Novel d-Annulated Pentacyclic Steroids: Regioselective Synthesis and Biological Evaluation in Breast Cancer Cells

The acid-catalyzed cyclization of benzylidenes based on 16-dehydropregnenolone acetate (16-DPA) was studied. It was found that these compounds readily undergo regioselective interrupted Nazarov cyclization with trapping chloride ion and an efficient method of the synthesis of d-annulated pentacyclic steroids based on this reaction was proposed. The structures of the synthesized pentacyclic steroids were determined by NMR and X-ray diffraction. It was found that the reaction affords a single diastereomer, but the latter can crystallize as two conformers depending on the structure. Antiproliferative activity of synthesized compounds was evaluated against two breast cancer cell lines: MCF-7 and MDA-MB-231. All tested compounds showed relatively high antiproliferative activity. The synthetic potential of the protocol developed was illustrated by the gram-scale experiment.

II. Biological assays. General procedure.
The breast cancer cell lines MCF-7 and MDA-MB-231 were obtained from the ATCC collection and were used to evaluate the antiproliferative activity of the synthesized compounds. The cultivation of the cells was performed in standard (4.5 g/L glucose) DMEM medium (Gibco) supplemented with 10% fetal bovine serum (HyClone), 0.1 mg/ml sodium pyruvate (Santa Cruz), 50 U/ml penicillin, and 50 µg/ml streptomycin. Cells were incubated at 37°C in the presence of 5% CO 2 at a relative humidity of 85-90% in a NuAire incubator.
The synthesized compounds were dissolved in dimethyl sulfoxide (DMSO) to a concentration of 5 mM using sonication, and the solutions were stored at -20°C before use. The MCF-7 and MDA-MB-231 cells were seeded onto a 24-well plate (Corning) at a density of 4*10 4 cells per well. After growth overnight, the compounds were added to the cells in the concentration range from 1.5 to 25 µM. Cisplatin was used as the reference compound. An appropriate solvent (DMSO) volume was added to the control cells; the final concentration of the solvent in the medium was less than 0.5%. The antiproliferative activity was evaluated using the MTT assay, which is based on the reduction of the MTT reagent (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) in living cells giving violet formazan crystals insoluble in the culture medium; [S1] the MTT assay was performed in a modified version as described previously. [S2] After 72 h growth in the presence of the tested compounds, the medium was removed and the MTT reagent (AppliChem) was added to the cells. After 2 h incubation with the MTT reagent, the cells were lysed in 100% DMSO (AppliChem). The plate was shaken to dissolve the formazan crystals that formed. The absorbance of the solutions was measured on a MultiScan FC spectrophotometer at 571 nm. Then the blank absorbance values (media only) were subtracted from the sample absorbance values; the absorbance of the solutions of the control samples was taken as 100%. The IC 50 values were calculated as the concentration of the compound that decreases the absorbance of the solution by 50% compared to the control sample using the GraphPad software. The experiments were repeated in triplicate.

