Vitamin D receptor (VDR) is a nuclear receptor that acts as a ligand-dependent transcription factor in response to 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3: 1), a metabolically activated form of vitamin D3 [1,2]. The VDR plays a crucial role in regulating calcium and phosphate homeostasis, and is also involved in cell proliferation, differentiation, and immune responses [3,4]. Since VDR dysfunction has been linked to various diseases, including osteoporosis, psoriasis, arthritis, and inflammation, synthetic VDR ligands, such as falecalcitriol (2), calcipotriol (3), maxacalcitol (4), and eldecalcitol (5), have been developed based on the structure of 1, and used clinically for the treatment of those diseases (Fig. 1) [5,6]. In addition, it has been suggested that the VDR plays a role in certain cancers, such as prostate, breast and colon cancers [[7], [8], [9]]. Thus, it remains an attractive target for drug discovery.

Structurally, all the clinically used VDR ligands have a secosteroidal skeleton, consisting of the A-ring bearing two hydroxyl groups, a conjugated triene moiety, the CD-ring, and a side chain. Most of the synthetic ligands also have modifications in the side chain or in the A-ring of 1. Although the secosteroidal framework is a privileged structure, VDR ligands other than secosteroidal derivatives, i.e., nonsecosteroidal VDR ligands, would be useful as clinical drug candidates because they are expected to have greater chemical and metabolic stability and a smaller risk of adverse side effects as compared with secosteroidal compounds. Based on these considerations, several nonsecosteroidal VDR ligands have been developed and reported [10]. Lithocholic acid (6), which was found to be the second endogenous VDR ligand [11], is one promising lead compound for the further development of VDR ligands. Though the potency of 6 itself is quite weak, we recently developed lithocholic acid derivatives 7 as a novel VDR ligands with potency comparable to that of 1 [12,13]. Mouriño and coworkers also developed potent VDR ligands by structural modification of lithocholic acid [14]. As for ligands that are different from the endogenous ligands, diphenylmethane derivatives, originating from LG190178 (8) [15], are representative examples. Kashiwagi et al. developed a diphenylmethane derivative 9 which is more potent than 1 in transactivation, HL-60 cell differentiation, and osteocalcin assays [16]. We recently investigated the structure-activity relationship (SAR) of the diphenylsilane scaffold (10), and found that slight differences in the hydrophobic structure, namely, modification of the diethyl-di(m-tolyl) core, significantly affect the VDR agonistic activity (Fig. 2) [17].

As VDR ligands with a chemotype quite different from the conventional ligands described above, we have developed a series of carborane-based VDR ligands [[18], [19], [20], [21]]. Icosahedral carboranes, more specifically dicarba-closo-dodecaboranes, are polyhedral carbon-containing boron clusters containing a bulky spherical structure with high hydrophobicity [[22], [23], [24], [25], [26]]. There are three isomeric carboranes, namely, o-carborane, m-carborane, and p-carborane, differing in the positions of the two carbon atoms. They all exhibit high hydrophobicity, as well as having high thermal and chemical stability, and are therefore promising scaffolds for three-dimensional compounds, especially for hydrophobic biologically active compounds such as nuclear receptor modulators, including VDR ligands [[27], [28], [29], [30]]. Based on these considerations, we previously developed a series of carborane-based VDR ligands consisting of a hydrophobic p-carborane core and two flexible alkyl chains with three hydroxyl groups, such as 11 [18]. X-Ray crystallographic analysis of compound 11 bound to the rat VDR ligand-binding domain (LBD) revealed that the hydrophobic, spherical surface of carborane functions as a hydrophobic anchor in binding to the protein surface. The hydrophobic interaction between carborane and the receptor protein counteracted the disadvantageous flexibility of the molecule, indicating that the carborane cage is promising hydrophobic core of distinctive nonsecosteroidal VDR ligands. However, there is still room for further structural development to increase the activity by reducing the entropic disadvantage arising from the flexibility. Based on these considerations, we planned further structural development of these carborane-based VDR ligands.