Synthesis, biological evaluation and in silico and in vitro mode-of-action analysis of novel dihydropyrimidones targeting PPAR- g †

Hepatocellular carcinoma, a fatal liver cancer, a ﬀ ects 600 000 people annually and ranks third in cancer-related lethality. In this work we report the synthesis and related biological activity of novel dihydropyrimidones. Among the tested compounds, 5-acetyl-4-(1 H - indol-3-yl)-6-methyl-3,4-dihydropyrimidin-2(1 H )-one ( 4g ) was found tobemost activetowardsthe HepG2 cellline(IC 50 ¼ 17.9 m M), being at the same time 7.6-fold selective over normal (LO2) liver cells (IC 50 ¼ 136.9 m M). Subsequently, we identi ﬁ ed peroxisome proliferator-activated receptor g as a target of compound 4g using an in silico approach, and con ﬁ rmed this mode-of-action experimentally.

The peroxisome proliferator-activated receptor (PPAR) belongs to the nuclear hormone receptor superfamily of transcription factors that includes 48 human transcription factors. Its activity is regulated by direct binding of steroid and thyroid hormones, vitamins, lipid metabolites, and xenobiotics. 1 PPARs heterodimerize with retinoid X receptors, which get activated and bind to specic response elements in the target DNA of various target genes, 2,3 PPARs consist of three different isoforms, PPAR-a, PPAR-b/d, and PPAR-g. Due to their central function in a variety of physiological processes, PPARs are important targets in drug discovery. 4 Small molecules frequently interact with more than a single PPAR isoform leading to unique proles of their biological effects. 5 In particular, PPAR-g has been established as a key regulator for inammation, 6 proliferation, 7 metabolism, 8 and differentiation, 9 and it is also upregulated by many tumor suppressor genes. 10 PPAR-g is overexpressed in several types of human cancers, including breast, colon, bladder, and prostate cancer, and it also induces apoptosis in several malignant cell lineages. 11 Furthermore, PPAR-g ligands such as 15d-PGJ2 as well as antidiabetic thiazolidinediones (TZDs) function as anti-proliferative and proapoptotic agents, suggesting that PPAR-g could be a promising therapeutic target for the treatment of cancer. [12][13][14] Non-thiazolidinedione derivatives, such as 2-cyano-3,12-dioxooleana-1,9-dien-28-oic acid, GW-7845, JTT-501, KRP-297, L-764406, MCC-555 and GW-0207 are further known synthetic ligands of PPAR-g. 15 On the other hand, the natural product ectoine, a tetrahydropyrimidine derivative, offers protection against the effects of ischemia-reperfusion injury in intestinal transplantation in vivo. 16 (AE)-Aplicyanin analogs bearing a pyrimidone moiety showed signicant anti-tumor activity via targeting p38 MAP kinase. [17][18][19] Similar small molecules, dihydropyrimidines (DHPs), have also been reported as anti-cancer agents. 20 Recently, pyrimidine-tethered curcumins showed anticancer activity by targeting EGFR tyrosine kinase. 21 In continuation of our interest to synthesize novel bioactive heteocycles, [22][23][24] we herein report the synthesis of DHP tethered to various functional heterocycles like indole, avone and benzofuran. Furthermore, and conceptually rather novel, we rationalized the mode-of-action for the lead DHP as PPAR-g via an in silico cheminformatics approach, followed by experimental validation. Details on synthesis, compound chararacterization, biological assays, and in silico methods are described in more detail in the ESI. † For synthesis, multi-component reactions (MCRs) have been employed which constitute an efficient synthetic strategy for rapid and effective laboratory organic transformations, because products are prepared in a one-pot and single step approach and the diversity can be obtained directly by changing the reacting components. 25 Here, we synthesized the dihydropyrimidone bearing small molecules via multi-component Biginelli reaction. 26 The mechanism for the formation of the title compounds involves the condensation between the aldehyde and urea to form an iminium intermediate, which acts as electrophile for the nucleophilic addition of the ketoester enol, and the ketone carbonyl of the resulting adduct undergoes condensation with the amine group of urea to give the cyclized product (Fig. 1A). Also further heterocycles such as indole, avone and benzofuran moieties were successful employed for the preparation of novel dihydropyrimidones (see ESI Table 1 †). The protocol was effective with aromatic amines having electron donating groups, and products were identied based on IR, LCMS, 1 H NMR, and 13 C NMR spectra (see ESI †). In the next step, the cytotoxic effects of dihydropyrimidones was investigated in HepG2 cell lines using the MTT assay. The cells were treated with 0, 10, 25, 50, and 100 mM concentrations of dihydropyrimidones for 72 h. DHPs were found to inhibit the viability of HepG2 cells in a dose-and timedependent manner. Compound 4g, 4k and 4p were found to be most effective with IC 50 values less than 50 mM. In addition, we were able to identify that all the dihydropyrimidone series of compounds inhibited the viability of immortalized normal human liver cells, LO2, at higher concentrations. Compound 4g was found have an IC 50 value of 17.9 mM against HepG2 and has the highest, 7.6-fold selectivity over the normal LO2 liver cell line (IC 50 ¼ 136.93 mM; see ESI Table 2 † for more details). Therefore, we considered 4g (Fig. 1B) in more detail and studied its modeof-action in HepG2 cells.
