Molecular Docking Studies of Spirostans as MAPK14 (P38α) Inhibitors and Their Potential Use against Cancer

Spirostans (SPs) are chemical products widely distributed in the plant kingdom; currently, they are studied by their medical applications. Cancer has a high incidence in humans; it reaches second place worldwide deaths. In molecular biology, it has been accepted that Mitogen- Activated Protein p38alpha Kinase (MAPK14 (p38α) is implicated in the regulation of cancer. This study aimed to identify SPs as potential MAPK14 (p38α) inhibitors. From a set of 133 modi fied SPs, SwissTargetPrediction platform, and molecular docking, it was obtained that 129 chemical structures had molecular interaction with the MAPK14 (p38α). From those molecules, 123 were bound to a specific inhibition site of MAPK14 (p38α), and 6 of th e structures resulted in inhibitors similarly to minocycline and dasatinib. One SP had binding couple energy (BCE, kcal/mol) as that of fostamatinib. In addition, 115 modified SPs had better BCE than the minocycline but not as that using fostamatinib. The key amino acids (aa) for the protein kinase MAPK14 (p38α) inhibition were Arg 70, Asp 168, Lys 53, His 148, and Ile 145, at a different interaction level. The BCE was enhanced when the H atom was substituted in C-2, C-11, and C-17 SPs positions. Similarly, the αOH group at C -5 and C-6 upgraded BCE. Stereochemistry and substitution at C-3, C-12, and C-25 did not present significant differences (Kruskal-Wallis test, p <0.05). From all this ensemble of results, be


INTRODUCTION
Spirostans (SPs) are obtained from the acidic hydrolysis of the corresponding saponins, which are abundant in the plant kingdom 1 . Some SPs (Figure 1) have shown biological activity in many human pathologies such as polycystic ovary syndrome, metabolic syndrome, rheumatoid arthritis, breast, cervical, and prostate cancer, among others 2-9 . The SPs bioactivity has been attributed to their chemical structure; in this order, specifically to substituents in various stereocenters of the steroidal nucleus and the ring F. The scientific literature has also reported the important role of SPs anticancer molecules 10-13 . Based on characteristics described previously, the SPs are interesting molecules and can be studied against cancer. In cancer, cells are exposed to constant changes by the local environment, changes in metabolism, and general, in cell physiology. Recent data have proposed that the Mitogen-Activated Protein Kinase family (MAPKs) is involved in the cancer process that affects many tissues involved in human diseases. The MAPK family, specifically MAPK (p38) proteins that include MAPK14 (p38α), MAP11 (p38β), MAPK12 (p38ϒ and

Molecular Docking Studies of Spirostans as MAPK14 (P38α) Inhibitors
and Their Potential Use against Cancer MAPK13 (p38δ) was reported in alterations in the growth signals, protection, and apoptotic cell death, de novo angiogenesis, and metastasis 14 .   (Figure 2) can open a new way to understand their action mechanism and test more accurately new compounds as a new treatment against cancer.

Receptors
The receptor used as the molecular target was MAPK14 (p38α) bound to SR348, downloaded from protein data bank (https://www.rcsb.org/) as PDB code 6ZWP 25 .

Docking protocol
The optimized energy and geometry for the ligand structures were calculated using MM2 26

Assessment
The biological potential for SPs was evaluated through STP. Then, the molecular coupling studies of SPs were analyzed compared to drugs reported in the DrugBank database. First, in a punctual way towards specific sites of MAPK14 (p38α) and interactional study energy, later, the binding coupling energy (BCE) values were obtained. The results were compared with Kruskal-Wallis 27 , and p <0.05 was considered statistically different. The describers were defined with a significant effect on the BCE value (Figure 3).

