PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
ChemMedChem. Author manuscript; available in PMC Jan 31, 2011.
Published in final edited form as:
PMCID: PMC3031428
NIHMSID: NIHMS217654
Derivation of a Retinoid X Receptor Scaffold from Peroxisome Proliferator-Activated Receptor γ Ligand 1-Di(1H-indol-3-yl)methyl-4-trifluoromethylbenzene
Marcia I. Dawson,*a Mao Ye,a Xihua Cao,a Lulu Farhana,b Qiong-Ying Hu,a Yong Zhao,c Li Ping Xu,b Alice Kiselyuk,d Ricardo G. Correa,a Li Yang,a Tingjun Hou,e John C. Reed,a Pamela Itkin-Ansari,d Fred Levine,d Michel F. Sanner,c Joseph A. Fontana,b and Xiao-Kun Zhanga
aCancer Center, Burnham Institute for Medical Research, 10901 North Torrey Pines Rd., La Jolla, CA 92037 (USA)
bDepartment of Veterans Affairs, John D. Dingell Medical Center, and Wayne State University School of Medicine, 4646 John R St., Detroit, MI 48201-1932 (USA)
cDepartment of Molecular Biology, The Scripps Research Institute, 10550 North Torrey Pines Rd., La Jolla, CA 92037 (USA)
dSanford Children’s Health Research Center, Burnham Institute for Medical Research, 10901 North Torrey Pines Rd., La Jolla, CA 92037 (USA)
eDepartment of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0359 (USA)
Fax: (+1) 858-646-3197, mdawson/at/burnham.org
1-Di(1H-indol-3-yl)methyl-4-trifluoromethylbenzene (DIM-Ph-4-CF3) is reported to inhibit cancer cell growth and to act as a transcriptional agonist of peroxisome proliferator-activated receptor γ (PPARγ) and nuclear receptor 4A subfamily member 1 (NR4A1). In addition, DIM-Ph-4-CF3 exerts anticancer effects independent of these receptors because PPARγ antagonists do not block its inhibition of cell growth, and the small pocket in the NR4A1 crystal structure suggests no ligand can bind. Because PPARγ and NR4A1 heterodimerize with retinoid X receptor (RXR), and several PPARγ ligands transcriptionally activate RXR, DIM-Ph-4-CF3 was investigated as an RXR ligand. DIM-Ph-4-CF3 displaces 9-cis-retinoic acid from RXRα but does not transactivate RXRα. Structure-based design using DIM-Ph-4-CF3 as a template led to the RXRα transcriptional agonist (E)-3-[5-di(1-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylic acid. Its docked pose in the RXRα ligand binding domain suggests that binding is stabilized by interactions of its carboxylate group with arginine 316, its indoles with cysteines 269 and 432, and its 1-methyl groups with hydrophobic residues lining the binding pocket. As is expected of a selective activator of RXRα, but not of RARs and PPARγ, this RXRα agonist, unlike DIM-Ph-4-CF3, does not appreciably decrease cancer cell growth or induce apoptosis at pharmacologically relevant concentrations.
Keywords: antitumor agents, dimethylarenes, receptors, retinoids, RXR, TR3
Only one retinoid X receptor (RXR)-selective retinoid, bexarotene,[14] is approved for cancer treatment.[5] Despite evaluation in clinical trials alone or in combination with other anticancer drugs, dose-limiting toxicities[6, 7] or lack of objective responses[810] limit its use to oral and topical treatment of cutaneous T-cell lymphoma.[11] In addition to transcriptionally activating RXR nuclear receptors, bexarotene can transactivate retinoic acid receptors (RARs) on trans-retinoic acid (trans-RA)-sensitive response elements (RAREs)[1] and permissive nuclear receptors such as peroxisome proliferator-activated receptors (PPARs) heterodimerized with RXR on PPAR response elements (PPREs).[12] Thus, to improve therapeutic indices, the search to identify a retinoid selective for RXR in the context of its heterodimer with a specific nuclear receptor has continued.
The orphan nuclear receptor (NR) family member 4A subtype 1 (NR4A1; human TR3, mouse Nur77, rat NGFI-Bα) exhibits genomic activity as a transcription factor in the cell nucleus[13, 14] and nongenomic activity as a cytosolic mediator of apoptosis. In the cytosol, it converts Bcl-2 function from mitochondrial protection to apoptosis induction.[1517] Recently, Safe and co-workers reported that several 4-substituted di(1H-indol-3-yl)methylbenzenes (DIM-arenes) induce colon and prostate cancer cell apoptosis and decrease cancer xenograft growth in animal models.[18] Using reporter assays in these cell lines and NR co-activator recruitment assays, they correlated NR4A1/TR3 transcriptional activation by DIM-arenes with their abilities to inhibit cell proliferation. The most potent transactivators were DIM-Ph-4-CF3 and DIM-Ph-4-OMe (1 and 2 in Figure 1, respectively). In contrast, DIM-Ph-4-OH (3) was inactive at concentrations up to 20 µm on the GAL4–Nur77(domain E/F) chimeric receptor bound to the pGAL4–LUC reporter construct, but antagonized transactivation induced by 1 or 2.
Figure 1
Figure 1
Structures of DIM-Ph-4-CF3 (1), DIM-Ph-4-OMe (2), DIM-Ph-4-OH (3), analogues 414, and RXRα agonist SR11237/BMS649 (15).
Reports on other NR4A subfamily members did not support small-molecule binding by the NR4A ligand binding domain (LBD or E domain), but suggested that NR4A interaction with small molecules was atypical: 1) Mutational studies showed that, rather than the LBD, the N-terminus or AB domain of NR4A subgroup members NR4A2/Nurr1 and NR4A3/NOR1, which contains the ligand-independent activation function-1 (AF-1), is responsible for transcriptional activation induced by a 6-thiopurine metabolite.[1921] 2) Although prostaglandin A2 transactivates NR4A3/NOR1 and interacts with the NR4A3/NOR1 LBD, its low affinity and AF-1 domain requirement suggested indirect effects.[22] 3) NMR studies suggested a unique binding surface on the NR4A2/Nurr1 LBD by showing that peptides derived from transcriptional co-repressors SMRT and NCoR and the NR4A2/Nurr1 N-terminal peptide (residues 1–13) interact with a hydrophobic surface located between helices H11 and H12.[23] 4) Other NMR studies suggested that several benzimidazole and isoxazolopyridinone NR4A2/Nurr1 activators, which were identified in reporter assays,[24] interact with a surface formed from regions of its LBD helices H4, H11, and H12.[25]
Moreover, crystallography of the NR4A2/Nurr1 LBD revealed that the region occupied by the ligand binding pocket (LBP) is filled with hydrophobic amino acid side chains.[26] The sequence identity of ~60% among the NR4A LBDs[21] suggests that the NR4A1/TR3 and NR4A3/NOR1 LBDs would have similar tertiary structures. Subsequently, crystal structures of the NR4A1/NGFI-B LBD[27] and NR4A1/TR3 LBD[28] confirmed the absence of accessible LBPs. For example, the latter structure (PDB code: 2QW4) revealed that the LBP region is predominantly occupied by hydrophobic side chains of H3 Phe75, H5 Phe112, β1–β2 loop Phe133, and H7 Phe148 and Tyr151.[28] These results suggest the absence of a typical NR LBP, and reports that, in addition to acting as a monomer, homodimer, and NR4A1/TR3–NR4A2/Nurr1 heterodimer, NR4A1/TR3 and NR4A2/Nurr1 heterodimerize with RXR[2932] suggested that DIM-Ph-4-CF3 (1) could function as an RXR ligand in the context of the RXR–NR4A1/TR3 heterodimer or another heterodimer on an RXR response element. Our premise was partly supported by reports that: 1) Although 1 functions as a PPARγ agonist in PPARγ–GAL4/pGAL4–LUC reporter assays,[33] it decreases the survival of several cancer cell lines in the presence of a PPARγ transcriptional antagonist.[34, 35] 2) Natural products that are PPARγ or PPARα transcriptional agonists—β-apo-14′-carotenal,[36] arachidonic acid,[37] docosahexaenoic acid,[38] and phytanic acid[39]—are also RXR agonists.[36, 4042] On this basis, if 1 were an RXR ligand, it could provide a new template for RXR ligand design. Herein we report the design, synthesis, RXRα LBD binding affinities, and biological activities of several 1-derived analogues (compounds 414 in Figure 1) in comparison with 1 and 3. Interestingly, 1 binds to the RXRα LBD but is unable to transcriptionally activate the RXRα homodimer on the TREpal or the RXRα–NR4A1/TR3 heterodimer on the βRARE in reporter assays.
Design and synthesis
Analogue design was based on docking of DIM-Ph-4-CF3 (1) to the RXRα LBD in the crystal structure (PDB code: 1MVC) of its complex with transcriptional agonist 2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)-2-(4-carboxyphenyl)-1,3-dioxolane (BMS649, 15),[43] first reported as SR11237.[44] The docked pose for 1 (Figure 2A) reveals that dihedral angles formed by the indole ring C2 and C3 atoms, the methine carbon, and the 1-aryl carbon permitted us to superpose 1 and 15 in the LBP. The indole ring of 1, with a dihedral angle of 47°, resides in same region of the LBP as the 5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl ring of 15, whereas the indole ring, with a dihedral angle of −100°, is adjacent to the 1,3-dioxolane ring of 15. These orientations cause the phenyl rings of 1 and 15 to overlap. In this docked pose, the electron-donating CF3 group of 1 is near the CO2H group of 15 (C–C: d=2.0 Å). Although 1 lacks the ability to form robust ionic and hydrogen bonding interactions with the guanidinium NH groups of Arg316 in helix H5 and the β sheet backbone NH group of Ala327, as observed for the carboxyl groups of 15 and 9-cis-retinoic acid (9-cis-RA) in the RXRα LBD,[43, 45] the distances (4–6 Å) between the guanidinium nitrogen atoms and the fluorine atoms of CF3 suggest that binding could be stabilized by N–H–F bonding if the Arg316 side chain shifts during formation of the complex between RXRα LBD and compound 1.
Figure 2
Figure 2
Docking of DIM-arenes into the RXRα ligand binding pocket (LBP) as found in the crystal structure of recombinant human RXRα LBD complexed with 15 (PDB code: 1MVC). A) Docked low-energy pose of DIM-Ph-4-CF3 (1, yellow C atoms) overlaps (more ...)
The docked pose for 1 suggests that DIM-arenes might assist in suppressing cancer cell growth by functioning as RXR ligands in RXR heterodimers with PPARγ, NR4A1/TR3, or another NR. To validate this hypothesis, we determined whether 1 could bind to RXRα. Next, because 1 was shown to have anticancer effects at 10–20 µm,[18] we employed structure-based design to identify analogues that could have higher affinities for RXRα LBD. The overlap of 1 with 15 (Figure 2A) suggested that binding affinity might be improved if the 4-CF3 group of 1 were replaced by CO2H, as found in analogue 4, whereas deletion of the phenyl ring from 4, as in 6, was expected to decrease activity. The overlap also suggested that alkylation at the 1-position of the indoles, as in 5, could enhance hydrophobic contacts with residues lining the LBP. To determine whether interactions between the oxygen atoms of CO2H and the guanidinium NH groups of Arg316 would be enhanced by decreasing the inter-group distance, the Ph-4-CO2H of 4 was replaced by Ph-3-(E-CH=CHCO2H), as in 7 and 8, and by 2-thienyl-5-(E-CH=CHCO2H), as in 914. The effects of increasing LBP contact were assessed by 1014, having a hydrophobic methyl group at the 1-, 5-, or 7-indole ring position. All compounds except 6 assumed similar poses on docking to the RXRα LBD in that their polar termini were directed toward the Arg316 guanidinium group and their indoles faced the hydrophobic portion of the LBP. In contrast, the pose for 6 had a reverse orientation.
