Cyclic 1H-Phospolane Oxides as a Potential Candidate for Cancer Therapy

: Organophosphorus compounds have been investigated for agricultural and medicinal applications for decades, and a considerable number of phosphorus-containing drugs have achieved commercial success. A recent review by P. Finkbeiner et al. has shown that phosphine oxides and related phosphorus-containing functional groups are valuable polar structural elements and that they deserve to be considered as a routine part of every medicinal chemist’s toolbox. A new approach to the synthesis of previously hard-to-obtain 3-alkyl-1H-phospholanes oxides was developed by us. In order to assess the potential of ﬁve-membered cyclic organophosphorus compounds in cancer therapy, we carried out docking 3-buthyl-1H-phospholanes oxide and 2,3-dihydrophosphole in the binding site of 24 human proteins involved in oncogenesis processes. Proteins were selected using the PharmMapper in-house pharmacophore model database. The results are presented in the article.


Introduction
It is well known that organophosphorus compounds are used in medicine, moreover, a significant number of phosphorus-containing drugs have achieved commercial success [1]. Recently, a review by P. Finkbeiner has shown [2] that most of the approved phosphoruscontaining pharmaceuticals, for example drugs 1-6, contain a phosphate, a phosphoramide, or a phosphonate group, while phosphines, phosphinates, and phosphine oxides are rare (Scheme 1). For example, the phosphinate-based drug used to treat hypertension is fosinopril (7). Recently, ridaforolimus (12), a dimethylphosphinic ester containing inhibitor of mammalian target of rapamycin (mTOR), progressed into phase III clinical studies for the treatment of sarcoma, and the anaplastic lymphoma kinase (ALK) inhibitor brigatinib (13) became the first drug containing a phosphine oxide motif that was approved for the treatment of patients with metastatic non-small-cell lung cancer (NSCLC) [3,4].
At the same time, new approaches to the synthesis of previously undescribed cyclic phospholane oxides are being developed.

Scheme 1. Selected examples of phosphorus-containing drugs.
We have accumulated significant experience in the development of effective one-pot methods for the synthesis of five-membered phosphacarbocycles via transmetalation of aluminacarbocycles, obtained by catalytic cycloalumination [5][6][7]  The synthesized compounds are chemically stable and may be promising in cancer therapy. In order to predict the biological properties for oncotherapy of a number of phospholane oxides, we screened using the PharmMapper. Then docking was employed using AutoDock to find out the mechanism of binding of the macromolecular targets to We have accumulated significant experience in the development of effective one-pot methods for the synthesis of five-membered phosphacarbocycles via transmetalation of aluminacarbocycles, obtained by catalytic cycloalumination [5][6][7] of olefins with AlEt 3 in the presence of Cp 2 ZrCl 2 as a catalyst, by PCl 3 (Scheme 2). We have accumulated significant experience in the development of effective one-pot methods for the synthesis of five-membered phosphacarbocycles via transmetalation of aluminacarbocycles, obtained by catalytic cycloalumination [5][6][7]  The synthesized compounds are chemically stable and may be promising in cancer therapy. In order to predict the biological properties for oncotherapy of a number of phospholane oxides, we screened using the PharmMapper. Then docking was employed using AutoDock to find out the mechanism of binding of the macromolecular targets to The synthesized compounds are chemically stable and may be promising in cancer therapy. In order to predict the biological properties for oncotherapy of a number of phospholane oxides, we screened using the PharmMapper. Then docking was employed using AutoDock to find out the mechanism of binding of the macromolecular targets to small active components under consideration, which made it possible to determine the role of the P=O(H) group in the interaction with targets.

Methods
A search for potential protein targets for the studied ligands was carried out using the PharmMapper in-house pharmacophore model database [8]. For this, the optimized ligand structures were saved as SDF files, which were then uploaded to a web server available at http://www.lilab-ecust.cn/pharmmapper/ (accessed on 22 October 2022). Pharmacophore mapping was carried out for the human protein targets set. From the resulting list of the potential human protein targets, only those involved in the processes of oncogenesis were selected for further study. AutoDock Tools (ADT) version 4.2.6 was used to carry out protein-ligand docking simulations [9]. The Discovery Studio Visualizer (version 21.1.0.20298, Dassault Systèmes, San Diego, CA, USA) [10] software was used to visualize the docking results.