RESULTS AND DISCUSSION
For commercial MAPK14 (p38α) inhibitors, minocycline, dasatinib, and fostamatinib, the BCE obtained were -8.0, -8.2, and -9.8 kcal/mol, respectively (Figure 4). These molecules were joined to the MAPK14 (p38α) specific site by interactions with the amino acid residues Lys A:53, Arg A:67, Leu A:74, Met A:78, Ile A:146, His A:148, Leu A: 167 and Asp A:168; in each case, it was different interaction types on the MAPK14's inhibition site in comparison to reference compounds. The BCE value was related to the interaction type and the amino acid residue it interacts with. Asp A:168 interacted by a conventional hydrogen bond with minocycline and dasatinib. Lys A:53 had the same type of bond with dasatinib and fostamatinib, and the ΔBCE~1.0 kcal/mol between minocycline and fostamatinib was related to amino acids residues Arg A:149 and Arg A:67 with unfavorable acceptor-acceptor and bump interactions, Asp A:168 and Glu A:71 was by sulfur-X interactions. In our studied SPs group (n=133); 26% are naturals, 24% synthetic and 50% are not reported so far. Two SPs presented the same BCE to minocycline. One is better but smaller than the dasatinib, six with similar BCE values, 123 with better BCE than dasatinib but worse than fostamatinib, and just one with the same energy as the last one ( Figure 5). The SPs present better potential inhibitory effect for MAPK14 (p38α) than dasatinib and minocycline but not than fostamatinib. Besides, 98% of the SPs were in the MAPK14 (p38α) specific site, between the central loop and the central ahelices of the A chain (Figure 6), except in three SPs with the worst BCE value. The arrangement of these compounds was found in the β-stacked sheets supported by the spiroketal moiety towards the nearby loop. In 3D models, the interaction site was amphipathic, the internal zone was hydrophobic, and the external zone hydrophilic. The steroids were located on the part of the steroidal nucleus in the hydrophobic zone and the oxygens of the spiroketal in the hydrophilic region, except for the 6-oxygenated derivatives compounds which have more significant molecular interaction on the hydrophilic site. The BCE depends on the interactions of each one of the compounds present with the target protein. Figure 7 shows the selected SPs with the descriptors, demonstrating the effect caused on the coupling force by a functional group with specific stereochemistry. In C-2, the coupling improvement SP-1, which has hydrogen in this carbon, due to an acetyl group in equatorial position (SP-17), promotes hydrogen bridge formation with the protein. In contrast, for SP-24, whose structural difference is the stereochemistry of C-3 in the axial position, the coupling force decreased even more than SP-1 and can be explained by the interactions Van der Waals. For the stereochemistry of C-25, better interaction was observed between amino acid residues and SP-114 and SP-123 with equatorial methyl with p-alkyl type interaction. Unlike SP-90, whose methyl is axial, which only has p-s interactions, these SPs presented better coupling than SP-1 and SP-24, which has a double bond at C-5 described before, comparing the hydrogen interactions in this same position was observed similarity by their nature.   For the C-12, the change in hybridization and polarity in the C ring rearranged the interaction with the MAPK14 (p38α) specific site. For SP-114 and SP-123, van der Waals and p-alkyl interactions stand out. For the first one, it is important to highlight that having equatorial hydrogen at C-5 and axial hydroxyl at C-6 was better for an unfavorable interaction with hydroxyl at C-3 28,29 . SS-123 with an axial hydroxyl group at C-5 promoted the hydrogen bridge-type interaction with the protein target, leaving the interaction with the carbonyl at C-6 and the hydroxyl at C-3. The grade of interactional similarity was found by analyzing the residues of the MAPK14 (p38α) with the commercial inhibitors and SPs interact ( Table I).
However, SP-24 presented the same BCE as minocycline; SP-24 only interacts with Met A:78 because the SP is outside the active site. For dasatinib, whose BCE was -8.2 kcal/mol, its energetic counterpart was SP-1. However, SPs had stronger hydrogen bridge and sulfur interactions than the control inhibitor. The interaction with amino acid residues presented similarity close to 80%, characterizing a compound with similar BCE to the commercial inhibitor drug without sulfur and nitrogen atoms in its structure. The same benefit can be observed in SP-17 and SP-90, whose BCE value was between dasatinib and fostamatinib energy. Besides, SP-90 presented a similar pattern in the interaction residues, whereas SP-17 presented interactional differences with the residues due to its low similarity, only in 33%. The presence of hydrogen bridges was shown with high coupling energy. For the fostamatinib energetic to SP-114, it is observed that the interaction residues have a close 80% match with van der Waals and p-alkyl type interactions. However, in this case, the unfavorable interaction with Arg A:189 stands out, which benefits the molecular coupling despite not being on the residues list of the commercial inhibitor. In the case of the SP-123 interactions (-9.7 kcal/mol), energetically was close to the commercial inhibitor fostamatinib, whose interactional match was 70%, this high value can be attributed to hydrogen bridge with Asp A:168, which is on the residues list of fostamatinib. for the inhibition of MAPK14 (p38α), a target important for the cancer treatment. These data, which also corroborated with the SP, could highlight the Asp A:168 residue whose interaction was observed with all the structures analyzed in the previous stage; Met A:78, which is only absent in one of the SPs. However, it is the only main residue observed with the SP-24 outside the active site, and its interaction remains due to the location of the amino acid in the lower zone of the α-helix.
Detailed studies permitted us to know the interaction between SPs and MAPK14 (p38α). Additionally, statistical analysis allows us to analyze the effect of each descriptor 22 . Non-parametric Kruskal-Wallis statistics were carried out for data that were not normally distributed, which was summarized in Table II. There were no changes in the stereogenic centers at C-3, C-12, and C-25. In C-25 (Figure 8), the side chain is exposed towards the area between chains A and B, which explains the null effect; the same case is for R2. In the case of C-12, the C ring is placed in this space between the chains, generating a null effect on the BCE 30 .
For descriptors with significant differences to the exposed region, the steroids compounds are coupled over the side, which explains why the modifications on it a more remarkable influence have, as the case of C-6 ( Figure 9) when this was oxygenated 31 . The position was exchanged, leaving the hydrophobic part of compounds exposed to the hydrophobic and hydrophilic sites, which considerably increased in the BCE. The same effect occurred in the C-5 hydroxyl group; it increased the hydrophilic region in rings A and B. In the case of substitution by hydrogen, there was no difference in its stereochemistry because it preserves the no polar environment of the region, but when it did present a double bond, the BCE decayed due to the hybridization change decreasing the contact area of receptor 32 . In C-17, it did present better energy when it has hydrogen as a substituent because it changes the polar environment in the D and E ring region, becoming polar when it is in the hydrophobic region 33 . At C-11, the carbonyl group or the hydrogen atom showed better BCE due to the hydrophobic region or only the hydrogen bond donor, but not the acceptor, as with hydroxyl substitutions in the molecular structure.