Compounds were synthesized as outlined in Scheme 1. The DIM moiety was introduced by electrophilic substitution reactions of indoles 1621 with 4-substituted benzaldehydes 2224, ethyl glyoxalate (27), 3-substituted benzaldehyde 30, or 5-substituted thiophene-2-carboxaldehyde 34 using a procedure described by Safe and co-workers,[18] namely, the cerium chloride and sodium iodide on silica gel catalyst developed by Bartoli et al.[46] Wittig mono-olefinations of 1,3-phthalaldehyde (29) and thiophene-2,5-dicarboxaldehyde (33) introduced the 3- and 5-(E-CH=CHCO2Et) groups of 30 and 34, respectively. While 1 was produced directly, base hydrolysis was used to remove the ester protecting groups of 25, 26, 28, 31, 32, and 3540 to provide 414, respectively. Phenol 3 was prepared by condensation of indole (16) and 4-hydroxybenzaldehyde (23) using iodine in acetonitrile as the catalyst.[47]
Scheme 1
Scheme 1
Synthesis of DIM-arenes 1 and 314. Reagents and conditions: a) 22, 24, 27, 30, 34: CeCl3·7H2O–NaI–SiO2, MeCN; b) 23: I2, MeCN; c) LiOH·H2O, MeOH, reflux; HCl(aq); d) Ph3P=CHCO2Et, PhMe, 80°C.
Competitive binding to the RXRα LBD
In competitive binding assays with RXR transcriptional agonist [11,12-3H2]9-cis-RA, the DIM-arenes 1, 4, and 10 most effectively displaced bound label from recombinant human RXRα LDB (Table 1). On the basis of IC50 values, 10 had the highest affinity for RXRα, whereas 1 and 4 had 53 and 17% of the affinity, respectively. Analogues 5, 7, 8, and 12 had IC50 values in the mid-micromolar range.
Table 1
Table 1
Effects of substituted DIM-arene analogues on competitive binding to RXRα and transcriptional activation of RXRα and/or NR4A1/TR3 on the TREpal, βRARE, and NurRE response elements.
Docking was conducted to understand how 4 and 10 could interact with RXRα LBP residues (Figure 2 B). The indole rings in the docked poses for 1, 4, and 10 reside in the same regions of the LBP, whereas superposing with RXR ligand 15 suggests that the CO2H carbon atoms of 4 and 10 are closer to that of 15 (0.7 and 0.5 Å, respectively). The N–O interatomic distances between the guanidinium group of Arg316 and the CO2H group of the ligand measured for 4 (3.4–4.6 Å) and 10 (2.7–3.4 Å) suggest why both DIM-arenes bind to RXRα and why 10, for which these distances are shorter, has higher affinity.
Most interesting in the docked pose for 1 is the proximity of one indole ring nitrogen to the sulfur atom of Cys269 on H3, and the other indole nitrogen atom to the sulfur of Cys432 on H11 (3.5 and 4.4 Å, respectively), which suggests possible N–H–S bonding interactions. The proximity between the cysteine S atoms and indole N atoms was recapitulated in the docked poses for 10 and 4 (S–N interatomic distances: 10, 3.5–4.2 Å; 4, 3.3–9.8 Å). The docked pose for 10 also suggests potential hydrophobic interactions between the 1-methyl group of one indole and a methyl group of H3 Val265, and the other 1-methyl group with alkyl side chains of H7 Ile345 and H11 Ile428. Such hydrophobic stabilizing interactions are absent in docked poses for 1 and 4.
Although the structural similarity of 9 to 10 suggests that 9 would bind RXRα with similar potency, compound 9, at concentrations as high as 100 µm, could not displace radiolabeled 9-cis-RA from the RXRα LBD. While the docked pose for 9 shows its CO2H group almost coincident with that of 10 and one indole N atom 3.5 Å from the Cys432 S atom, the second indole N is 5.5 Å from Cys432 S and 5.9 Å from C269 S. In addition, the second indole and the thiophene of 9 are almost orthogonal to those rings in the docked pose for 10. Thus, the decreased N–H–S interactions and dissimilar docking poses suggest why 9 does not displace 9-cis-RA from RXRα. Similarly, the docked pose for 3 suggests that the O–N interatomic distances of 4.9–5.1 Å between its phenolic oxygen atom and the NH groups of Arg316 would not permit robust N–H–O interactions required for high-affinity binding.
Activation of RXRα and NR4A1/TR3 in reporter assays
Reporter assays were conducted to determine whether the DIM-arenes transactivate RXRα or NR4A1/TR3 (Figure 3). CV-1 kidney cells were transfected with the nr4a1/tr3 gene plasmid and the everted repeat NurRE–thymidine kinase–chloramphenicol acetyl transferase (NurRE–tk–CAT) reporter, which is activated by the NR4A1/TR3 homodimer;[32] the rxrα gene construct and the palindromic TREpal–tk–CAT reporter, which is activated by an RXR homodimer–agonist complex; or with genes for both receptors and the DR-5 βRARE–CAT reporter, which we found to be activated by the RXR–NR4A1/TR3 heterodimer–RXR agonist complex.[30] Cells were then treated with a DIM-arene: 1 and 314 at 20 µm, which was the highest concentration of 1 and 2 used by Safe and co-workers to demonstrate NurRE–LUC reporter activation.[18] In this group of compounds, only 10 at 1.0 µm effectively activated RXRα on the TREpal (63% of that produced by 9-cis-RA at 0.1 µm, or 7-fold above the vehicle-treated control; Figure 3A and Table 1). The other analogues were essentially inactive. Activation by 10 at 20 µm rose to 1.9-fold that of 0.1 µm 9-cis-RA, whereas the other DIM-arenes only weakly activated (16–36%) or were inactive (Table 1). Although capable of binding to RXRα in the micromolar range, compound 1 did not activate the RXRα homodimer on the TREpal. Because PPAR target genes containing PPREs have been reported to be activated in vivo by RXR homodimers in the presence of 9-cis-RA,[48] the converse may occur so that 1 could function as an RXR transcriptional antagonist in certain contexts to prevent PPRE activation by RXR homodimers.
Figure 3
Figure 3
Effects of 1,1′-Me2-DIM-2-thienyl-5-(E-CH=CHCO2H) (10) and other DIM-arenes on RXRα and NR4A1/TR3 transcriptional activation of reporter constructs. A) Fold-activation of RXRα on the TREpal–tk–CAT reporter construct (more ...)
At 20 µm, neither 1 nor its analogues significantly activated the NR4A1/TR3 homodimer on the NurRE above the basal level (termed 100%) in nr4a1/tr3 plus NurRE–tk–CAT-transfected CV-1 kidney cells (Figure 3 B). The weak decrease (30%) in basal activation of NR4A1/TR3 by 3 and 10 suggested that in the context of this homodimer both function as weak antagonists either directly by binding to NR4A1/TR3 or indirectly by binding to another NR that preferentially recruits NR4A1/TR3 co-activators. The antagonism demonstrated by 3 partially supports the results reported by Safe and colleagues, that compound 3 at 20 µm, while unable to activate endogenous NR4A1/Nur77 on the NurRE or the chimeric GAL4–Nur77 receptor on the pGAL4–LUC in transfected pancreatic cancer cells, completely blocked activation of the latter reporter induced by 20 µm 1 or 2.[18] The differences between the present results and those previously reported[18] may be due to differences among cell lines or the possibility that endogenous PPARγ or another NR was also activated by the DIM-arenes.
Earlier, the research group of Zhang established that the βRARE response element, which is typically activated by binding of an RXR–RARα heterodimer–RARα agonist complex, was also activated by an RXRα–NR4A1/TR3 heterodimer–RXR agonist complex.[30] Thus, in CV-1 cells transfected with plasmids for both rxrα and nr4a1/tr3 genes and the βRARE–tk–CAT reporter construct, compound 10 at 20 µm was the most potent transactivator (83% that of 0.1 µm 9-cis-RA) followed by 3 and 79 (47–56%) (Table 1). Least active were compounds 1, 11, and 12.
DIM-arene 10 was next examined for its abilities to transactivate the RARs on the TREpal in CV-1 cells. Although it does not significantly transactivate any RAR subtype alone (Figure 4A–C) and does not significantly enhance the activation of RARα (Figure 4A) or RARβ (Figure 4B) induced by 1.0 µm trans-RA, compound 10 at 1.0 and 10 µm does enhance trans-RA-induced RARγ activation (1.28- and 1.54-fold, respectively; Figure 4 C). At 10 µm, 10 suppresses RARβ basal activation (29% decrease) and very weakly inhibits (5% decrease) that induced by trans-RA.
Figure 4
Figure 4
Effects of 1,1′-Me2-DIM-2-thienyl-5-(E-CH=CHCO2H) (10) on RAR and PPARγ transcriptional activation of their reporter constructs. Fold-activation of A) RARα, B) RARβ, and C) RARγ on the TREpal–tk–CAT (more ...)
DIM-arenes 1, 3, and 10 were then examined for PPARγ transactivation on the PPRE (Figure 4D). The RXR–PPAR heterodimer on a PPRE can be activated by an agonist binding to RXR or PPAR or by both agonists binding, in which case activation can be enhanced.[4952] Analogues 1 and 3 behaved as transcriptional agonists, with 1 being the more potent, but at 20 µm, compound 1 is cytotoxic. When compared on the basis of fold-activation above that of the PPRE-transfected vehicle-alone control, 1, at 10 µm, was as potent (18.3-fold) as the PPARγ agonist troglitazone at 20 µm (15.6-fold) and 3 at 40 µm (18.5-fold) in the PPRE and PPARγ co-transfected cells. In contrast, 10 behaved as a weak partial agonist of the RXRα–PPARγ heterodimer on the PPRE (4.5-fold activation above the PPRE control at 2.0 µm and 3.5-fold at 20 µm). At 40 µm, 10 was cytotoxic.
Growth inhibition and apoptosis induction
We next investigated the abilities of the DIM-arenes shown in Figure 1 to inhibit the proliferation of androgen-dependent retinoid-sensitive LNCaP and androgen-independent retinoid-resistant DU-145 prostate cancer cells, and drug-resistant KG-1 acute myelocytic leukemia (AML) cells. According to Western blot analysis, LNCaP cells express NR4A1/TR3 protein.[18] Treatment of LNCaP cells with compound 1 at 10 or 20 µm for 48 h was reported to inhibit cell growth[35] and to induce apoptosis.[18] Values for 50% growth inhibition by 1 and 314 were determined under similar conditions except that cells were grown in 10% delipidized serum (Table 2). On the basis of IC50 values after treatment for 48 h, compound 12 (IC50=2.4 µm) was the most potent inhibitor of LNCaP cell growth followed by 1 (IC50=13 µm), whereas 11 was less active (IC50 value 3.6-fold higher than that of 12). Analogues 5, 7, and 9 had IC50 values in the 20–40 µm range, whereas 20 µm 6 and 10 had only low inhibitory activity, and 3 was inactive up to 100 µm.
Table 2
Table 2
Effects of substituted DIM-arene analogues on inhibition of prostate cancer and leukemia cell growth and induction of leukemia cell apoptosis.