Results and Discussion
The potential human protein targets were identified for model compound 15a. Two diastereomers were taken into consideration with energy energetically lowest twist conformation (Scheme 3). The screening results showed 17 ranked targets listed in Table 1, confirming the possibility of interaction between the model compound and some indications. The highest fit scores for both isomers was characterized the androgen receptor, which is a member of the steroid/nuclear receptor superfamily and which functions as a transcription factor [11]. This receptor is activated by binding to androgenic hormones that regulate male sex development [12]. Reactivation of the androgen receptor occurs in recurrent prostate cancer [13], making this protein a potential target for prostate cancer therapy.
Chem. Proc. 2022, 12, 49 3 of 6 small active components under consideration, which made it possible to determine the role of the P=O(H) group in the interaction with targets.

Methods
A search for potential protein targets for the studied ligands was carried out using the PharmMapper in-house pharmacophore model database [8]. For this, the optimized ligand structures were saved as SDF files, which were then uploaded to a web server available at http://www.lilab-ecust.cn/pharmmapper/ (accessed on 22 October 2022). Pharmacophore mapping was carried out for the human protein targets set. From the resulting list of the potential human protein targets, only those involved in the processes of oncogenesis were selected for further study. AutoDock Tools (ADT) version 4.2.6 was used to carry out protein-ligand docking simulations [9]. The Discovery Studio Visualizer (version 21.1.0.20298, Dassault Systèmes, San Diego, CA, USA) [10] software was used to visualize the docking results.

Results and Discussion
The potential human protein targets were identified for model compound 15a. Two diastereomers were taken into consideration with energy energetically lowest twist conformation (Scheme 3). The screening results showed 17 ranked targets listed in Table 1, confirming the possibility of interaction between the model compound and some indications. The highest fit scores for both isomers was characterized the androgen receptor, which is a member of the steroid/nuclear receptor superfamily and which functions as a transcription factor [11]. This receptor is activated by binding to androgenic hormones that regulate male sex development [12]. Reactivation of the androgen receptor occurs in recurrent prostate cancer [13], making this protein a potential target for prostate cancer therapy. The receptor was selected for the molecular docking simulation (Figure 1). Accordingly, the bioactive molecule in lowest conformation forms intermolecular interactions between the P=O group and the residues. The active sites of binding region in the receptor for the co-crystallized structure, which taken for comparison, were differ. The receptor was selected for the molecular docking simulation (Figure 1). Accordingly, the bioactive molecule in lowest conformation forms intermolecular interactions between the P=O group and the residues. The active sites of binding region in the receptor for the co-crystallized structure, which taken for comparison, were differ. An free binding energy and final intermolecular energy, as well as an inhibition constant for each of the docked bioactive molecules, were estimated ( Table 2). In terms of inhibitory activity, phospholane is clearly lower to the co-crystalized ligand (FBE = −10.04 kcal/mol, Ki = 43.69 nM).  In the case of RR phospholane interaction with the active site of the androgen receptor, the hydrogen bonds were formed with the P=O functional group. Out of the total interactions, there was a lack of hydrophobic contacts, obviously; therefore, with an increase in chain length of the alkyl substituent, an increase in the binding energy was observed. It should be noted that the effect of stereochemistry on the energy parameters was also manifested. Moreover, we have docked the tautomeric form P-OH [14], which An free binding energy and final intermolecular energy, as well as an inhibition constant for each of the docked bioactive molecules, were estimated (Table 2). In terms of inhibitory activity, phospholane is clearly lower to the co-crystalized ligand (FBE = −10.04 kcal/mol, Ki = 43.69 nM). In the case of RR phospholane interaction with the active site of the androgen receptor, the hydrogen bonds were formed with the P=O functional group. Out of the total interactions, there was a lack of hydrophobic contacts, obviously; therefore, with an increase in chain length of the alkyl substituent, an increase in the binding energy was observed. It should be noted that the effect of stereochemistry on the energy parameters was also manifested. Moreover, we have docked the tautomeric form P-OH [14], which can exist at the equilibrium concentration (denoted as 15 ) known for phosphine oxides (Table 2).

Conclusions
In summary, the potential anticancer activity for new 1H-phospolane oxides was identified. The androgen receptor was selected for the molecular docking simulation, as a result a binding site between the P=O and protein was shown. It was found that the design of the substituent in position 3 helped to model the binding activity.