Synthesis, Characterization and in vitro Anticancer Evaluation of 5‐Sulfinyl(sulfonyl)‐4‐arylsulfonyl‐Substituted 1,3‐Oxazoles

A novel series of 5‐sulfinyl(sulfonyl)‐4‐arylsulfonyl‐substituted 1,3‐oxazoles has been synthesized, characterized and subjected to NCI in vitro assessment. Cancer cell lines of all subpanels were most sensitive to 2‐{[4‐[(4‐fluorophenyl)sulfonyl]‐2‐(2‐furyl)‐1,3‐oxazol‐5‐yl]sulfinyl}acetamide (3 l). Its antiproliferative and cytotoxic activity averaged over each subpanel was manifested in a very narrow range of concentrations (GI50: 1.64–1.86 μM, TGI: 3.16–3.81 μM and LC50: 5.53–7.27 μM), i. e. practically did not depend on the origin of the cancer cell line. The COMPARE matrix using TGI vector showed a high positive correlation of 3 l (r=0.88) with the intercalating agent aclarubicin, which inhibits topoisomerases. The absence in the database of standard agents that have a high correlation with the cytotoxicity of this compound suggests that it may have a unique mechanism of action. According to the results of the docking analysis, the most promising anticancer target for compound 3 l is DNA topoisomerase IIβ. The obtained results indicate the anticancer activity of 5‐sulfinyl(sulfonyl)‐4‐arylsulfonyl‐substituted 1,3‐oxazoles, which may be useful for the development of new anticancer drugs. 2‐{[4‐[(4‐Fluorophenyl)sulfonyl]‐2‐(2‐furyl)‐1,3‐oxazol‐5‐yl]sulfinyl}acetamide (3 l), as the most active, can be recommended for further in‐depth studies.


Introduction
Due to the structural and chemical diversity of molecules synthesized from 1,3-oxazole and, consequently, their ability to interact with different molecular targets of the cell, 1,3-oxazole derivatives occupy one of the central position in medicinal chemistry. [1][2][3] This is reflected in the manifestation of their broad biological activity and demonstrates the massive potential of such derivatives in the discovery of new potential therapeutic agents. [4,5] This is confirmed by the wide variety of oxazole-containing compounds used as therapeutic drugs, presented in DrugBank (https://go.drugbank.com/), which are actively used throughout the world. However, it should be noted that only 3.4 % of all cancer drug development programs from 2000 to 2015 received regulatory approval, even though oncology accounted for 42 % of all drug development programs in this dataset. [6] Therefore, the determination of the sensitivity of human tumor cells to newly synthesized compounds provides only preliminary in vitro data on their anticancer activity. Cell culture studies are of limited utility because they do not reflect the heterogeneity of the cancer itself. It is clear that factors other than the in vitro chemosensitivity of tumor cells significantly affect the efficacy of in vivo chemotherapy (e. g., tumor microenvironment and/or lack of specific targeting ability, etc.). Such factors are not present in the in vitro cell line screening assay, but are a prerequisite for the selection of the most active compounds for further in vivo testing.
Important targets for search and development of anticancer drugs are DNA topoisomerases, in particular DNA topoisomerases type I and II. [7] Examples of efficient anticancer compounds with topoisomerase activity are camptothecin, topotecan, etoposide, doxorubicin, amsacrine, merbarone, and dexrazoxane. [8] DNA topoisomerases are nuclear enzymes that regulate DNA topology during the processes of transcription and replication. Thus, topoisomerases are convenient therapeutic targets and specific biomarkers for proliferating cancer cells. [9] Taking into account the pronounced antitumor activity of aryloxazoles, [10,11] we have synthesized new 5-sulfinyl(sulfonyl)-4-arylsulfonyl-substituted 1,3-oxazoles, hoping that they will exhibit a significant antitumor activity.
The synthesis of 5-arylsulfonyloxazoles 9 is presented on Scheme 2. The three-stage reaction sequence involves treatment of N-(2,2-dichloro-1-(arylsulfonyl)vinyl)amides 1 with a threefold excess of arenethiol 5 in the presence of triethylamine to obtain N-(1,2,2-tris(p-tolylthio)vinyl)benzamides 6 followed by cyclization in the presence of thionyl chloride to form 2-aryl-4,5-bis(arylthio)oxazoles 8. The mechanism of this transformation has not been studied in detail, but it is clear that thionyl chloride is a strong electrophilic agent capable of interacting with the S-nucleophilic centers of enamides (transition states 6 and 7), which can lead to a decrease in the electron density at the C-2 center of the alkenyl fragment and contribute to the nucleophilic attack on this center by the oxygen atom of the acylamine residue.
Compounds 8 were converted into the corresponding 5sulfonyl derivatives 9 by oxidation with hydrogen peroxide. [13] Data of synthesized novel 1,3-oxazole derivatives are presented in the SI. NMR ( 1 H NMR and 13 C NMR), chromato-mass and elemental analyses reliably confirm the structure of the obtained compounds.
The synthesized compounds 3, 4, 9 were submitted for in vitro anticancer screening at the National Cancer Institute (NCI), USA, against full NCI 60 cell lines panel and were granted the NCS codes shown in Table S1.