III. X-ray diffraction studies
X-ray diffraction data were collected at 100K on a Bruker Quest D8 diffractometer equipped with a Photon-III area-detector (graphite monochromator, shutterless and ω-scan technique), using Mo K α -radiation (0.71073 Å). The intensity data were integrated by the SAINT program [S3] and were corrected for absorption (semi-empirical from equivalents by multi-scan techniques) using TWINABS [S3] for 2b and SADABS [S4] for 2g.The structures were solved by direct methods using SHELXT/SHELXS-2013 [S5] and refined by full-matrix least-squares on F 2 using SHELXL-2018. [S6] All non-hydrogen atoms were refined with anisotropic displacement parameters. Hydrogen atoms were placed in ideal calculated positions and refined as riding atoms with relative isotropic displacement parameters.A rotating group model was applied for methyl groups.The studied crystal of 2b was refined as a 2-component twin with the twin law of [-1 0.01 0, 0.14 1 0.01, 0 0.01 -1] (the second domain was rotated from the first domain by 178° about reciprocal axis 0 1 0.01). The domain ratio was not found based on collected data since the total number of collected reflections (127213 at resolution down to 0.69Å) contained less than 0.18% single reflections (225) corresponding to the first domain and no single reflection for the second domain; all other reflections were composites of both domains. A non-coordinating methanol molecule in 2g was disordered over at least four overlapping positions with the overall occupancy of 0.5, forming infinite 1D chains in the crystal channels of 2g. This molecule was removed by the SQUEEZE method [S7] implemented in the PLATON program. [S8] The absolute structures of chiral centers unchanged in course of reactions were confirmed by anomalous X-ray scattering. The absolute structure parameter (Flack x) was determined by classical fit [S9] for the twinned crystal 2b:I H (calc)=(1-x)|F H (calc)| 2 +x|F -H (calc)| 2 , and by using 3396 quotients [S10] [(I+)-(I-)]/[(I+)+(I-)] for 2g. The SHELXTL program suite [S3] was used for molecular graphics.
Crystal data, data collection and structure refinement details for 2'b and 2''b are summarized in Table  S1.   (2'a) (2"a) (2'g)      (2) Cl(2)-C(23) 1.739(2) C(10)-C(11) 1.522 (2) Cl (3)  One of the reasons for the absence of conformer 2"g (similar in structure to 2"b) in the crystal lattice of 2g may be its significantly lower stability compared to 2'g. Either a very short distance C4-H4A···Cl2 or short contacts C5-H5···Cl2 and C6-H6···Cl2 should be present in 2"g, which significantly increases the intramolecular Van der Waals repulsion, but the rotation of the phenyl group about the C5-C22 bond (with a deviation of the C4-C5-C22-C27 torsion angle from its optimal value of 15.7-16.7°) cannot provide a decrease in the total intramolecular repulsion due to the appearance of additional Van der Waals interactions between the hydrogens of the terminal five-membered ring and the ortho-hydrohen atoms of the phenyl group.

IV.1. DFT calculations
The geometries of two conformers of 2b, established by X-ray diffraction, were optimized with GAUSSIAN09 [S11] using the ωB97X-D functional [S12] and the 6-31+G(d,p) basis set both in the gas phase and in chloroform (SMD-PCM continuum solvation model).
Calculation of vibrational frequencies was performed to prove that the optimized structure corresponds to a true minimum on the potential energy surface. The X-ray atomic coordinates were taken as the starting coordinates. The energies of the conformers, which were optimized in vacuum and in solution, and those determined by X-ray diffraction were calculated at the ωB97X-D/6-311++G(d,p) level of theory..
Signal (5.9 ppm) is estimated as CH of enolate and could be seen in CH 2 Cl 2 (in CDCl 3 not obtained this signal). As evidence we could see that equilibrium without acid turned into ketone (integral of the signal above 0.1-0.2) and with addition of 2,2 eq. of TiCl 4 the equilibrium shifts to enolate. Moreover we suggest, that this signal estimates to cyclohexenone ring, not to cyclopentanone due presence it in starting spectra of benzylidene 1c. Figure S5. 1 H-NMR-monitoring of compound 2b in 1,2-dichlorobenzene.
As you can see, the signal at ~ 2.4 ppm (at 300 K) does not disappear at 373 K, which is proof that this is not a set of two conformer doublets. Figure S6. 1 H-NMR-monitoring of 1c reaction with AlCl 3 in CH 2 Cl 2 Figure S7. Comparison of 1 H NMR spectra of the compounds 1c, 2c and crude reaction mixture of compound 1c. Figure S8. 1 H NMR spectra of the crude reaction mixture of compound 1c with TMS.
We used TMS as internal standard (Note: TMS should be added just before the spectrum registration) to calculate the yield of products. As seen the spectra of crude reaction mixture had incomplete conversion, but any extra signals of byproducts or another diastereomer (the area from 2.5 to 3.2 ppm are clear from another compounds) are not observed. Moreover the sum of the integrals of single proton from cyclopentanone ring 2c (red or orange marked) and divinylketone 1c (blue marked) converges with the integral of single proton from ring A (violet marked) which refers for both compounds.
VI. Copies of 1 H and 13 C NMR spectra.