DIM-arenes 1, 3, 10, 11, and 14 inhibited DU-145 cell growth with IC50 values in the 40–90 µm range. Analogues 5, 6, and 12 were least active (~20% growth inhibition at 100 µm). Only compound 1 inhibited KG-1 AML cell growth at 48 h (IC50= 15 µm and 81% inhibition at 20 µm). The similar EC50 value (23 µm) for KG-1 apoptosis induction by 1 at 48 h suggests that apoptosis was a major contributor to growth inhibition.
Because of its relative potency in inhibiting cancer cell growth, the efficacy of 1 was assessed against other human cancer cell lines: radiation- and cisplatin-resistant HeLa-S cervical adenocarcinoma; androgen- and anchorage-independent, FAS- and cyclophosphamide-resistant metastatic PPC-1 prostate adenocarcinoma with defective p53; and TRAIL-, cisplatin-, and radiation-resistant U2OS osteosarcoma. Potency in these tumor cell lines was compared with that in the Ki-ras mutated, 5-fluoro-2′-deoxyuridine-resistant HCT-116 colon cancer cell line. Safe and co-workers reported that 1 decreased HCT-116 cell numbers with an IC50 value of 10 µm when cells were grown in medium containing 2.5% charcoal-stripped fetal bovine serum (FBS).[18] We elected to evaluate the cell lines in the presence of 10% FBS, which would provide more robust growth. As shown in Figure 5, HeLa-S, PPC-1, U2OS, and HCT-116 cell vitalities were decreased by 50% after 22 h with compound 1 at 10.0, 17.6, 12.6, and 9.3 µm, respectively, in comparing ATP levels with those of vehicle-alone-treated controls. These results indicate that even under more optimal growth conditions, 1 effectively inhibits the proliferation of these cancer cell lines by inducing cell death.
Figure 5
Figure 5
DIM-Ph-4-CF3 (1) decreases cancer cell viability as measured by ATP levels. Cells were treated with 1 for 22 h before their ATP levels were assessed as described in the Experimental Section. Viability is expressed as the average of triplicate experiments (more ...)
The results presented herein demonstrate that, in addition to several natural products, a synthetic ligand that activates PPARγ, DIM-Ph-4-CF3 (1), can interact with RXRα. Although unable to potently transactivate RXRα on the TREpal and βRARE, 1 was structurally modified by using docking to enhance pocket contacts to afford the higher-affinity RXRα ligand (1.9-fold binding enhancement) and transcriptional agonist 1,1′-Me2-DIM-2-thienyl-5-(E-CH=CHCO2H) (10), which at 1.0 µm, increased activation of the RXRα homodimer on the TREpal eightfold above the basal level. Similarly, structural modification of RAR-selective agonist 4-[3-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)phenyl]benzoic acid (AC50 values: RARα, 0.2 nm; RARβ, 4.3 nm; RARγ, 2.3 nm; RXRα 2.3 µm)[53] by replacing its benzoic acid terminus by (2E,4E)-2-methyl-2,4-pentadienoic acid decreased RAR activation at 10 µm (RARα, 5%; RARβ, 30%; RARγ, 68% and AC50 4.5 µm) but enhanced that of RXRα (93% and AC50 1.5 µm). These combined results suggest that increasing ligand flexibility next to its CO2H group to allow a decrease in the O–N distance between the CO2H group and Arg316 enhances both RXR selectivity and transactivational activity.
Although RXR ligands terminated by phenyl-3-(E-CH=CHCO2H) and 2-furyl-5-(E-CH=CHCO2H) groups and equipped with an alkyl-substituted sp3 carbon bridge to the hydrophobic tetrahydrotetramethylnaphthalene ring have been reported,[5456] compound 10 is the first RXR agonist with an indole as the hydrophobic ring and another indole as the bridge substituent. A 9-cis-RA analogue with a 1-(methylethyl)-1,2,3,4-tetrahydroquinolin-6-yl group in place of the β-cyclogeranylidene ring and the 7,8E double bond was reported to selectively activate RXRs.[57] The docked pose of its low-energy conformer suggested to us that the quinoline N atom could stabilize binding by hydrogen bonding with the Cys432 SH group (N–S interatomic distance: 4.6 Å), as was observed in the docked pose for 10. Recently, virtual ligand screening employing molecular interaction fingerprinting identified several new RXR scaffolds.[58] Interestingly, these compounds possessed indole, benzothiophene, and indane rings.
After the present studies were completed, cytosporone B [ethyl 2-(3,5-dihydroxy-2-octanoylphenyl)acetate] was reported to be an NR4A1/Nur77 agonist, which, at 1.0 µm, activated the GAL4–Nur77 LBD on a GAL4 reporter 13.5-fold and Nur77 on a NurRE reporter 10-fold, and bound GST–Nur77 LBD with a Kd value of 1.9 µm.[59] Cytosporone B also docked to the crystal structure of the NR4A1/TR3 LBD (PDB code: 2QW4) in the region typically occupied by NR LBPs, suggesting that other NR4A1/TR3 ligands could be identified. Interestingly, we observed that compound 1 could dock to the same region.
The retinoid bexarotene, which is capable of activating both RARs and RXRs,[1] is used clinically to inhibit CTCL cell growth. DIM-arene 10 activated RXRα but not the RARs or PPARγ. Thus, as expected, a correlation between the abilities of 10 to transactivate RXRα and inhibit cancer cell growth or induce KG-1 cell apoptosis was not found. Previously, we observed that the induction of either activity by a selective RXR agonist required the presence of both the RXR heterodimeric partner RARα or PPARγ and its agonist. In those cases, the RXR agonist enhanced the activity induced by the agonist of its partner. For example, in head-and-neck squamous-cell carcinoma cell lines lacking RARβ, RXR-selective retinoids, which were inactive or poorly active alone, enhanced the ability of RAR-selective retinoids to inhibit cell growth.[60] Similar combinatorial effects were observed in leukemia cell lines.[61, 62] PPARγ ligand citaglitazone and RXR-selective agonist 15 cooperatively inhibited CaLu-6 lung and ZR-75-1 and T-47D breast cancer cell line viability.[12] However, RXR-selective retinoids alone can have growth inhibitory effects in certain contexts. In NR4A1/TR3-expressing, RARα-null MDA-MB-231 breast cancer cells, 15 induced NR4A1/TR3 activation of the βRARE and inhibited cell viability;[30] however, in LNCaP cells, which express RARα, RXRα, and NR4A1/TR3, 15 did not induce NR4A1/TR3 activation, nuclear export, or apoptosis, leading to the conclusion that RARα out-competes NR4A1/TR3 for RXRα.[15]
Of the DIM-arenes 414, only 12 inhibits LNCaP growth with an IC50 value <3 µm. The data in Table 1 suggest that its selective activity against LNCaP cells is not due to affinity for RXRα or activation of RXRα or NR4A1/TR3. The similarity of PPARγ expression levels in the LNCaP and DU-145 cell lines and their comparable growth inhibition by PPARγ agonist ciglitazone[63] further suggests that 12 does not function by activating PPARγ. As LNCaP cells are reported to be somewhat more sensitive to all-trans-RA-induced growth inhibition than DU-145 cells,[64] and 12 is a poor inhibitor of DU-145 growth (Table 2), 12 may perhaps function through RAR activation. The other DIM-arene analogues only inhibit cell growth at high, non-pharmacologic concentrations and were not further explored.
Of the DIM-arenes evaluated, only compound 1 inhibited growth of both prostate cancer cell lines and KG-1 leukemia cells, and induced KG-1 apoptosis. Whether 1 exerts its effects by activating PPARγ or inducing NR4A1/TR3 nuclear export remains to be explored. However, it appears unlikely that 1 functions by antagonizing RXRα–NR4A1/TR3 basal activation on the βRARE, because 11 and 12, while similarly decreasing activation, have disparate effects on LNCaP growth inhibition. Using transfected CV-1 cells, we were unable to confirm that compound 1 at 20 µm activated the NR4A1/TR3 homodimer on an NurRE reporter construct as was reported.[18] The requirement for a particular co-activator protein may have caused the differences between the results observed for this cell line and those reported earlier.[18] More recently, using RNA interference and receptor antagonists, the Safe research group also determined that DIM-arenes exert cancer growth inhibitory and apoptotic effects through PPARγ- and NR4A1/TR3-independent pathways,[14] which include inducing: 1) CCAAT/enhancer binding protein homologous protein (CHOP)/GADD153 expression,[65] 2) pro-apoptotic nonsteroidal anti-inflammatory drug-activated gene-1 expression,[66] 3) c-Jun N-terminal kinase activation,[65] 4) endoplasmic reticulum-mediated stress,[67] and 5) loss of mitochondrial membrane potential and cytochrome c release.[66, 68, 69] Thus, in keeping with other anticancer drugs, the DIM-arenes appear to function through a variety of different signaling pathways; however, initial target(s) remain to be identified.
Chemistry
General
Chemicals and solvents from commercial sources were used without purification unless specified. Reactions were conducted under Ar and monitored by thin-layer chromatography on silica gel (mesh size 60, F254) with visualization under UV light. Organic extracts were dried using Na2SO4 unless otherwise specified and then concentrated at reduced pressure. Standard and flash column chromatography employed silica gel (Merck 60, 230–400 mesh). Experimental procedures were not optimized. Melting points of compounds were determined in capillaries using a Mel-Temp II apparatus and are uncorrected. 1H NMR spectra were recorded on a 300 MHz Varian Unity Inova spectrometer, and shift values are expressed in ppm (δ) relative to CHCl3 as an internal standard. Unless stated otherwise, NMR spectra were taken in CDCl3 solution. IR spectra were obtained on powdered samples using an FTIR Mason Satellite spectrophotometer. HRMS data were recorded on an Agilent ESI-TOF mass spectrometer at The Scripps Research Institute (La Jolla, CA). ESMS was performed on an ABI EPI-3000 instrument.
1-Isopropyl-1H-indoles 18 and 20
A reported method[70] was adapted. N,N-Dimethylformamide (DMF; 8 mL) was added to the indole (10.0 mmol) and 60% NaH dispersion (600 mg, 15 mmol) in mineral oil at 0 °C with stirring. 2-Bromopropane (1.23 g, 10 mmol) was then added dropwise with stirring. The reaction mixture was stirred with slow warming to room temperature until evolution of H2 ceased. After the reaction was complete as indicated by TLC (~1 h), the solvent was removed, and the residue was extracted (CH2Cl2). The extract was washed (H2O and brine) and dried. The residue obtained after concentration was subjected to chromatography (hexane) to give the 1-isopropylindole.