One-dose assay
One-dose assay mean graph data are presented in Table S1.
Among the synthesized compounds, derivatives 3 m (94 growth % for the total panel) and 9 b (104 %) practically did not inhibit the growth of cancer cells (mean graph data not shown). According to the number of sensitive lines, all compounds were arranged in a row: more than 80 % of the panel cell lines were highly sensitive. Based on the above statistical analysis, these 7 compounds were selected for a five-dose assay.
It should be noted that according to the one-dose assay results, among sulfonylsulfinyl derivatives (3 a, 3 i, 3 l and 3 m), the presence of furyl at the 2 nd position, sulfinylacetamide at the 5th position and simultaneous fluorination of the phenyl ring at the 4th position of oxazole (compound 3 l) significantly increased both the activity and the number of sensitive lines. Whereas the replacement of the furyl cycle with a thienyl at the 2 nd position and sulfinylacetamide with sulfinylphenylethanone at the 5 th position of oxazole (3 m) completely eliminated the cytotoxic activity.
Among the active disulfonyl derivatives, neither changes of aryl substituents at the 2 nd and 4 th positions, nor variations of the sulfonyl fragment at the 5 th position of the oxazole did not change the number of cell lines sensitive to the resulting compounds.
Regarding the inhibitory activity of these derivatives, the replacement of the methylsulfonyl in compound 4 a with oxopyrrolidinethylsulfonyl (4 b) or sec-butylsulfonyl (4 c) increased the cytotoxicity of the last compounds. At the same time, phenyl methylation and demethylation at the 2 nd and 4 th positions of compound 4 c, respectively, did not significantly affect the antitumor activity. The same result was obtained by replacing the methylsulfonyl in 4 d with butylsulfonyl and simultaneous fluorination of the phenyl at the 4 th position of the oxazole, giving 4 e. When the phenyl at the 4 th position of compound 4 f was chlorinated to give 4 g, the last one lost its cytotoxicity and exhibited only a cytostatic effect. A similar result was obtained by replacing the methylsulfonyl in compound 4 j with N-benzylsulfonyl-Nmethylacetamide, giving 4 k. The derivative in which the oxazole scaffold has been functionalised at the 4 th and 5 th positions with methylphenylsulphonyl gave compound 9 b inactive against all cell lines of the general panel. However, chlorination of the phenyls at the 4 th and 5 th positions instead of methylation slightly increased the activity of the resulting derivative 9 a.
It should be noted that due to the simultaneous functionalization of oxazole with various substituents at all positions (2, 4 and 5), a comparative structural-functional analysis of oxazole sulfonyl and sulfinyl derivatives is impossible.

