitazone were obtained from Cayman Chemical,and GI262570 was synthesized in our laboratory accord-ing to the published methods [10,11]. All of the test com-pounds were dissolved in dimethyl sulfoxide (DMSO) as20 mM stock solution for use.

Yeast nitrogen base without amino acids, yeast syn-thetic dropout medium supplement without tryptophan,yeast synthetic dropout medium supplement without leu-cine and tryptophan, D(+)-glucose, and p-nitrophenyl a-DD-galactopyranoside (PNP-a-Gal) all were purchasedfrom Sigma Chemical. All yeast media were preparedaccording to the Yeast Protocols Handbook (PT3024-1, Clontech). Assay buffer was prepared freshly beforeeach use by combining 2 volumes of 0.5 M NaAc buffer,pH 4.5, with 1 volume of 100 mM PNP-a-Gal solution.

Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259 255

Strains and plasmids construct

The Clontech MATCHMAKER Two-Hybrid Sys-tem 3, which includes the yeast host strain AH109 (MA-

Ta:trp1-901, leu2-3, 112, ura3-52, his3-200, gal4D,gal80D; LYS2: GAL1UAS–GAL1TATA–HIS3, GA-L2UAS–GAL2TATA–ADE2; URA3: MEL1UAS–MEL1-

TATA–lacZ MEL1) and yeast expression vectorspGBKT7 and pGADT7 encoding the GAL4 DNA-binding domain (DBD) and activation domain (AD),respectively, was a generous donation from ProfessorGong (Shanghai Institutes for Biological Sciences, Chi-nese Academy of Sciences, China). pCDNA3.1-hPPARcvector harboring the full length of hPPARc coding se-quence is owned by our lab, and CMV–mCBP (mouseCREP-binding protein) vector harboring the full lengthof mCBP coding sequence was a gift from ProfessorRosenfeld (Howard Hughes Medical Institute, Univer-sity of California, USA).

Oligonucleotides PPARc(+)[5 0-GTAATACGACTCACTATAGGGCGA-3 0] and PPARc(�)[5 0-TCCGAGTCACTTTAAAATTTGTAT-3 0] were used to amplifythe human PPARc-LBD coding sequence (193-475aa)from pCDNA3.1-hPPARc vector. The PCR productswere digested with NdeI and SalI and were subclonedinto pGBKT7 vector digested with the same enzymesto construct pGBKT7-hPPARc-LBD, which expressesPPARc-LBD as a C-terminal fusion protein followingGAL4-DBD. Oligonucleotides CBP(+)[5 0-AACATATGATGGCCGAGAACTTGCTGGACG-3 0] and CBP(�)[50- AAGGATCCCTGTTGCCCTGCACCAACAG-3 0] were used to amplify the cDNA at the N terminus ofCBP(1–464aa) from CMV-mCBP vector. The PCRproducts were digested with NdeI and BamHI and wereinserted into pGADT7 using the NdeI and BamHI sitesin the polylinkers to produce pGADT7-mCBP express-ing mCBP as a C-terminal fusion protein followingGAL4 AD. Both clones were performed in Escherichiacoli strain DH5a and sequenced.

Recombinant plasmid of pGBKT7-hPPARc-LBDwas introduced into AH109 using the lithium acetatemethod [26]. Selection was done on SD minimal mediumlacking Trp (SD-T) solidified with agar at 30 �C for 2–3days to get transformant P1. Thereafter, transformantP1C2 was produced by introducing pGADT7-mCBPinto transformant P1 by the same method, and the selec-tion of transformants harboring both of the plasmidswas processed on SD minimal medium plate lackingLeu and Trp (SD-LT) at 30 �C for 2–3 days.