1-Isopropyl-1H-indole (18):[70]
Compound 16, after reaction and chromatography, afforded 18 as a colorless oil (1.05 g, 66 %): TLC (20% EtOAc/hexane): Rf = 0.75; 1H NMR: δ = 1.51 (d, J = 6.6 Hz, 6H, (CH3)2CH), 4.67 (sept, J = 6.6 Hz, 1H, (CH3)2CH), 6.51 (d, J = 3.3 Hz, 1H, 2-InH), 7.09 (dd, J = 6.9, 7.8 Hz, 1H, 5-ArH), 7.20 (dd, J = 6.9, 7.8 Hz, 1H, 6-ArH), 7.21 (d, J = 3.3 Hz, 1H, 3-InH), 7.37 (d, J = 7.8 Hz, 1H, 7-ArH), 7.63 ppm (d, J = 7.8 Hz, 1H, 4-ArH); IR: [nu with tilde] =2974, 1509, 1459, 1221 cm−1; in agreement with reported data.[70]
1-Isopropyl-5-methyl-1H-indole (20)
5-Methyl-1H-indole (19), after reaction and chromatography, afforded 20 as a colorless oil (1.11 g, 64 %): TLC (20% EtOAc/hexane): Rf = 0.77; 1H NMR: δ = 1.49 (d, J = 6.9 Hz, 6H, (CH3)2CH), 2.44 (s, 3H, CH3-Ar), 4.61 (sept, J = 6.9 Hz, 1H, (CH3)2CH), 6.41 (d, J = 3 Hz, 1H, 2-InH), 7.02 (d, J = 8.1 Hz, 1H, 6-ArH), 7.16 (d, J = 3.0 Hz, 1H, 3-InH), 7.25 (d, J = 8.1 Hz, 1H, 7-ArH), 7.41 ppm (s, 1H, 4-ArH); IR: [nu with tilde] =2973, 1480, 1318, 1221 cm−1; HRMS: m/z [M+H]+ calcd for C12H15N: 174.1277, found: 174.1275.
3-Arylacrylates 30 and 34 from arylaldehydes 29 and 33, respectively
A reported method[71] was adapted. Thus, ethyl(triphenyl-phosphoranylidene) acetate (3.48 g, 10.0 mmol) was added to a stirred mixture of the aldehyde (10.0 mmol) in PhMe (10 mL). This mixture was stirred at 80 °C for 1 h, quenched with H2O (10 mL), and extracted with EtOAc (3 × 10 mL). The extract was washed (brine) and dried. The residue obtained after concentration was subjected to chromatography (33% EtOAc/hexane) to give the 3-arylacrylate.
Ethyl (E)-3-(3-formylphenyl)acrylate (30):[72]
Isophthalaldehyde (29) (1.34 g, 10.0 mmol), after reaction and chromatography, afforded 30 as a white solid (1.65 g, 81 %): TLC (20% EtOAc/hexane): Rf = 0.43; mp: 37–39 °C; 1H NMR: δ = 1.35 (t, J = 6.9 Hz, 3H, CH2CH3), 4.29 (q, J = 6.9 Hz, 2H, CH2CH3), 6.54 (d, J = 15.9 Hz, 1H, CH=CHCO), 7.58 (t, J = 7.5 Hz, 1H, 5-ArH), 7.73 (d, J = 15.9 Hz, 1H, CH=CHCO), 7.70 (d, J = 8.7 Hz, 1H, 4-ArH), 7.89 (d, J = 7.5 Hz, 1H, 5-ArH), 8.03 (s, 1H, 2-ArH), 10.05 ppm (s, 1H, CHO); IR: [nu with tilde] =2981, 1695, 1639, 1172 cm−1; in agreement with reported data.[72]
Ethyl (E)-3-(5-formyl-2-thienyl)acrylate (34)
Thiophene-2,5-dicarb-aldehyde (33), after reaction and chromatography, afforded 34 as a yellow solid (1.74 g, 83%): TLC (20% EtOAc/hexane): Rf = 0.31; mp: 45–47 °C; 1H NMR: δ = 1.34 (t, J = 7.2 Hz, 3H, CH2CH3), 4.27 (q, J = 7.2 Hz, 2H, CH2CH3), 6.43 (d, J = 15.9 Hz, 1H, CH=CHCO), 7.32 (d, J = 3.9 Hz, 1H, 3-TpH), 7.70 (d, J = 3.9 Hz, 1H, 4-TpH), 7.75 (d, J = 15.9 Hz, 1H, CH=CHCO), 9.91 ppm (s, 1H, CHO); IR: [nu with tilde] = 1706, 1665, 1627, 1166 cm−1; HRMS: m/z [M+H]+ calcd for C10H10O3S: 211.0423, found: 211.0422.
Condensation of indoles 16–21 with 4-substituted benzaldehydes 22 and 24, ethyl glyoxalate (27), or arylaldehydes 30 and 33 to give di(indolyl)methylbenzenes 1, 25, and 26, di-(indolyl)acetate 28, 3-di(indolyl)methylcinnamates 31 and 32, or 3-[5-di(indolyl)methyl-2-thienyl]acrylates 35–40, respectively
A reported procedure[46] was used. Briefly, silica gel (0.500 g) was added to a stirred mixture of CeCl3·7H2O (0.113 g, 0.300 mmol) and NaI (0.045 g, 0.300 mmol) in MeCN (8 mL). The mixture was stirred overnight and then concentrated to give a yellow solid, to which was then added the indole (2.00 mmol) and aldehyde (1.00 mmol) dissolved in MeCN (15 mL). The mixture was stirred for 4 h, diluted with Et2O (20 mL), and filtered through a short pad of Celite (ether rinse). The crude product obtained on concentration was purified by chromatography.
3,3′-[(4-Trifluoromethylphenyl)methylene]bis(1H-indole) (1):[46]
Indole (16) (0.234 g) and 4-(trifluoromethyl)benzaldehyde (22) (0.174 g), after reaction and chromatography (33% EtOAc/hexane), gave 1 as a white solid (0.63 g, 86%): TLC (33% EtOAc/hexane): Rf = 0.43; mp: 85–87 °C; 1H NMR: δ = 5.95 (s, 1H, PhCH), 6.67 (d, J = 1.5 Hz, 2H, 2-ArH), 7.00–7.04 (m, 2H, ArH), 7.16–7.22 (m, 2H, ArH), 7.35–7.39 (m, 4H, ArH), 7.46 (d, J = 8.1 Hz, 2H, 2, 6-ArH), 7.53 (d, J = 8.1 Hz, 2H, 3,5-ArH), 7.98 ppm (s, 2H, NH); IR (CHCl3): [nu with tilde] = 3425, 3415, 3048, 2666, 2367, 1324 cm−1; in agreement with reported data.[46]
Methyl 4-[di(1H-indol-3-yl)methyl]benzoate (25)
16 (0.234 g) and methyl 4-formylbenzoate (24) (0.164 g) gave, after reaction and chromatography (33% EtOAc/hexane), 25 as a white solid (0.677 g, 89%): TLC (33% EtOAc/hexane): Rf = 0.41; mp: 98–100 °C; 1H NMR: δ = 3.89 (s, 3H, CH3), 5.94 (s, 1H, CH), 6.60 (d, J = 1.5 Hz, 2H, C=CH-NH), 6.98–7.03 (m, 2H, ArH), 7.15–7.20 (m, 2H, ArH), 7.34–7.37 (m, 2H, ArH), 7.37 (d, J = 8.4 Hz, 2H, 2, 6-ArH), 7.42 (d, J = 8.4 Hz, 2H, 3,5-ArH), 7.94–7.97 (m, 2H, ArH), 7.97 ppm (s, 2H, NH); IR (CHCl3): [nu with tilde] = 3425, 3403, 1702, 1282 cm−1; HRMS: m/z [M−H] calcd for C25H20N2O2 : 379.1452, found: 379.1452.
Methyl 4-[di(1-methyl-1H-indol-3-yl)methyl]benzoate (26)
1-Methyl-1H-indole (17) (0.262 g) and 24 (0.164 g) gave, after reaction and chromatography (20% EtOAc/hexane), 26 as a white solid (0.342 g, 84%): TLC (20% EtOAc/hexane): Rf = 0.34; mp: 185–187 °C; 1H NMR: δ = 3.68 (s, 6H, NCH3), 3.89 (s, 3H, OCH3), 5.92 (s, 1H, ArCH), 6.51 (s, 2H, 2-ArH), 7.00 (t, J = 7.2 Hz, 2H, ArH), 7.18–7.23 (m, 2H, ArH), 7.25–7.28 (m, 2H, ArH), 7.31–7.36 (m, 2H, ArH), 7.41 (d, J = 8.1 Hz, 2H, 3,5-PhH), 7.95 ppm (d, J = 8.1 Hz, 2H, 2,6-PhH); IR: [nu with tilde] = 1716, 1608, 1274 cm−1; HRMS: m/z [M+Na]+ calcd for C27H24N2O2 : 431.1730, found: 431.1728.
Ethyl 2,2-di(1H-indol-3-yl)acetate (28):[46]
16 (0.234 g) and a 50% solution of ethyl glyoxalate (27) (1.0 mmol) in PhMe (0.198 mL) gave, after reaction and chromatography (33% EtOAc/hexane), 28 as a white solid (0.522 g, 82%): TLC (50% EtOAc/hexane): Rf = 0.62; mp: 69–71 °C; 1H NMR: δ = 1.21 (t, J = 7.2 Hz, 3H, CH2CH3), 3.48 (q, J = 7.2 Hz, 2H, CH2CH3), 5.54 (s, 1H, Ar-CH), 7.07–7.10 (m, 2H, 5-ArH), 7.14 (s, 2H, 2-ArH), 7.17–7.21 (m, 2H, 6-ArH), 7.35 (d, J = 7.5 Hz, 2H, 7-ArH), 7.63 (d, J = 8.4 Hz, 2H, 4-ArH), 8.01 ppm (br s, 2H, NH); IR: [nu with tilde] =3406, 1715, 1174 cm−1; in agreement with reported data.[46]
Ethyl (E)-3-[3-di(1H-indol-3-yl)methylphenyl]acrylate (31)
16 (0.234 g) and ethyl (E)-3-(3-formylphenyl)acrylate (30) (0.204 g) gave, after reaction and chromatography (20% EtOAc/hexane), 31 as a white solid (0.365 g, 87%): TLC (50% EtOAc/hexane): Rf = 0.69; mp: 90–92 °C; 1H NMR: δ = 1.27 (t, J = 7.2 Hz, 3H, CH2CH3), 4.23 (q, J = 7.2 Hz, 2H, CH2CH3), 5.90 (s, 1H, ArCH), 6.35 (d, J = 16.2 Hz, 1H, CH=CHCO), 6.66 (s, 2H, 2-ArH), 7.01 (t, J = 7.2 Hz, 2H, ArH), 7.18 (t, J = 7.5 Hz, 2H, ArH) 7.31 (d, J = 7.5 Hz, 1H, ArH), 7.36–7.38 (m, 6H, ArH), 7.51 (s, 1H, 2-PhH), 7.62 (d, J = 16.2 Hz, 1H, CH=CHCO), 7.95 ppm (br s, 2H, NH); IR: [nu with tilde] = 3399, 1693, 1635, 1272, 1182 cm−1; HRMS: m/z [M+Na]+ calcd for C28H24N2O2 : 443.1730, found: 443.1715.
Ethyl (E)-3-[3-di(1-methyl-1H-indol-3-yl)methylphenyl]acrylate (32)
1-Methyl-1H-indole (17) (0.262 g) and 30 ( 0.204 g) gave, after reaction and chromatography (20% EtOAc/hexane), 32 as a white solid (0.352 g, 84%); TLC (20% EtOAc/hexane): Rf = 0.34; mp: 93–95 °C; 1H NMR: δ = 1.29 (t, J = 7.2 Hz, 3H, CH2CH3), 3.69 (s, 6H, NCH3), 4.23 (q, J = 7.2 Hz, 2H, CH2CH3), 5.89 (s, 1H, ArCH), 6.35 (d, J = 15.9 Hz, 1H, CH=CHCO), 6.52 (s, 2H, 2-ArH), 7.01 (t, J = 7.2 Hz, 2H, ArH), 7.20 (t, J = 7.2 Hz, 2H, ArH) 7.30 (d, J = 8.4 Hz, 2H, ArH), 7.35–7.38 (m, 4H, ArH), 7.50 (s, 1H, 2-PhH), 7.63 ppm (d, J = 15.9 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 1705, 1633, 1614, 1468, 1176 cm−1; HRMS: m/z [M+Na]+ calcd for C30H28N2O2 : 471.2043, found: 471.2055.