Five-dose assay
The results of the antitumor activity of these compounds against cancer cell lines in a five-dose assay are presented in Table S2. For a comparative assessment of the antitumor activity of the compounds, a statistical analysis of their effect on the tested parameters of the total NCI panel ( Table 1) and individual subpanels was carried out (Table S3). Undefined data with a parameter value of > 100 μM were excluded from the analysis.
Statistical analysis of the parameters of the antitumor activity of the tested compounds shows that the 5-sulfinylacetamide derivative of 4-sulphonyl-1,3-oxazole 3 l is significantly superior compared to 5-sulfonyl-1,3-oxazoles 4, both in terms of antiproliferative activity (GI 50 and TGI) and cytotoxicity. Among sulfonylsulfonyl derivatives, 4 c and 4 e showed the same and the highest potency. That is, the simultaneous replacement of methylphenyl group and phenyl at the 2 nd and 4 th positions in oxazole 4 c with fluorophenyl groups, as well as S-sec-butyl replacement Statistical data (Table S2) make it possible to draw the following conclusions regarding the lines of antitumor activity of the tested compounds for individual subpanels.
Leukemia. All compounds showed high antiproliferative activity in terms of the GI 50 and TGI parameters to this subpanel, with the average values lying in the micromolar range, which were not statistically significantly different from each other, except for the less active compounds 4 h and 4 j, especially in TGI parameter. But all tested lines were sensitive to compound 3 l (the melanoma cell line SK-MEL-5 was not tested), which cannot be said about sulfonyl derivatives. Moreover, the latter, unlike 3 l (the leukemia cell line RPMI-8226 was not tested), did not show cytotoxicity against all subpanel cell lines. Therefore, oxazole 3 l is the most active antitumor compound against the cell lines of this subpanel.
The cell lines of the remaining subpanels were also most sensitive to compound 3 l. Its antiproliferative and cytotoxic activity, averaged over each subpanel, was manifested in a [a] Numbers of sensitive cell lines are given in parentheses; parameter values are expressed in μM (M � m). by S-butyl at the 5 th position, did not affect the antitumor activity of the obtained compound (4 e). However, the attachment of a pyrrolidinoxoethyl (4 b and 4 f) or methyl group (4 h and 4 j) to the 5sulfonyl substituent instead of butyl or its isomer markedly reduced the activity of the obtained derivatives. Apparently, the presence of butyl groups is the main factor influencing the activity of 4,5-bissulfonyl derivatives of oxazole, since the functionalization of phenyls at neither the 2 nd nor the 4 th positions (methylation, halogenation or methoxylation) practically had no effect on the antitumor activity of these derivatives. very narrow concentration range (GI 50 : 1.64-1.86 μM, TGI: 3.16-3.81 μM and LC 50 : 5.53-7.27 μM), i. e. in practice, a high antitumor activity was observed, independent of the nature of the carcinogenic cell line. The remaining compounds were significantly inferior to 3 l in all tested parameters of antitumor activity. Moreover, if for some subpanels (colon cancer, melanoma and prostate cancer) some compounds (4 c and 4 e in particular) approached that of 3 l in terms of antiproliferative activity, then only for a limited number of cell lines within each subpanel. However, even in this case, the cytotoxicity of 3 l significantly exceeded that of the other tested compounds. Thus, statistical analysis of antitumor activity parameters showed that among the tested compounds, 3 l showed the highest activity against all cell lines of cancer subpanels without noticeable selectivity. During the evaluation of selectivity, the approach proposed by Acton et al. was used as a criteria for the degree of selectivity. [14] According to these authors, if 3 3 PG(t)/PG(s) � 6, then the compound has moderate selectivity. When PG(t)/PG(s) > 6, then the compound is highly selective, and when PG(t)/ PG(s) < 3, it is non-selective, where PG(t) is the average cancer cell growth parameter of the total panel, and PG(s) is the same for a separate subpanel. In this context, compounds 3 l and 4 c were found to be non-selective against the nine tumor subpanels tested with selectivity ratios (SR) � 1.5 (Tables 1 and S2). Compounds 4 b, 4 f and 4 h showed moderate GI 50 selectivity for the leukemia subpanel with SR selectivity ranging from 3.1 to 4.2. In addition, compound 4 b was moderately selective toward this subpanel and at the LC 50 level (SR = 3.2). Compound 4 e showed moderate selectivity at the cytotoxicity parameter against the breast cancer subpanel (3.4) and compound 4 f showed the same selectivity against the last one at the TGI and LC 50 parameters (5.0). Only compound 4 j showed equally high selectivity against the ovarian and renal cancer subpanels.

COMPARE correlations
We have performed COMPARE analyses using the GI 50 , TGI and LC 50 vectors for the most active compounds in order to investigate the similarity of their potency, efficacy and cytotoxicity patterns with those of known NCI anticancer standard agents present in public available databases (Table S4). Quantitative evaluation of the obtained results was carried out according to the Chaddock scale. [15] Compound 3 I's GI 50 vector showed a high correlation with antitumor alkylating drugs that add alkyl groups to DNA bases, resulting in cross-linking of the DNA strand and thus inhibition of cancer cell growth. The absence of these in the tested compounds does not allow us to consider them as alkylating agents. However, the high correlation (r = 0.88) of the antiproliferative activity of compound 3 l with the intercalating agent Aclarubicin, which inhibits topoisomerases, suggests that this mechanism is the main one, and served as the basis for a docking analysis of this potential anticancer target. As for cytotoxicity, the question remains open, since among the standard preparations there was none that showed a high correlation with compound 3 l. This result suggests that compound 3 l may have a unique mechanism of cytotoxic action requiring appropriate experimental investigation. A clear moderate correlation of antiproliferative activity with Rifamycin SV was observed for compounds containing a sulfonylsulfonyl group. Their antiproliferative effect may be mediated to some extent through inhibition of DNA-dependent RNA synthesis. Similarly, it can be assumed that the cytotoxicity of these compounds may be partially realized through molecular mechanisms inherent in Geldanamycin or Caracemide. Geldanamycin induces apoptosis through targets sensitive to the heat shock protein Hsp90 [16] causing proteasomal degradation of oncogenic signaling proteins that have mutated in tumor cells exhibiting increased HSP90 activity. [17] Caracemide inhibits Ribonuclease reductase, which plays an important role in cell proliferation by inhibiting DNA synthesis and repair, and also causes cell cycle arrest and apoptosis. [18] It is therefore possible that the likely molecular targets for compounds containing a sulfonylsulfonyl group may be, at least in part, the apoptotic pathways controlled by these proteins. However, a NCI standard drug database search using the COMPARE algorithm did not find among them drugs with antitumor activity highly similar to the average plot of seed compounds containing a sulfonylsulfonyl group.