Human protein targets were predicted for the most bioactive compound 4g, and of all predicted targets (ESI Table 3 †), PPARg was the only target with score of 26.02, exceeding the significance cut-off of 10. Given the involvement of PPAR-g in the induction of apoptosis and cancer development, as described in previous studies, 4-6 this target was selected for further analysis.
PPAR-g has been extensively shown to be associated with anti-cancer effects in a variety of cancer types including HCC. HepG2 cells were treated with different concentrations of compound 4g for 8 h and then analyzed for the expression of PPAR-g by western blot analysis. It was found that compound 4g could substantially increase PPAR-g expression in a dosedependent manner, with maximum effect at 15 mM concentration ( Fig. 2A). Compound 4g also increased PPAR-g expression in a time-dependent manner, with maximum activity at 8 h (Fig. 2B). Evidently, the DNA-binding assay for PPAR-g in nuclear extracts showed that compound 4g signicantly enhanced PPAR-g DNA binding ability in a time dependent manner (Fig. 2C).
DHPs are well known ligands for PPAR-g. In order to determine the interaction of dihydropyrimidones towards PPAR-g, the co-crystal structure of human PPAR-g in complex with rosiglitazone was considered for our studies. 27 Using Discovery Studio (DS) default tools and settings, the ligand binding domain (LBD) of PPAR-g was identied and ligands were prepared for docking utilizing the CDOCKER programme of the ligand-receptor interaction module of Discovery Studio version 2.5, the dihydropyrimidones were docked into the LBD of PPAR. The CDOCKER energies for all docked ligands (ESI Table 4 †) indicate that compound 4g bound to the LBD region with the lowest (and thus most favorable) score of À24.0 kcal mol À1 (ESI Table 3 †). Furthermore, the binding pose of compound 4g was found to overlap with the hydrophobic tail of TZDs (rosiglitazone; Fig. 2D). Compound 4g forms mainly hydrophobic contacts with residues at the LBD region of PPAR-g including  Ile281, Met348, Ile353, Val339, Cys285, Ile341, Ser342, Gly284, Tyr321, Ile326, Lys367, Phe363, Met364, Leu330, Ser289, and Arg288 (Fig. 2E). Thereby, CH-p interactions with Leu-330 and van der Waals contacts with the alkyl chain of Arg-288 are proposed as main driving forces of the ligand. This molecular docking study revealed that compound 4g could be a suitable non-TZD ligand for PPAR-g.
We investigated whether activation of PPAR-g in HepG2 cells by compound 4g leads to apoptosis. In HepG2 cells treated with compound 4g, there was a time-dependent reduction in the levels of procaspase-3 (Fig. 3A). This suggests that cleavage events had occurred, indicating the activation of caspase-3, the levels of which were shown to be increased aer 48 h. Activation of caspase-3 led to the cleavage of a 118 kDa Poly (ADP-ribose) polymerase (PARP) protein into an 85 kDa fragment (Fig. 3B). This result clearly suggests that compound 4g induces caspase-3-dependent apoptosis in HepG2 cells. In addition, compound 4g was found to downregulate the anti-apoptotic and proliferative proteins, Bcl-2 and Cyclin D1 in a time dependent manner (Fig. 3C), with maximum effect observed at 48 h. Pronounced expression of PPAR-g was demonstrated in HCC cells treated with rosiglitazone and such induction markedly suppressed the migration of HCC cells. 28 The effect of compound 4g on the migration of HepG2 cells was investigated using a wound healing assay. We found that compound 4g treatment signicantly suppressed the migration of HepG2 cells, and the pretreatment with GW9662, a pharmacological PPAR-g inhibitor, reversed the anti-migratory effects of compound 4g as shown in Fig. 3D and E. It is well known that PPAR-g ligands may have therapeutic value for the treatment of highly invasive breast cancer by targeting invasion. 29 An in vitro Bio-Coat Matrigel invasion assay was performed (BD Biosciences, San Jose, CA, USA), and it was found that upon treatment with compound 4g there was a signicant reduction in the number of cells that could invade the Matrigel coated polycarbonate membrane, indicating that compound 4g could indeed signicantly inhibit the invasive property of HepG2 cells (Fig. 3F and  G). Moreover, we found that pretreatment with the PPAR-g antagonist GW9662 reversed the anti-invasive potential of compound 4g in HCC cells, conrming that the activity is mediated through inhibition of PPAR-g pathway.
In conclusion, we herein report the synthesis of novel dihydropyrimidones and found 5-acetyl-4-(1H-indol-3-yl)-6-methyl-3,4-dihydropyrimidin-2(1H)-one (4g) to be most active against HCC cells. Furthermore, compound 4g down-regulated antiapoptotic and proliferative proteins such as Bcl-2 and Cyclin D1 in a time-dependent manner. In silico studies predicted PPAR-g as mode-of-action of compound 4g, which could be validated by in vitro studies. Molecular docking furthermore suggested the binding pose of 4g within the ligand binding domain of PPAR-g, which overlaps with the tail of rosiglitazone. Compound 4g inhibited the invasive property of HepG2 cells and the pretreatment with GW9662 reversed the anti-invasive potential of compound 4g in HCC cells in vitro, thereby conrming that the activity is mediated through inhibition of PPAR-g pathway. Hence, both in silico and experimental methods agree on the mode-of-action of compound 4g via PPAR-g.