a-Galactosidase activity assay

The quantitative a-galactosidase assay was per-formed by using PNP-a-Gal as substrate. Yeast cellswere precultured in 3 ml of liquid medium overnight(16–18 h) at 30 �C with shaking (250 rpm). The over-

night cultures were diluted with fresh medium, and then5 ll of DMSO or test compound was added into 495-lldiluted cultures independently followed by well mixing.The obtained test cultures were incubated at 30 �C fora given time period. After measuring OD600 using aspectrophotometer (Hitachi U-2010), cells were centri-fuged and 16 ll of cell culture medium supernatants orunused culture media (as a reagent blank) was trans-ferred into a 96-well plate (Corning Costar 96-wellflat-bottom plate). Then 48 ll of assay buffer was addedinto each sample. After a 60-min incubation at 30 �C,the reaction was terminated by the addition of 136 llof stop buffer (1 M Na2CO3) and OD410 was measured(Benchmark Plus microplate spectrophotometer, Bio-Rad). One unit of a-galactosidase is defined as theamount of enzyme that hydrolyzes 1 lM PNP-a-Galto p-nitrophenol and DD-galactose in 1 min at 30 �C inacetate buffer, pH 4.5 [27]. The a-galactosidase activityin milliunits per cell was calculated according to the fol-lowing formula:

a�Galactosidase ActivityðmilliunitsÞ¼ OD410 � V f � 1000=ðe� b� t � V i �OD600Þ;

where t is elapsed time (min) of incubation, Vf is the finalvolume of assay (200 ll), Vi is the volume of culturemedium supernatant added (16 ll), and OD600 is the celldensity at the start of the assay, e · b is p-nitrophenolmolar absorptivity at 410 nm · the light path(cm) = 10.5 (ml/mol) (Yeast Protocols Handbook).

In determining EC50 or IC50 values, 200-ll cell cul-tures incubated with ligands were transferred into a96-well plate and OD600 was obtained using the micro-plate spectrophotometer. Thereafter, the plate was cen-trifuged in a microplate centrifuge, and the generatedsupernatants were used for a-galactosidase assay as de-scribed above.

Results and discussion

Ligand-dependent hPPARc–LBD/CBP interaction

It is well known that PPARc demonstrates its ligand-dependent transactivation activity against its targetgenes by recruiting one of the several types of nuclearreceptor coactivator complex. These coactivators in-clude CBP, SRC-1, and TIF2 [3,4], and the LBD ofPPARc mediates the ligand-dependent interactions be-tween PPARc and coactivators [3]. It has been reportedthat CBP participates in its interaction with PPARcthrough its N-terminal moiety [28]. In our current builtyeast two-hybrid system, hPPARc-LBD and the N-ter-minal moiety of mCBP were coexpressed in AH109 asC-terminal fusion proteins following GAL4-DBD andAD separately, and the ligand-dependent interactions

Fig. 2. Time course of the effect of rosiglitazone treatment on theinteractions between hPPARc-LBD and mCBP. Transformant P1C2was precultured in SD-TL overnight. Cells were diluted to variedinitial OD600 values of 0.6, 0.45, 0.30, 0.22, 0.15, 0.11, and 0.09 andwere treated with 0.1% DMSO (as vehicle control) or 10 lMrosiglitazone for different time intervals of 2, 4, 6, 8, 10, 12, and 13 hin sequence. Relative a-galactosidase activity corresponds to the cellstreated with 0.1% DMSO, and the values shown are the means ± stan-dard deviations from three independent experiments.

256 Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259

between them were detected (Fig. 1). Four kinds ofPPARc agonists—15d-PGJ2 (natural agonist), rosiglit-azone (TZD type), GI262570 (tyrosine-based syntheticligand), and indomethacin (NSAID)—were found tobe capable of inducing interactions between hPPARc-LBD and mCBP. In tranformant P1, no a-galactosidaseactivity could be detected regardless of any kinds ofagonists added in the test, thereby eliminating the possi-bility that bait protein itself could activate the expres-sion of the MEL1 reporter gene nonspecifically.DMSO, the solvent for treating all of the ligands usedin our assays, was determined to have no effects on theinteractions between hPPARc-LBD and mCBP (datanot shown).

To this point, several yeast two-hybrid systems havebeen established to quantitatively detect ligand-depen-dent interactions between nuclear receptors and coacti-vators and with estrogen receptors in particular[29,30]. However, manipulation details regarding ligandtreatments in these assays are quite different, for exam-ple, the terminal concentration of DMSO in test culturesand the time elapsed in incubation with chemicals. Forthis reason, we evaluated the DMSO influence in the as-say using different concentrations of DMSO (1–5%), butno influence was detected (data not shown). Further-more, it was found that the long incubation of yeast cellswith ligands might promote the ligand-dependent inter-actions between hPPARc-LBD and mCBP in a directratio, as shown in Fig. 2, thereby increasing the sensitiv-ity of our current assay system. During the assay, theovernight culture was diluted until the test culture couldhave an OD600 value between 0.8 and 1.0 after incuba-