Ethyl (E)-3-[5-(di(1H-indol-3-yl)methyl-2-thienyl]acrylate (35)
16 (0.234 g) and ethyl (E)-3-(5-formyl-2-thienyl)acrylate (34) (0.210 g) gave, after reaction and chromatography (20% EtOAc/hexane), 35 as a white solid (0.360 g, 87%): TLC (50% EtOAc/hexane): Rf = 0.67; mp: 95–97 °C; 1H NMR: δ = 1.28 (t, J = 7.2 Hz, 3H, CH2CH3), 4.20 (q, J = 7.2 Hz, 2H, CH2CH3), 6.05 (d, J = 15.9 Hz, 1H, CH=CHCO), 6.12 (s, 1H, ArCH), 6.60 (d, J = 3.6 Hz, 1H, TpH), 6.88 (s, 2H, 2-ArH), 7.02–7.06 (m, 2H, ArH), 7.07 (d, J = 3.6 Hz, 1H, TpH), 7.17–7.22 (m, 2H, ArH), 7.38 (d, J = 7.8 Hz, 2H, ArH), 7.46 (d, J = 7.8 Hz, 2H, ArH), 7.68 (d, J = 15.9 Hz, 1H, CH=CHCO), 8.00 ppm (br s, 2H, NH); IR: [nu with tilde] = 3400, 1690, 1619, 1456, 1175 cm−1; HRMS: m/z [M+Na]+ calcd for C26H22N2O2S: 427.1475, found: 427.1455.
Ethyl (E)-3-[5-di(1-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylate (36)
1-Methyl-1H-indole (17) (0.262 g) and 34 (0.210 g) gave, after reaction and chromatography (20% EtOAc/hexane), 36 as a white solid (0.366 g, 83%): TLC (20% EtOAc/hexane): Rf = 0.33; mp: 158–160 °C; 1H NMR: δ = 1.28 (t, J = 7.2 Hz, 3H, CH2CH3), 3.71 (s, 6H, NCH3), 4.20 (q, J = 7.2 Hz, 2H, CH2CH3), 6.04 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.10 (s, 1H, ArCH), 6.73 (s, 2H, 2-ArH), 6.84 (d, J = 2.7 Hz, 1H, TpH), 7.01–7.06 (m, 2H, ArH), 7.07 (d, J = 2.7 Hz, 1H, TpH), 7.20–7.25 (m, 2H, ArH), 7.31 (d, J = 8.1 Hz, 2H, ArH), 7.45 (d, J = 8.1 Hz, 2H, ArH), 7.69 ppm (d, J = 15.9 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 1702, 1618, 1465, 1159 cm−1; HRMS: m/z [M+Na]+ calcd for C28H26N2O2S: 455.1788, found: 455.1779.
Ethyl (E)-3–5-[di(1-isopropyl-1H-indol-3-yl)methyl-2-thienyl]acrylate (37)
1-Isopropyl-1H-indole (18) (0.318 g) and 34 (0.210 g) gave, after reaction and chromatography (50% EtOAc/hexane), 37 as a white solid (0.914 g, 92%): TLC (50% EtOAc/hexane): Rf = 0.86; mp: 83–85 °C; 1H NMR: δ = 1.28 (t, J = 7.2 Hz, 3H, CH2CH3), 1.46 (d, J = 6.6 Hz, 12 H, CH(CH3)2), 4.20 (q, J = 7.2 Hz, 2H, CH2CH3), 4.63 (sept, J = 6.6 Hz, 2H, CH(CH3)2), 6.04 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.10 (s, 1H, ArCH), 6.82 (d, J = 3.3 Hz, 1H, 4-TpH), 6.91 (s, 2H, 2-ArH), 7.01 (dd, J = 7.8, 7.2 Hz, 2H, 6-ArH), 7.06 (d, J = 3.3 Hz, 1H, 3-TpH), 7.18 (dd, J = 7.8, 7.2 Hz, 2H, 5-ArH), 7.37 (d, J = 7.8 Hz, 2H, 7-ArH), 7.43 (d, J = 7.8 Hz, 2H, 4-ArH), 7.69 ppm (d, J = 15.6 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 2974, 1704, 1620, 1461, 1194 cm−1; HRMS: m/z [M+Na]+ calcd for C32H34N2O2S: 533.2233, found: 533.2232.
Ethyl (E)-3-[5-(di(1-isopropyl-5-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylate (38)
1-Isopropyl-5-methyl-1H-indole (20) (0.346 g) and 34 (0.210 g) gave, after reaction and chromatography (20% EtOAc/hexane), 38 as a white solid (0.969 g, 93%): TLC (50% EtOAc/hexane): Rf = 0.86; mp: 86–88 °C; 1H NMR: δ = 1.29 (t, J= 7.2 Hz, 3H, CH2CH3), 1.44 (d, J = 7.2 Hz, 12H, CH(CH3)2), 2.36 (s, 6H, CH3-Ar), 4.21 (q, J = 7.2 Hz, 2H, CH2CH3), 4.58 (sept, J = 7.2 Hz, 2H, CH(CH3)2), 6.04 (s, 1H, ArCH), 6.05 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.82 (d, J = 3.6 Hz, 1H, 4-TpH), 6.84 (s, 2H, 2-ArH), 7.04 (d, J = 8.1 Hz, 2H, 6-ArH), 7.07 (d, J = 3.3 Hz, 1H, 3-TpH), 7.22 (s, 2H, 4-ArH), 7.26 (d, J = 8.1 Hz, 2H, 7-ArH), 7.70 ppm (d, J = 15.6 Hz, 1H, CH=CHCO); IR: [nu with tilde] =2976, 1708, 1662 cm−1; HRMS: m/z [M+Na]+ calcd for C34H38N2O2S: 561.2546, found: 561.2542.
Ethyl (E)-3-[5-(di(5-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylate (39)
5-Methyl-1H-indole (19) (0.262 g) and 34 (0.210 g) gave, after reaction and chromatography (20% EtOAc/hexane), 39 as a white solid (0.479 g, 89%): TLC (50% EtOAc/hexane): Rf = 0.66; mp: 96–98 °C; 1H NMR: δ = 1.28 (t, J = 7.2 Hz, 3H, CH2CH3), 2.38 (s, 6H, CH3-Ar), 4.20 (q, J = 7.2 Hz, 2H, CH2CH3), 6.05 (s, 1H, ArCH), 6.05 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.81 (s, 2H, 2-ArH), 6.84 (d, J = 3.6 Hz, 1H, 4-TpH), 7.02 (d, J = 7.8 Hz, 2H, 6-ArH), 7.08 (d, J = 3.6 Hz, 1H, 3-TpH), 7.25 (s, 2H, 4-ArH), 7.26 (d, J = 7.8 Hz, 2H, 7-ArH), 7.69 (d, J = 15.6 Hz, 1H, CH=CHCO), 7.90 ppm (s, 2H, NH); IR: [nu with tilde] =3418, 1696, 1619 cm−1; HRMS: m/z [M+H]+ calcd for C28H26N2O2S: 455.1788, found: 455.1776.
Ethyl (E)-3-[5-(di(7-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylate (40)
7-Methyl-1H-indole (21) (0.262 g) and 34 (0.210 g) gave, after reaction and chromatography (20% EtOAc/hexane), 40 as a white solid (0.468 g, 87%): TLC (50% EtOAc/hexane): Rf = 0.61; mp: 94–96 °C; 1H NMR: δ =1.28 (t, J=6.3 Hz, 3H, CH2CH3), 2.49 (s, 6H, CH3-Ar), 4.20 (q, J=6.3 Hz, 2H, CH2CH3), 6.03 (d, J = 15.3 Hz, 1H, CH=CHCO), 6.10 (s, 1H, ArCH), 6.84 (d, J = 3.6 Hz, 1H, 4-TpH), 6.86 (s, 2H, 2-ArH), 6.98 (dd, J = 6.3, 6.3 Hz, 2H, 5-ArH), 6.99 (d, J = 6.3 Hz, 2H, 6-ArH), 7.06 (d, J = 3.6 Hz, 1H, 3-TpH), 7.31 (d, J = 6.3 Hz, 2H, 4-ArH), 7.68 (d, J = 15.6 Hz, 1H, CH=CHCO), 7.90 ppm (s, 2H, NH); IR (CHCl3): [nu with tilde] = 2970, 2945, 2360, 2342, 1734, 1053 cm−1; HRMS: m/z [M+H]+ calcd for C28H26N2O2S: 455.1788, found: 455.1788.
4-Di(1H-indol-3-yl)methylphenol (3):[47]
A reported procedure[47] was used. 4-Hydroxybenzaldehyde (23) (0.233 g, 1.00 mmol), indole (16) (0.234 g, 2.00 mmol), and I2 (0.051 mg, 0.2 mmol) in MeCN (3 mL) was stirred for 1 min, at which time TLC [(33% EtOAc/hexane): Rf = 0.25] indicated complete reaction. The mixture was treated with 5% Na2S2O3(aq) (5 mL) and extracted with EtOAc (3 × 5 mL). The extract was washed (H2O and brine) and dried. Concentration and chromatography (33% EtOAc/hexane) afforded 3 as a red solid (0.331 g, 98%): TLC (33% EtOAc/hexane): Rf = 0.25; mp: 110–114 °C; 1H NMR: δ = 4.66 (s, 1H, ArOH), 5.83 (s, 1H, ArCH), 6.65 (s, 2H, 2-ArH), 6.73 (d, J = 8.4 Hz, 2H, 2,6-PhH), 6.98–7.03 (m, 2H, 5-ArH), 7.15 (d, J = 7.2 Hz, 2H, 4-ArH), 7.20 (d, J = 8.4 Hz, 2H, 3,5-PhH), 7.34–7.40 (m, 2H, 6-ArH), 7.38 (d, J = 8.4 Hz, 2H, 7-ArH), 7.91 ppm (s, 2H, NH); IR (CHCl3): [nu with tilde] = 2966, 2215 cm−1; in agreement with reported data.[47]
Hydrolysis of esters 25, 26, 28, 31, 32 and 35–40 to carboxylic acids 4–14, respectively
A reported procedure[73] was adapted. Briefly, LiOH·H2O (0.033 g, 0.789 mmol) was added to a solution of ester (0.158 mmol) in MeOH (1.5 mL). This mixture was heated at reflux under Ar for 3 h, cooled to room temperature, acidified with 1 n HCl (pH ~ 1), and extracted with EtOAc (10 mL). The extract was washed (H2O and brine). Concentration and drying or chromatography (50% EtOAc/hexane) gave the pure carboxylic acid. Acids were found to be unstable to analytical RP-HPLC conditions. Therefore, the corresponding esters were analyzed. Chromatograms indicated that esters were > 98% pure (see Supporting Information).