CONCLUSION
Molecular docking studies predict a favorable interaction between MAPK14 (p38α) and SPs, with BCE in between the range of two widely used commercial inhibitors. Only two of the 133 SPs did not bind at the specific MAPK14 (p38α) protein site. This result can be attributed to the most repetitive interaction van der Waals type between the key amino acids: Lys 53, Arg 67, Arg 70, Leu 74, Met 78, Ile 46, His 148, Leu 167, and Asp 168, also the carbonyl, hydroxyl, hydrogen, and methyl substituents which are in specific positions in the study compounds. Hydrogen bond interaction between Asp 168 and the hydroxyl group of SP-123 presented the best coupling with the MAPK14 (p38α) protein. From the analysis of interaction energies, it was very favorable the presence of hydrogen in C-2, C-17, and C-11. In addition, the last one also improved being a carbonyl group. The same effect was observed in C-6 and C-5; the presence of a hydroxyl group favored the interaction energy. These results support that SPs 17, 114 y 123 were the better MAPK14 (p38α) inhibitors and suggest that they can be potential therapeutic agents in cancer treatment.

CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare that are relevant to the content of this article.

DATA AVAILABILITY
All data are available from the authors. The software tools and version for docking analyses and specifics modules used. AutoDock Vina and SwissTargetPrediction can be used with a free license.