Fig. 1. Ligand-dependent interactions between hPPARc-LBD andmCBP. Yeast transformants P1 and P1C2 were precultured overnightin 3 ml liquid medium SD-T or SD-TL independently. Cells werediluted to an initial OD600 value of 0.3 to 0.4 with fresh medium andwere incubated for 5–6 h at 30 �C with different PPARc agonists or0.1% DMSO (as vehicle control). The shown values, obtained via a-galactosidase assays, represent the means ± standard deviations ofthree independent sets of experiments.

tion. Considering the sensitivity of the assay and the fea-sibility of manipulation, we adopted overnightincubation (12–14 h) in our subsequent assays.

Evaluation of the constructed system by PPARctypical agonists

To assess the reliability of the constructed yeast two-hybrid screening system with which we have determinedthe EC50 values of four kinds of PPARc known agon-ists, the results are shown in Table 1 and Fig. 3. In com-parison with the EC50 values of the test compoundsgenerated by PPAR-GAL4 transactivation assays, itwas found that the EC50 values for 15d-PGJ2(0.82 lM) and indomethacin (15.2 lM) determined by

Table 1Summary of EC50 values determined via the constructed yeast two-hybrid system and the reported PPARc-GAL4 transactivation assays

Compound EC50 in this studya (lM) EC50 in referenceb (lM)

GI262570 0.0133 ± 0.000265 0.00034c

Rosiglitazone 6.37 ± 0.130 0.089d

Troglitazone 10.1 ± 0.300 0.54d

Pioglitazone 10.4 ± 0.224 0.59d

15d-PGJ2 0.816 ± 0.100 1e

Indomethacin 15.2 ± 1.20 40f

a Each EC50 value is the mean of three independent sets ofexperiments.

b EC50 values were generated via PPARc-GAL4 transactivationassays.

c Ref. [1].d Ref. [10].e Ref. [21].f Ref. [13].

Fig. 3. Dose-dependent effects of PPARc agonists on the interactionsbetween hPPARc-LBD and mCBP. The P1C2 overnight cultures inSD-LT were diluted to an initial OD600 value at 0.07 with fresh mediaand treated with increasing concentrations of various PPARc agonistsor 0.1% DMSO (as vehicle control) for 14 h at 30 �C. Relative a-galactosidase activity corresponds to the cells treated with 0.1%DMSO. The values shown are the means ± standard deviations fromthree independent experiments.

Fig. 4. GW9662 inhibiting the agonistic effects of rosiglitazone on thehPPARc-LBD/mCBP interaction dose-dependently. The P1C2 over-night cultures in SD-LT were diluted to an initial OD600 value of 0.07with fresh media. Dilutions were pretreated with 20 lM rosiglitazoneand subsequently were exposed to increasing concentrations ofGW9662 at 0.01–100 lM (s) or treated with GW9662 only (d). Afterincubation at 30 �C for 14 h, a-galactosidase assay was performed. Therelative a-galactosidase activity corresponds to the cells treated with0.1% DMSO, and the values are the means ± standard deviations fromthree independent experiments.

Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259 257

the current method are nearly within the same quantita-tive grade with the literature values. However, the EC50

values for troglitazone (10.1 lM) and pioglitazone(10.4 lM) are roughly 19-fold higher than the literatureresults, whereas those values for GI262570 (0.0133 lM)and rosiglitazone (6.37 lM) are observed to be 39- and72-fold higher, respectively, than the literature results(Table 1). Such differences may be tentatively attributedto the ligand type-specific interactions for PPARc andCBP. As stated in the early article by Kodera et al. [3],15d-PGJ2 is more efficient than troglitazone in inducingPPARc interactions with the highly related p300 proteinof CBP in yeast and mammalian two-hybrid assays.Other recent reports even show that TZD compoundswith structural similarity, such as rosiglitazone andpioglitazone, induced distinct modes of interactions withthe coactivator PGC-1 [4] and have disparate actions onhigh-density lipoprotein (HDL) metabolism [31,32]. Onthe other hand, drug penetration and yeast compoundefflux may also be somewhat related to the higherEC50 values obtained from the current built yeast two-hybrid screening approach for GI262570 and TZDs [33].