4-Di(1H-indol-3-yl)methylbenzoic acid (4)
After hydrolysis, concentration, and drying, 25 (0.060 g, 0.158 mmol) yielded 4 as a white solid (0.049 g, 85%): TLC (50% EtOAc/hexane): Rf = 0.28; mp: 248–250 °C (dec); 1H NMR: δ = 5.92 (s, 1H, CH), 6.66 (s, 2H, C=CHNH), 6.86–6.90 (m, 2H, ArH), 7.03–7.08 (m, 2H, ArH), 7.26 (d, J = 7.8 Hz, 2H, 2,6-ArH), 7.33 (d, J = 7.8 Hz, 2H, 3,5-ArH), 7.41–7.43 (m, 2H, ArH), 7.91–7.94 (m, 2H, ArH), 10.17 ppm (s, 1H, CO2H); IR (CHCl3): [nu with tilde] = 2970, 2945, 2360, 2342, 1734, 1053 cm−1; HRMS: m/z [M+Na]+ calcd for C24H18N2O2 : 389.1260, found: 389.1273.
4-Di(1-methyl-1H-indol-3-yl)methylbenzoic acid (5)
After hydrolysis and chromatography, 26 (0.122 g, 0.30 mmol) yielded 5 as a pale-yellow solid (0.105 g, 89%): TLC (50% EtOAc/hexane): Rf = 0.49; mp: 257–259 °C (dec); 1H NMR (CD3OD): δ = 3.70 (s, 6H, NCH3), 5.92 (s, 1H, ArCH), 6.60 (s, 2H, 2-ArH), 7.00 (t, J = 7.2 Hz, 2H, ArH), 6.89–6.94 (m, 2H, ArH), 7.11–7.16 (m, 2H, ArH), 7.26–7.28 (m, 2H, ArH), 7.43 (d, J = 8.4 Hz, 2H, 3,5-PhH), 7.93 ppm (d, J = 8.4 Hz, 2H, 2,6-PhH); IR: [nu with tilde] =3384, 1680, 1607, 1291 cm−1; HRMS: m/z [M−H] calcd for C26H22N2O2: 393.1608, found: 393.1610.
2,2-Di(1H-indol-3-yl)acetic acid (6):[74]
After hydrolysis and chromatography, 28 (0.095 g, 0.30 mmol) yielded 6 as a pale-yellow solid (0.076 g, 87%): TLC (50% EtOAc/hexane): Rf = 0.29; mp: 187–189 °C; 1H NMR ([D6]DMSO): δ = 5.34 (s, 1H, ArCH), 6.95 (dd, J = 7.5, 6.6 Hz, 2H, 5-ArH), 7.06 (dd, J = 7.2, 6.6 Hz, 2H, 6-ArH), 7.21 (s, 2H, 2-ArH), 7.36 (d, J = 7.2 Hz, 2H, 7-ArH), 7.55 (d, J = 7.5 Hz, 2H, 4-ArH), 10.92 (s, 2H, NH), 12.36 ppm (br s, 1H, CO2H); IR: [nu with tilde] = 3400, 1702, 1338, 1198 cm−1; in agreement with reported data.[74]
(E)-3-[3-Di(1H-indol-3-yl)methylphenyl]acrylic acid (7)
After hydrolysis and chromatography, 31 (0.105 g, 0.25 mmol) yielded 7 as a pale-yellow solid (0.088 g, 90 %): TLC (50% EtOAc/hexane): Rf = 0.25; mp: 137–139 °C; 1H NMR (CD3OD): δ = 5.89 (s, 1H, ArCH), 6.35 (d, J = 16.2 Hz, 1H, CH=CHCO), 6.66 (s, 2H, 2-ArH), 6.85–6.90 (m, 2H, ArH), 7.03–7.08 (m, 2H, ArH) 7.26–7.29 (m, 2H, ArH), 7.32 (s, 2H, NH), 7.32–7.34 (m, 4H, ArH), 7.42–7.45 (m, 2H, ArH), 7.59 ppm (d, J = 16.2 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 3427, 3400, 1683, 1630, 1274 cm−1; HRMS: m/z [M−H] calcd for C26H20N2O2 : 391.1452, found: 391.1448.
(E)-3-[3-Di(1-methyl-1H-indol-3-yl)methylphenyl]acrylic acid (8)
After hydrolysis and chromatography, 32 (0.09 g, 0.20 mmol) yielded 8 as a pale-yellow solid (0.076 g, 90%): TLC (50% EtOAc/hexane): Rf = 0.36; mp: 140–142 °C; 1H NMR: δ = 3.69 (s, 6H, NCH3), 5.89 (s, 1H, ArCH), 6.35 (d, J = 15.9 Hz, 1H, CH=CHCO), 6.52 (s, 2H, 2-ArH), 7.00 (t, J = 7.2 Hz, 2H, ArH), 7.21 (t, J = 7.2 Hz, 2H, ArH) 7.29–7.31 (m, 2H, ArH), 7.35–7.38 (m, 4H, ArH), 7.51 (s, 1H, 2-PhH), 7.71 (d, J = 15.9 Hz, 1H, CH=CHCO), 10.28 ppm (br s, 1H, CO2H); IR: [nu with tilde] = 3382, 1691, 1469, 1014 cm−1; HRMS: m/z [M+Na]+ calcd for C28H24N2O2 : 443.1730, found: 443.1732.
(E)-3-[5-Di(1H-indol-3-yl)methyl-2-thienyl]acrylic acid (9)
After hydrolysis and chromatography, 35 (0.082 g, 0.20 mmol) yielded 9 as a pale-yellow solid (0.07 g, 91 %): TLC (50% EtOAc/hexane): Rf = 0.26; mp: 196–198 °C; 1H NMR (CD3OD): δ = 6.05 (d, J = 15 Hz, 1H, CH=CHCO), 6.09 (s, 1H, Ar-CH), 6.60 (d, J = 3.6 Hz, 1H, TpH), 6.88 (s, 2H, 2-ArH), 7.02–7.06 (m, 2H, ArH), 7.07 (d, J = 3.6 Hz, 1H, TpH), 7.17–7.22 (m, 2H, ArH), 7.38 (d, J = 7.8 Hz, 2H, ArH), 7.46 (d, J = 7.8 Hz, 2H, ArH), 7.68 (d, J = 15.9 Hz, 1H, CH=CHCO), 8.00 ppm (br s, 2H, NH); IR: [nu with tilde] = 3366, 3335, 1678, 1454, 1015 cm−1; HRMS: m/z [M+H]+ calcd for C24H18N2O2S: 399.1162, found: 399.1147.
(E)-3-[5-Di(1-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylic acid (10)
After hydrolysis and chromatography, 36 (0.088 g, 0.20 mmol) yielded 10 as a pale-yellow solid (0.077 g, 93%): TLC (50% EtOAc/hexane): Rf = 0.39; mp: 120–122 °C; 1H NMR (CD3OD): δ = 3.72 (s, 6H, N(CH3)2), 6.01 (d, J = 15.9 Hz, 1H, CH=CHCO), 6.11 (s, 1H, ArCH), 6.81 (s, 2H, 2-InH), 6.86–6.88 (m, 1H, TpH), 6.93–6.98 (m, 2H, ArH), 7.12–7.15 (m, 1H, TpH), 7.15–7.17 (m, 2H, ArH), 7.34 (d, J = 8.1 Hz, 2H, ArH), 7.37 (d, J = 8.1 Hz, 2H, ArH), 7.69 ppm (d, J = 15.9 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 3362, 1677, 1613, 1262 cm−1; HRMS: m/z [M+Na]+ calcd for C26H22N2O2S: 449.1294, found: 449.1287.
(E)-3-[5-(Di(1-isopropyl-1H-indol-3-yl)methyl-2-thienyl]acrylic acid (11)
After hydrolysis and chromatography, 37 (0.497 g, 1.00 mmol) yielded 11 as a pale-yellow solid (0.413 g, 88 %): TLC (50% EtOAc/hexane): Rf = 0.41; mp: 248–250 °C (dec); 1H NMR: δ = 1.40 (d, J = 6.3 Hz, 12H, CH(CH3)2), 4.71 (sept, J = 6.3 Hz, 2H, CH(CH3)2), 5.96 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.16 (s, 1H, ArCH), 6.92 (d, J = 7.8 Hz, 2H, 7-ArH), 6.94 (d, J = 3.6 Hz, 1H, 4-TpH), 7.09 (dd, J = 6.6, 7.8 Hz, 2H, 6-ArH), 7.28 (d, J = 9 Hz, 2H, 4-ArH), 7.30 (d, J = 3.6 Hz, 1H, 3-TpH), 7.45 (s, 2H, 2-ArH), 7.46 (dd, J = 9.0, 6.6 Hz, 2H, 5-ArH), 7.63 (d, J = 15.6 Hz, 1H, CH=CHCO), 12.22 ppm (s, 1H, CO2H); IR: [nu with tilde] = 3571, 2976, 1677, 1610, 1460, 1194 cm−1; HRMS: m/z [M+Na]+ calcd for C30H30N2O2S: 505.1920, found: 505.1923.
(E)-3-[5-(Di(1-isopropyl-5-methyl-1H-indol-3-yl)methyl-2-thienyl] acrylic acid (12)
After hydrolysis and chromatography, 38 (0.088 g, 0.20 mmol) yielded 12 as a pale-yellow solid (0.077 g, 93%): TLC (50% EtOAc/hexane): Rf = 0.40; mp: 251–253 °(dec); 1H NMR: δ = 1.39 (d, J = 6.3 Hz, 12H, CH(CH3)2), 2.28 (s, 6H, CH3-Ar), 4.67 (sept, J = 6.3 Hz, 2H, CH(CH3)2), 5.96 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.07 (s, 1H, ArCH), 6.89 (d, J = 3.6 Hz, 1H, 4-TpH), 6.91 (d, J = 7.8 Hz, 2H, 6-ArH), 7.19 (s, 2H, 2-ArH), 7.20 (s, 2H, 4-ArH), 7.30 (d, J = 3.6 Hz, 1H, 3-TpH), 7.35 (d, J = 7.8 Hz, 2H, 7-ArH), 7.62 (d, J = 15.6 Hz, 1H, CH=CHCO), 12.20 ppm (s, 1H, CO2H); IR: [nu with tilde] = 3617, 2971, 1680, 1613, 1197 cm−1; HRMS: m/z [M+Na]+ calcd for C32H34N2O2S: 533.2233, found: 533.2237.
(E)-3-[5-(Di(5-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylic acid (13)
After hydrolysis and chromatography, 39 (0.091 g, 0.20 mmol) yielded 13 as a pale-yellow solid (0.078 g, 91%): TLC (50% EtOAc/hexane): Rf = 0.18; mp: 172–174 °C; 1H NMR (CD3OD): δ = 2.32 (s, 6H, CH3-Ar), 6.00 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.04 (s, 1H, ArCH), 6.80 (s, 2H, 2-ArH), 6.84 (d, J = 3.6 Hz, 1H, 4-TpH), 6.91 (d, J = 7.2 Hz, 2H, 6-ArH), 7.15 (s, 2H, 4-ArH), 7.16 (d, J = 3.6 Hz, 1H, 3-TpH), 7.23 (d, J = 7.8 Hz, 2H, 7-ArH), 7.70 ppm (d, J = 15.6 Hz, 1H, CH=CHCO); IR: [nu with tilde] =3589, 3407, 1677, 1612, 1265 cm−1; HRMS: m/z calcd [M+H]+ for C26H22N2O2S: 427.1475, found: 422.1468.