Molecular docking study
Since the LC 50 vector of compound 3 l showed a high (close to very high correlation, r = 0.88) with that of aclarubicin interacting with topoisomerase, the latter was chosen to determine the possibility of interaction of compound 3 l with this molecular target by in silico method. Table 2 presents the predicted binding energy of the complexation of ligand 3 l into the active centers of three potential molecular anticancer targets.

Docking analysis of potential anticancer targets
The resulting ligand-protein complexes demonstrated the predicted binding energy from À 8.7 to À 10.0 kcal/mol ( Table 2). The most energetically favorable complexation was marked for compound 3 l in the active site of DNA topoisomerase IIβ with a binding energy of À 10.0 kcal/mol. This particularly suggests a potential relationship between the level of in vitro anticancer activity of compound 3 l and the level of binding energy of this compound in the DNA topoisomerase IIβ active center.

Molecular docking ligand 3 l into the active site of DNA topoisomerase IIβ
As a docking center, the co-crystallized inhibitor of DNA topoisomerase IIβ-amsacrine [19] was utilized. For the comparative analysis of the potential of the complexation mechanism, molecular docking of the ligand 3 l into the DNA topoisomerase IIβ active site was performed ( Figure S44, S45). Docking results show the formation of the proteinligand-DNA complex with a predicted binding energy of À 10.0 kcal/mol.
Thus, complexation of compound 3 l occurs in the active site of human DNA topoisomerase IIβ, similar to the amsacrine inhibitor co-crystallized with this enzyme (PDB ID: 4G0 U). The received high values of binding energies and similar docking positions of compound 3 l and the amsacrine inhibitor confirm the correct docking procedure.
Features of the complexation of ligand 3 l into the DNA topoisomerase IIβ active site are shown in Figure S45.
The docking results of compound 3 l ( Figure S44) demonstrate the formation of the ligand-protein-DNA complex with the predicted binding energy of À 10.0 kcal/mol. The formed complex was stabilized by two hydrogen bonds (2.51-2.88 Å) between the acetamide group and amino acids Gln778 and Arg503, one electrostatic interaction (4.80 Å) between the phenyl ring and amino acid Glu522, and four hydrogen bonds (2.72-3.08 Å) between sulfoxide and sulfonyl groups with nucleotide bases DC8, DA12, and DG13. Also, eight hydrophobic interactions (3.73-5.57 Å) were formed between oxazole and furane rings with nucleotide bases DA12 and DG13. It should be noted, the oxazole and furane rings and sulfinyl and sulfonyl groups of the 3 l ligand interact with nucleotide bases. And the acetamide group and the phenyl ring of the ligand interact with the amino acid residues.

Conclusion
A new series of 5-sulfinyl(sulfonyl)-4-arylsulfonyl-substituted 1,3-oxazoles were synthesized in good yields and showed various anticancer activities against NCI 60 panel cell lines. Among them, the potency of seven compounds (3 l, 4 b, 4 c,  4 e, 4 f, 4 h and 4 j) in the micromolar range exceeds that of other synthesized derivatives. Statistical analysis of the obtained data showed the highest anticancer potency of compound 3 l, compared to other tested compounds, which makes it possible to consider it as a lead compound for further in-depth studies and synthesis of new 5-sulfinyl-(sulfonyl)-4-arylsulfonyl-substituted 1,3-oxazoles with antitumor activity. COMPARE analysis showed a high correlation (r = 0.88) of the antiproliferative activity of compound 3 l with the intercalating agent aclarubicin, which inhibits topoisomerases, suggesting that the latter may be the main molecular target in the mechanism of action of this compound. This was the basis for the docking analysis. According to the cytotoxicity vector, no standard agent was found to show a high correlation with compound 3 l, which probably indicates a specific mechanism of its cytotoxic activity that requires further study. Investigation of the anticancer mechanism of action of the oxazole-containing compound 3 l by docking into three likely targets made it possible to identify DNA topoisomerase as the potential target. The complex formation of compound 3 l in the active site of the DNA topoisomerase IIβ was accompanied by a predicted binding energy of À 10.0 kcal/mol. Thus, the ligand-protein-DNA complex was stabilized by the formation of hydrogen bonds, as well as by electrostatic and hydrophobic interactions with amino acids Arg503, Glu522, Gln778 and nucleotide bases DC8, DA12 and DG13. The major groups involved in complexation are oxazole, furane and phenyl rings, sulfoxide and sulfonyl groups.

Experimental Section
Experimental methods, synthetic procedures and compound characteristics are given in the Supporting Information.