Now, it is known that the broad variety of biologicalfunctions mediated by PPARc may differ according tothe agonists used [31,32,34,35] and that the expressionlevels of various coactivators in target tissues playimportant roles in the modulation of the agonist activityof PPARc [36,37]. Ligand binding to PPARc inducesthe formation of a unique conformation in the coactiva-tor-binding surface of the receptor, allowing selectivecoactivator recruitment. It has been considered thatthe molecular mechanism underlies the type-specificpharmacological profiles of PPARc agonists [4].

Accordingly, within this respect, our constructed yeasttwo-hybrid screening system might also contribute tosome potential information pertaining to the molecularinteraction mechanism for the gained agonists.

Evaluation of the constructed system by PPARctypical antagonist GW9662

To evaluate the constructed yeast two-hybrid systemin the application of PPARc antagonist screening, weinvestigated the antagonistic feature for PPARc antago-nist GW9662 by use of this system. GW9662 is identifiedas a potent irreversible PPARc ligand and profiles as afunctionally selective PPARc antagonist [38]. As shownin Fig. 4, GW9662 efficiently inhibited the agonistic ef-fects exerted by 20 lM rosiglitazone with an IC50 valueof 1.6 lMbut had little effect when added alone. This ob-tained IC50 value is comparable with the result obtainedbyGupta et al. [39], where GW9662 antagonized the acti-vation response elicited by 1 lM rosiglitazone at concen-trations of 1–5 lM in MS colon carcinoma cell line.Moreover, it was found that the agonistic effect of rosig-litazone could be antagonized to the baseline level byGW9662 in our yeast system (Fig. 4), which further veri-fies the full antagonist characteristic of GW9662 reportedpreviously [38]. Therefore, all of these results shed lighton the feasibility of applying this built yeast two-hybridsystem to PPARc-specific antagonist screening.

In conclusion, in this article we have successfullydeveloped a yeast two-hybrid system for PPARc-spe-

258 Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259

cific agonist and antagonist screening. To our knowl-edge, our work is the first to employ the MEL1 repor-ter gene used in a yeast two-hybrid system designedfor drug screening and quantitative detection of pro-tein–protein interactions through conducting the con-venient a-galactosidase assays. In addition,optimization of the yeast two-hybrid system wasachieved by prolonging the incubation time of yeastcells with test chemicals. This screening platform wasfully evaluated by the agonists from four differentchemical classes tested in this system by inducing thetype-specific interaction of hPPARc-LBD with mCBPdose-dependently. The fact that agonistic effects of ros-iglitazone could be antagonized by GW9662 makes itfeasible to apply this built system to PPARc-specificantagonist screening. Moreover, this system is designedfor use in a microtiter plate format where large num-bers of chemicals can be readily screened; thus, itmight be adaptable for a high-throughput system. Fi-nally, for the sake of the homological cases, this sys-tem might be modified to find possible applicationsin screening ligands of other PPAR subtypes such asPPARa and PPARd.

Acknowledgments

This work was supported by the State Key Programof Basic Research of China (2002CB512801,2002CB512802, and 2002CB512807), the National Nat-ural Science Foundation of China (20372069), theShanghai Basic Research Project from the Shanghai Sci-ence and Technology Commission (02DJ14070,03DZ19222, and 03DZ19212), and the 863 Hi-TechProgram (2002AA233011, 2001AA235030, and2001AA235071).

References

[1] T.M. Willson, P.J. Brown, D.D. Sternbach, B.R. Henke, ThePPARs: from orphan receptors to drug discovery, J. Med. Chem.43 (2000) 527–550.

[2] G. Murphy, G.J. Holder, PPAR-? agonists: therapeutic role indiabetes, inflammation, and cancer, Trends Pharmacol. Sci. 21(2000) 469–474.

[3] Y. Kodera, K. Takeyama, A. Murayama, M. Suzawa, Y.Masuhiro, S. Kato, Ligand type-specific interactions of peroxi-some proliferator-activated receptor gamma with transcriptionalcoactivators, J. Biol. Chem. 275 (2000) 33201–33204.