(E)-3-[5-(Di(7-methyl-1H-indol-3-yl)methyl-2-thienyl]acrylic acid (14)
After hydrolysis and chromatography, 40 (0.091 g, 0.20 mmol) yielded 14 as a pale-yellow solid (0.077 g, 90%): TLC (50% EtOAc/hexane): Rf = 0.18; mp: 175–177 °C; 1H NMR (CD3OD): δ = 2.48 (s, 6H, CH3-Ar), 6.00 (d, J = 15.6 Hz, 1H, CH=CHCO), 6.08 (s, 1H, ArCH), 6.84 (d, J = 3.6 Hz, 1H, 4-TpH), 6.87 (dd, J = 7.8, 7.8 Hz, 2H, 5-ArH), 6.86 (s, 2H, 2-ArH), 7.20 (d, J = 7.8 Hz, 2H, 6-ArH), 7.15 (d, J = 3.6 Hz, 1H, 3-TpH), 7.20 (d, J = 7.8 Hz, 2H, 4-ArH), 7.69 ppm (d, J = 15.6 Hz, 1H, CH=CHCO); IR: [nu with tilde] = 3629, 3409, 1679, 1612, 1263 cm−1; HRMS: m/z [M+H]+ calcd for C26H22N2O2S: 427.1475, found: 422.1466.
Computational methods
Three-dimensional structures of potential DIM-arenes were constructed using the SKETCH module in Sybyl 7.1 (Tripos, St. Louis, MO, USA). Energy was minimized with the Tripos force-field using the conjugated gradient method until a convergence value of 0.005 kcalÅ−1mol−1 was achieved. Docking was performed using the automated docking program AutoDock 4.0.[75] Grid maps for docking simulations were generated with a grid of 50 points in the x, y, and z directions using the Arg316 residue of the RXRα LBD (PDB code: 1MVC) as the central residue and 0.375 Å as the grid spacing. A distance-dependent function of the dielectric constant was used for the energetic map calculations. The AutoDockTools suite[75] was used to identify the torsion angles of the ligands and assign Gasteiger and Kollman partial atomic charges to DIM-arenes and the RXRα LBD, respectively. One hundred independent docking runs were performed for each DIM-arene using the Lamarckian genetic algorithm and a maximum number of 2 500 000 energy evaluations. Other parameters were assigned default values, implemented by the program. Cluster analysis performed on the results from 100 runs gave a root-mean-square tolerance of 2 Å.
Biological assays
Competitive binding to RXRα
Purified polyhistidine-tagged human RXRα LBD(223–462) (50 ng) was incubated with [11,12-3H2]9-cis-RA (Amersham Biosciences, Piscataway, NJ, USA) in the presence of various concentrations of unlabeled 9-cis-RA or DIM-arene at 4 °C for 12 h. RXRα LBD was then captured by incubation with Ni-coated beads. The amount of specifically bound radiolabel was determined by scintillation counting of the beads followed by subtraction of nonspecifically bound radiolabel as described.[76]
Transient-transfection assays for transcriptional activation of RXRα, RARs, and NR4A1/TR3
CV-1 cells (5 × 104 per well) were plated in 24-well plates in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% charcoal-stripped FBS. After 24 h, cells were transfected with the TREpal–tk–CAT reporter construct (100 ng) alone or with an rxrα, rarα, rarβ, or rarγ gene plasmid (25 ng); NurRE–tk–CAT alone or with the nr4a1/tr3 plasmid; or βRARE–tk–CAT alone or with rxrα and/or nr4a1/tr3 as described[12, 15, 17, 76] and, at the same time, the β-galactosidase gene (gal) (100 ng) was included as a control for transfection efficiency. Cells were then treated for 48 h with DMSO vehicle alone (control) or compound dissolved in DMSO as specified in Figures 3 and 4A–C, then collected for measurement of CAT enzyme activity, which was normalized for transfection efficiency on the basis of βGAL enzyme activity as a measure of co-transfected βgal gene activity. Results reported are averages of duplicates.
PPRE–tk–LUC reporter assays
PPARγ transcriptional activation was monitored by transfection of an (AOx)3tk–LUC plasmid, containing three copies of the acyl-CoA oxidase PPRE upstream of the basal thymidine kinase (tk) promoter in the tk–LUC plasmid.[76, 77] HeLa cells were maintained in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin. Cells (7 × 104) in 1.0 mL of medium were transfected in 24-well plates using 1 µg polyethylenimine (PEI, Polyscience Inc., Warrington, PA, USA), 0.2 µg (AOx)3tk–LUC plasmid,[60] and either 0.3 µg of human pparγ[78] or pMSCVhph vector in 50 µL serum-free DMEM per well. This PEI mixture had been incubated for 20 min at room temperature before addition to the cells, which previously had their medium replaced by fresh medium. Transfections also included Renilla luciferase (pRL–tk) plasmid as a control for transfection efficacy. After overnight incubation at 37 °C under 10% CO2, medium was replaced with DMEM containing 10% FBS alone, or combined with 20 µm troglitazone as the positive control, 0.2, 2.0, 10, 20, or 40 µm (final concentration) 1, 3, or 10, or DMSO vehicle alone up to a final maximal concentration of 0.5%, which did not affect the assay results. After 48 h incubation, cells were lysed in lysis buffer and assayed for luciferase activity using the Dual Luciferase kit (Promega Corp., Madison, WI, USA) according to the manufacturer’s protocol. Luminescence was measured using a Veritas Microplate Luminometer (Turner Biosystems, Sunnyvale, CA, USA). Data were normalized to pRL–tk and expressed as fold-activation above (AOx)3tk–LUC activation induced by vehicle alone. Results expressed in Figure 4D are the averages of triplicates ±SD.
Growth inhibition
LNCaP and DU-145 prostate cancer cells were obtained from ATCC (Manassas, VA, USA) and grown at 37°C as described.[15, 76] Compound-induced growth inhibition was assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in 96-well microtiter plates. Cells were allowed to attach and then treated for 48 h with DMSO alone or by DIM-arene in DMSO at concentrations specified in Table 1. At 4 h before the end of treatment, MTT was added to each well to a final concentration of 1.0 mg mL−1, and incubation was continued at 37 °C. The MTT reaction was terminated by removing the supernatant, and the dye was dissolved by adding DMSO (200 µL) followed by thorough mixing. Plates were read at 550 nm using a microtiter plate reader. Means and standard deviations were derived from eight replicates.
KG-1 cells in medium containing 5% FBS were allowed to attach overnight, treated for the indicated times with analogues at the specified concentrations in DMSO or DMSO alone (0.1% final concentration), and then harvested as described.[62] Cell numbers were counted in triplicate wells using a hemocytometer.
Cell viability
HeLa-S cervical adenocarcinoma, PPC-1 prostate adenocarcinoma, U2OS bone osteosarcoma, and HCT-116 colon carcinoma cell lines (ATCC) were cultured in DMEM supplemented with 10% heat-inactivated FBS and 1% penicillin/streptomycin mixture in 10 cm dishes at 37°C, 10% CO2, and 90% relative humidity. Cell suspensions (1 × 104 cells in 50 µL per well) in white-bottomed Greiner 384-well plates were pre-incubated for 1 h under the same culture conditions and then treated with various concentrations of 1 that had been dissolved in DMSO (0.125% maximum concentration, which had no effect on cell growth) and diluted into DMEM containing 10% FBS. After 24 h incubation, plates were transferred to room temperature for 15 min. The luciferase assay was performed by adding ATPLite™ Assay System reagent (20 µL per well; PerkinElmer, Boston, MA, USA) and 10 min later measuring light emission using a FlexStation 3 Microplate Reader (Molecular Devices, Sunnyvale, CA, USA). Experiments were performed in triplicate. Viability was calculated on the basis of maximum luminescence intensity observed for each cell line in the absence of compound 1 (100% value).
Cell apoptosis
KG-1 acute myelocytic leukemia cells were seeded and incubated as described in the growth inhibition experiment. Apoptotic cells were identified using acridine orange staining on at least three replicates as described.[77]
Supplementary Material
Supplementary Data
Acknowledgements
The (AOx)3-tk-LUC plasmid and human PPARγ expression vector were kind gifts from Professor Christopher K. Glass (University of California, San Diego). This work was supported by the following grants: DoD Prostate Cancer Research Grant W81XWH-04-1-0161 (M.I.D. and X.-K.Z.) and USPHS Grant R01 CA107039 (M.I.D. and X.-K.Z.).
Glossary
AMLacute myelocytic leukemia
DIMdi(1H-indol-3-yl)methyl
GALβ-galactosidase
Hhelix
LBDligand binding domain
LBPligand binding pocket
LUCluciferase
NBRENur77-binding response element
NGFI-Bnerve growth factor-induced clone B
NOR1neuron-derived orphan receptor 1
NRnuclear receptor
NurRENur response element
Nurr1Nur-related factor 1
PPARperoxisome proliferator-activated receptor
PPREPPAR response element
RARretinoic acid receptor
RAREretinoic acid response element
RXRretinoid X receptor

Footnotes
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.200800447.
1. Dawson MI, Hobbs PD, Jong L, Cameron JF, Lehmann JM, Fanjul A, Lu XP, Haefner P, Pfahl M. US Patent. 5 466 861. 1995.
2. Dawson MI, Jong L, Hobbs PD, Cameron JF, Chao WR, Pfahl M, Lee MO, Shroot B, Pfahl M. J. Med. Chem. 1995;38:3368–3383. [PubMed]
3. Boehm MF, Zhang L, Badea BA, White SK, Mais DE, Berger E, Suto CM, Goldman ME, Heyman RA. J. Med. Chem. 1994;37:2930–2941. [PubMed]