[4] Y. Wu, W.W. Chin, Y. Wang, T.P. Burris, Ligand and coactivatoridentity determines the requirement of the charge clamp forcoactivation of the peroxisome proliferator-activated receptor, J.Biol. Chem. 278 (2003) 8637–8644.

[5] L. Nagy, P. Tontonoz, J.G. Alvarez, H. Chen, R.M. Evans,Oxidized LDL regulates macrophage gene expression throughligand activation of PPARc, Cell 93 (1998) 229–240.

[6] B. Hulin, P.A. McCarthy, E.M. Gibbs, The glitazone family ofantidiabetic agents, Curr. Pharma. Des. 2 (1996) 85–102.

[7] J.M. Lehmann, L.B. Moore, T.A. Smith-Oliver, W.O. Wilkison,T.M. Willson, S.A. Kliewer, An antidiabetic thiazolidinedione isa high affinity ligand for peroxisome proliferator-activated recep-tor c (PPARc), J. Biol. Chem. 270 (1995) 12953–12956.

[8] B.M. Spiegelman, PPAR-c: adipogenic regulator and thiazolidin-edione receptor, Diabetes 47 (1998) 507–514.

[9] B.B. Lowell, PPARc: an essential regulator of adipogenesis andmodulator of fat cell function, Cell 99 (1999) 239–242.

[10] B.R. Henke, S.G. Blanchard, M.F. Brackeen, K.K. Brown, J.E.Cobb, J.L. Collins, W.W. Harrington Jr., M.A. Hashim, E.A.Hull-Ryde, I. Kaldor, S.A. Kliewer, D.H. Lake, L.M. Leesnitzer,J.M. Lehmann, J.M. Lenhard, L.A. Orband-Miller, J.F. Miller,R.A. Mook, Jr., S.A. Noble, W. Oliver Jr., D.J. Parks, K.D.Plunket, J.R. Szewczyk, T.M. Willson, N-(2-Benzoylphenyl)-LL-tyrosine PPARc agonists: I. Discovery of a novel series of potentantihyperglycemic and antihyperlipidemic agents, J. Med. Chem.41 (1998) 5020–5036.

[11] J.L. Collins, S.G. Blanchard, G.E. Boswell, P.S. Charifson, J.E.Cobb, B.R. Henke, E.A. Hull-Ryde, W.M. Kazmierski, D.H.Lake, L.M. Leesnitzer, J. Lehmann, J.M. Lenhard, L.A. Orband-Miller, Y. Gray-Nunez, D.J. Parks, K.D. Plunkett, W.Q. Tong,N- (2-Benzoylphenyl)-LL-tyrosine PPARc agonists: II. Structure–activity relationship and optimization of the phenyl alkyl ethermoiety, J. Med. Chem. 41 (1998) 5037–5054.

[12] J.E. Cobb, S.G. Blanchard, E.G. Boswell, K.K. Brown, P.S.Charifson, J.P. Cooper, J.L. Collins, M. Dezube, B.R. Henke,E.A. Hull-Ryde, D.H. Lake, J.M. Lenhard, W. Oliver Jr., J.Oplinger, M. Pentti, D.J. Parks, K.D. Plunket, W.Q. Tong, N-(2-Benzoylphenyl)-LL-tyrosine PPARc agonists: III. Structure–activ-ity relationship and optimization of the N-aryl substituent, J.Med. Chem. 41 (1998) 5055–5069.

[13] J.M. Lehmann, J.M. Lenhard, B.B. Oliver, G.M. Ringold, S.A.Kliewer, Peroxisome proliferator-activated receptors alpha andgamma are activated by indomethacin and other non-steroidalanti-inflammatory drugs, J. Biol. Chem. 272 (1997) 3406–3410.

[14] J.M. Malinowski, S. Bolesta, Rosiglitazone in the treatment oftype 2 diabetes mellitus: a critical review, Clin. Ther. 22 (2000)1151–1168.

[15] N. Kubota, Y. Terauchi, H. Miki, H. Tamemoto, T. Yamauchi,K. Komeda, S. Satoh, R. Nakano, C. Ishii, T. Sugiyama, K.Eto, Y. Tsubamoto, A. Okuno, K. Murakami, H. Sekihara, G.Hasegawa, M. Naito, Y. Toyoshima, S. Tanaka, K. Shiota, T.Kitamura, T. Fujita, O. Ezaki, S. Aizawa, T. Kadowaki, PPARcmediates high-fat diet-induced adipocyte hypertrophy and insulinresistance, Mol. Cell 4 (1999) 597–609.

[16] P.D. Miles, Y. Barak, W. He, R.M. Evans, J.M. Olefsky,Improved insulin sensitivity in mice heterozygous for PPAR-cdeficiency, J. Clin. Invest. 105 (2000) 287–292.

[17] S.S. Deeb, L. Fajas, M. Nemoto, J. Pihlajamaki, L. Mykkanen,J. Kuusisto, M. Laakso, W. Fujimoto, J. Auwerx, A Pro12Alasubstitution in PPARc2 associated with decreased receptoractivity, lower body mass index, and improved insulin sensitivity,Nat. Genet. 3 (1998) 284–287.

[18] K. Hara, T. Okada, K. Tobe, K. Yasuda, Y. Mori, H. Kadowaki,R. Hagura, Y. Akanuma, S. Kimura, C. Ito, T. Kadowaki, ThePro12Ala polymorphism in PPARc2 may confer resistance to type2 diabetes, Biochem. Biophys. Res. Commun. 271 (2000) 212–216.

[19] D. Altshuler, J.N. Hirschhorn, M. Klannemark, C.M. Lindgren,M.C. Vohl, J. Nemesh, C.R. Lane, S.F. Schaffner, S. Bolk, C.Brewer, T. Tuomi, D. Gaudet, T.J. Hudson, M. Daly, L. Groop,E.S. Lander, The common PPARc Pro12Ala polymorphism isassociated with decreased risk of type 2 diabetes, Nat. Genet. 26(2000) 76–80.

[20] P.W. Young, D.R. Buckle, B.B. Cantello, H. Chapman, J.C.Clapham, P.J. Coyle, D. Haigh, R.M. Hindley, J.C. Holder, H.Kallender, A.J. Latter, K.W. Lawrie, D. Mossakowska, G.J.Murphy, L.R. Cox, S.A. Smith, Identification of high-affinity

Q. Chen et al. / Analytical Biochemistry 335 (2004) 253–259 259

binding sites for the insulin sensitizer rosiglitazone (BRL-49653)in rodent and human adipocytes using a radioiodinated ligand forperoxisomal proliferator-activated receptor gamma, J. Pharmacol.Exp. Ther. 284 (1998) 751–759.

[21] G. Krey, O. Braissant, F.L. Horset, M. Perroud, E. Kalkhoven,M. Parker, W. Wahli, Fatty acids, eicosanoids, and hypolipidemicagents identified as ligands of peroxisome proliferator-activatedreceptors by coactivator-dependent receptor ligand assay, Mol.Endocrinol. 11 (1997) 779–791.

[22] J.S. Nicholsa, D.J. Parksa, T.G. Conslerb, S.G. Blancharda,Development of a scintillation proximity assay for peroxisomeproliferator-activated receptorc ligand binding domain, Anal.Biochem. 257 (1998) 112–119.

[23] C.Y. Yu, L.L. Chen, H.B. Luo, J. Chen, F. Cheng, C.S. Gui,R.H. Zhang, J.H. Shen, K.X. Chen, H.L. Jiang, X. Shen, Bindinganalyses between human PPARc-LBD and ligands surface plas-mon resonance biosensor assay correlating with circular dichroicspectroscopy determination and molecular docking, Eur. J.Biochem. 271 (2004) 386–397.

[24] D.P. McDonnell, E. Vegeto, M.A.G. Gleeson, Nuclear hormonereceptors as targets for new drug discovery, Biotechnology 11(1993) 1256–1261.

[25] T. Taniguchi, J. Mizukami, Method for identifying or screeningagonist and antagonist to PPAR, U.S. Patent 6,365,361 (2002).

[26] R.D. Gietz, R.A. Woods, Transformation of yeast by lithiumacetate/single-stranded carrier DNA/polyethylene glycol method,Methods Enzymol. 350 (2002) 87–96.

[27] P.S. Lazo, A.G. Ochoa, S. Gascon, a-Galactosidase fromSaccharomyces carlsbergensis, cellular localization, and purifica-tion of the external enzyme, Eur. J. Biochem. 77 (1977) 375–382.