4. Boehm MF, Heyman RA, Zhi L, Koch SC. US Patent . 6 320 074. 2001.
5. Hurst RE. Curr. Drug Metab. Curr. Opin. Investig. Drugs. 2000;1:514–523. [PubMed]
6. Smit JW, Stokkel MP, Pereira AM, Romijn JA, Visser TJ. J. Clin. Endocrinol. Metab. 2007;92:2496–2499. [PubMed]
7. Heald P. Clin. Lymphoma. 2000;1 Suppl. 1:S45–S49. [PubMed]
8. Blumenschein GR, Jr, Khuri FR, von Pawel J, Gatzemeier U, Miller WH, Jr, Jotte RM, Le Treut J, Sun SL, Zhang JK, Dziewanowska ZE, Negro-Vilar A. J. Clin. Oncol. 2008;26:1879–1885. [PubMed]
9. Wildi JD, Baggstrom MQ, Suresh R, Read W, Fracasso PM, Govindan R. Chemotherapy. 2008;54:125–130. [PubMed]
10. Esteva FJ, Glaspy J, Baidas S, Laufman L, Hutchins L, Dickler M, Tripathy D, Cohen R, DeMichele A, Yocum RC, Osborne CK, Hayes DF, Hortobagyi GN, Winer E, Demetri GD. J. Clin. Oncol. 2003;21:999–1006. [PubMed]
11. Duvic M, Martin AG, Kim Y, Olsen E, Wood GS, Crowley CA, Yocum RC. Arch. Dermatol. 2001;137:581–593. [PubMed]
12. James SY, Lin F, Kolluri SK, Dawson MI, Zhang X-K. Cancer Res. 2003;63:3531–3538. [PubMed]
13. Kuang AA, Cado D, Winoto A. Eur. J. Immunol. 1999;30:3722–3728. [PubMed]
14. Cho SD, Yoon K, Chintharlapalli S, Abdelrahim M, Lei P, Hamilton S, Khan S, Ramaiah SK, Safe S. Cancer Res. 2007;67:674–683. [PubMed]
15. Li H, Kolluri SK, Gu J, Dawson MI, Cao X, Hobbs PD, Lin B, Chen G, Lu J, Lin F, Xie Z, Fontana JA, Reed JC, Zhang X. Science. 2000;289:1159–1164. [PubMed]
16. Lin B, Kolluri SK, Lin F, Liu W, Han YH, Cao X, Dawson MI, Reed JC, Zhang X-K. Cell. 2004;116:527–540. [PubMed]
17. Cao X, Liu W, Lin F, Li H, Kolluri SK, Lin B, Han YH, Dawson MI, Zhang X-K. Mol. Cell. Biol. 2004;24:9705–9725. [PMC free article] [PubMed]
18. Chintharlapalli S, Burghardt R, Papineni S, Ramaiah S, Yoon K, Safe S. J. Biol. Chem. 2005;280:24903–24914. [PubMed]
19. Wansa KD, Harris JM, Muscat GE. J. Biol. Chem. 2002;277:33001–33011. [PubMed]
20. Ordentlich P, Yan Y, Zhou S, Heyman RA. J. Biol. Chem. 2003;278:24791–24799. [PubMed]
21. Wansa KD, Harris JM, Yan G, Ordentlich P, Muscat GE. J. Biol. Chem. 2003;278:24776–24790. [PubMed]
22. Kagaya S, Ohkura N, Tsukada T, Miyagawa M, Sugita Y, Tsujimoto G, Matsumoto K, Saito H, Hashida R. Biol. Pharm. Bull. 2005;28:1603–1607. [PubMed]
23. Codina A, Benoit G, Gooch JT, Neuhaus D, Perlmann T, Schwabe JW. J. Biol. Chem. 2004;279:53338–53345. [PubMed]
24. Hintermann S, Chiesi M, von Krosigk U, Mathe D, Felber R, Hengerer B. Bioorg. Med. Chem. Lett. 2007;17:193–196. [PubMed]
25. Poppe L, Harvey TS, Mohr C, Zondlo J, Tegley CM, Nuanmanee O, Cheetham J. J. Biomol. Screening. 2007;12:301–311. [PubMed]
26. Wang Z, Benoit G, Liu J, Prasad S, Aarnisalo P, Liu X, Xu H, Walker NP, Perlmann T. Nature. 2003;423:555–560. [PubMed]
27. Flaig R, Greschik H, Peluso-Iltis C, Moras D. J. Biol. Chem. 2005;280:19250–19258. [PubMed]
28. Min JR, Schuetz A, Loppnau P, Weigelt J, Sundström M, Arrow-smith CH, Edwards AM, Bochkarev A, Plotnikov AN. RCSB Protein Data Bank. 2007 accession code 2QW4, DOI: 10.2210/pdb2qw4/pdb.
29. Okabe T, Nawata H. Nippon Rinsho. 1996;54:1768–1772. [PubMed]
30. Wu Q, Dawson MI, Zheng Y, Hobbs PD, Agadir A, Jong L, Li Y, Liu R, Lin B, Zhang X-K. Mol. Cell. Biol. 1997;17:6598–6608. [PMC free article] [PubMed]
31. Maira M, Martens C, Philips A. J. Drouin, Mol. Cell. Biol. 1999;19:7549–7557. [PMC free article] [PubMed]
32. Maxwell MA, Muscat GE. Nucl. Recept. Signal. 2006;4:e002. [PMC free article] [PubMed]
33. Chintharlapalli S, Smith R, III, Samudio I, Zhang W, Safe S. Cancer Res. 2004;64:5994–6001. [PubMed]
34. Chintharlapalli S, Papineni S, Baek SJ, Liu S, Safe S. Mol. Pharmacol. 2005;68:1782–1792. [PubMed]
35. Chintharlapalli S, Papineni S, Safe S. Mol. Pharmacol. 2007;71:558–569. [PubMed]
36. Ziouzenkova O, Orasanu G, Sukhova G, Lau E, Berger JP, Tang G, Krinsky NI, Dolnikowski GG, Plutzky J. Mol. Endocrinol. 2007;21:77–88. [PubMed]
37. Lin Q, Ruuska SE, Shaw NS, Dong D, Noy N. Biochemistry. 1999;38:185–190. [PubMed]
38. Yamazaki Y, Kawano Y, Uebayasi M. Life Sci. 2008;82:290–300. [PubMed]
39. Heim M, Johnson J, Boess F, Bendik I, Weber P, Hunziker W, Fluhmann B. FASEB J. 2002;16:718–720. [PubMed]
40. Lengqvist J, Mata de Urquiza A, Bergman AC, Willson TM, Sjövall J, Perlmann T, Griffiths WJ. Mol. Cell. Proteomics. 2004;3:692–703. [PubMed]
41. Mata de Urquiza A, Liu S, Sjöberg M, Zetterström RH, Griffiths W, Sjövall J, Perlmann T. Science. 2000;290:2140–2144. [PubMed]
42. Lemotte PK, Keidel S, Apfel CM. Eur. J. Biochem. 1996;236:328–333. [PubMed]
43. Egea PF, Mitschler A, Moras D. Mol. Endocrinol. 2002;16:987–997. [PubMed]
44. Lehmann JM, Jong L, Fanjul A, Cameron JF, Lu XP, Haefner P, Dawson MI, Pfahl M. Science. 1992;258:1944–1946. [PubMed]
45. Egea PF, Mitschler A, Rochel N, Ruff M, Chambon P, Moras D. EMBO J. 2000;19:2592–2601. [PubMed]
46. Bartoli G, Bosco M, Foglia G, Giuliani A, Marcantoni E, Sambri L. Synthesis. 2004;6:895–900.
47. Bandgar BP, Shaikh KA. Tetrahedron Lett. 2003;44:1959–1961.
48. IJpenberg A, Tan NS, Gelman L, Kersten S, Seydoux J, Xu J, Metzger D, Canaple L, Chambon P, Wahli W, Desvergne B. EMBO J. 2004;23:2083–2091. [PubMed]
49. Diab A, Hussain RZ, Lovett-Racke AE, Chavis JA, Drew PD, Racke MK. J. Neuroimmunol. 2004;148:116–126. [PubMed]
50. Yamazaki K, Shimizu M, Okuno M, Matsushima-Nishiwaki R, Kanemura N, Araki H, Tsurumi H, Kojima S, Weinstein IB, Moriwaki H. Gut. 2007;56:1557–1563. [PMC free article] [PubMed]
51. Zapata-Gonzalez F, Rueda F, Petriz J, Domingo P, Villarroya F, de Madariaga A, Domingo JC. J. Immunol. 2007;178:6130–6139. [PubMed]
52. Lu J, Chen M, Stanley SE, Li E. Biochem. Biophys. Res. Commun. 2008;365:42–46. [PMC free article] [PubMed]
53. Dawson MI, Jong L, Hobbs PD, Xiao D, Feng KC, Chao WR, Pan C, Fontana JA, Zhang X-K. Bioorg. Med. Chem. Lett. 2000;10:1311–1313. [PubMed]
54. Bernardon J-M, Diaz P. US Patent. 6 258 775. 2001.
55. Bernardon J-M, Diaz P. US Patent. 6 825 360. 2004.
56. Haffner CD, Lenhard JM, Miller AB, McDougald DL, Dwornik K, Ittoop OR, Gampe RT, Jr, Xu HE, Blanchard S, Montana VG, Consler TG, Bledsoe RK, Ayscue A, Croom D. J. Med. Chem. 2004;47:2010–2029. [PubMed]
57. Hibi S, Kikuchi K, Yoshimura H, Nagai M, Tai K, Hida T. J. Med. Chem. 1998;41:3245–3252. [PubMed]
58. Venhorst J, Nunez S, Terpstra JW, Kruse CG. J. Med. Chem. 2008;51:3222–3229. [PubMed]
59. Zhan Y, Du X, Chen H, Liu J, Zhao B, Huang D, Li G, Xu Q, Zhang M, Weimer BC, Chen D, Cheng Z, Zhang L, Li Q, Li S, Zheng Z, Song S, Huang Y, Ye Z, Su W, Lin S-C, Shen Y, Wu Q. Nat. Chem. Biol. 2008;4:548–556. [PubMed]
60. Sun S-Y, Yue P, Dawson MI, Shroot B, Michel S, Lamph WW, Heyman RA, Teng M, Chandraratna RAS, Shudo K, Hong WK, Lotan R. Cancer Res. 2007;67:4931–4939. [PubMed]
61. Kizaki M, Dawson MI, Heyman R, Elstner E, Morosetti R, Pakkala S, Chen DL, Ueno H, Chao W, Morikawa M, Ikeda Y, Heber D, Pfahl M, Koeffler HP. Blood. 1996;87:1977–1984. [PubMed]
62. Shiohara M, Dawson MI, Hobbs PD, Sawai N, Higuchi T, Koike K, Komiyama A, Koeffler HP. Blood. 1999;93:2057–2066. [PubMed]
63. Laidler P, Dulińska J, Mrozicki S. Arch. Biochem. Biophys. 2007;462:1–12. [PubMed]
64. Hsu JY, Pfahl M. Cancer Res. 1998;58:4817–4822. [PubMed]
65. Lei P, Abdelrahim M, Cho SD, Liu S, Chintharlapalli S, Safe S. Carcinogenesis. 2008;29:1139–1147. [PMC free article] [PubMed]
66. Cho SD, Lei P, Abdelrahim M, Yoon K, Liu S, Guo J, Papineni S, Chintharlapalli S, Safe S. Mol. Carcinog. 2008;47:252–263. [PubMed]
67. York M, Abdelrahim M, Chintharlapalli S, Lucero SD, Safe S. Clin. Cancer Res. 2007;13:6743–6752. [PubMed]
68. Hong J, Samudio I, Chintharlapalli S, Safe S. Mol. Carcinog. 2008;47:492–507. [PMC free article] [PubMed]
69. Safe S, Papineni S, Chintharlapalli S. Cancer Lett. 2008;269:326–338. [PMC free article] [PubMed]
70. Lane BS, Brown MA, Sames D. J. Am. Chem. Soc. 2005;127:8050–8057. [PubMed]
71. Kumar SP, Nagaiah K. Tetrahedron Lett. 2007;48:1391–1394.
72. Greenman KL, Van Vranken DL. Tetrahedron Lett. 2005;61:6438–6441.
73. Corey EJ, Székely I, Shiner CS. Tetrahedron Lett. 1977;18:3529–3532.
74. Sato S, Sato T. Carbohydr. Res. 2005;340:2251–2255. [PubMed]
75. Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK, Olson AJ. J. Comput. Chem. 1998;19:1639–1662.
76. Kolluri SK, Corr M, James SY, Bernasconi M, Lu D, Liu W, Cottam HB, Leoni LM, Carson DA, Zhang XK. Proc. Natl. Acad. Sci. USA. 2005;102:2525–2530. [PubMed]
77. Dawson MI, Xia Z, Liu G, Fontana JA, Farhana L, Patel BB, Arumugarajah S, Bhuiyan M, Zhang X-K, Han YH, Stallcup WB, Fukushi J, Mustelin T, Tautz L, Su Y, Harris DL, Waleh N, Hobbs PD, Jong L, Chao WR, Schiff LJ, Sani BP. J. Med. Chem. 2007;50:2622–2639. [PMC free article] [PubMed]
78. Ricote M, Li AC, Willson TM, Kelly CJ, Glass CK. Nature. 1998;391:79–82. [PubMed]