[28] J. Yanagisawa, Y. Yanagi, Y. Masuhiro, M. Suzawa, T. Toriyabe,K. Kashiwagi, M. Watanabe, M. Kawabata, K. Miyazono, S.Kato, Convergence of transforming growth factor-b and vitaminD signaling pathways on SMAD transcriptional coactivators,Science 283 (1999) 1317–1321.

[29] K. Eguchi, M. Ozawa, Y.S. Endoh, J. Nishikawa, T. Nishihara,K. Goto, H. Yoshimura, Validity test for a yeast two-hybridassay to screen for estrogenic activity and its application toinsecticides and disinfectants for veterinary use, Bull. Environ.Contam. Toxicol. 70 (2003) 226–232.

[30] E. Dornstauder, E. Jisa, I. Unterrieder, L. Krenn, W. Kubelka,A. Jungbauer, Estrogenic activity of two standardized red cloverextracts intended for large scale use in hormone replacementtherapy, J. Steroid Biochem. Mol. Biol. 78 (2001) 67–75.

[31] P.J. Boyle, A.B. King, L. Olansky, A. Marchetti, H. Lau, R.Magar, J. Martin, Effects of pioglitazone and rosiglitazone onblood lipid levels and glycemic control in patients with type 2diabetes mellitus: a retrospective review of randomly selectedmedical records, Clin. Ther. 24 (2002) 378–396.

[32] C.G. Gegick, M.D. Altheimer, Comparison of effects of thiazo-lidinediones on cardiovascular risk factors: observations from aclinical practice, Endocrinol. Pract. 7 (2001) 162–169.

[33] B. Rogers, A. Decottignies, M. Kolaczkowski, E. Carvajal, E.Balzi, A. Goffeau, The pleitropic drug ABC transporters fromSaccharomyces cerevisiae, J. Mol. Microbiol. Biotechnol. 3 (2001)207–214.

[34] M. Adams, C.T. Montague, J.B. Prins, J.C. Holder, S.A. Smith,L. Sanders, J.E. Digby, C.P. Sewter, M.A. Lazar, V.K. Chatterjee,S. O�Rahilly, Activators of peroxisome proliferator-activatedreceptor-c have depot-specific effects on human preadipocytedifferentiation, J. Clin. Invest. 100 (1997) 3149–3153.

[35] B. Lecka-Czernik, E.J. Moerman, D.F. Grant, J.M. Lehmann,S.C. Manolagas, R.L. Jilka, Divergent effects of selective peroxi-some proliferator-activated receptor-c2 ligands on adipocyteversus osteoblast differentiation, Endocrinology 143 (2002)2376–2384.

[36] N. Takahashi, T. Kawada, T. Yamamoto, T. Goto, A. Taimatsu,N. Aoki, H. Kawasaki, K. Taira, K.K. Yokoyama, Y. Kamei, T.Fushiki, Overexpression and ribozyme-mediated targeting oftranscriptional coactivators CREB-binding protein and p300revealed their indispensable roles in adipocyte differentiationthrough the regulation of peroxisome proliferator-activatedreceptor, J. Biol. Chem. 277 (2002) 16906–16912.

[37] G. Castillo, R.P. Brun, J.K. Rosenfield, S. Hauser, C.W. Park,A.E. Troy, M.E. Wright, B.M. Spiegelman, An adipogeniccofactor bound by the differentiation domain of PPARc, EMBOJ. 18 (1999) 3676–3687.

[38] L.M. Leesnitzer, D.J. Parks, R.K. Bledsoe, J.E. Cobb, J.L.Collins, T.G. Consler, R.G. Davis, E.A. Hull-Ryde, J.M.Lenhard, L. Patel, K.D. Plunket, J.L. Shenk, J.B. Stimmel, C.Therapontos, T.M. Willson, S.G. Blanchard, Function conse-quences of cysteine modifcation in the ligand binding sites ofperoxisome proliferator activated receptors by GW9662, Bio-chemistry 41 (2002) 6640–6650.

[39] R.A. Gupta, J.A. Brockman, P. Sarraf, T.M. Willson, R.N.DuBois, Target genes of peroxisome proliferator-activated recep-tor gamma in colorectal cancer cells, J. Biol. Chem. 276 (2001)29681–29687.