Synthesis and Biological Evaluation of Novel Dispiro-Indolinones with Anticancer Activity

Novel variously substituted thiohydantoin-based dispiro-indolinones were prepared using a regio- and diastereoselective synthetic route from 5-arylidene-2-thiohydantoins, isatines, and sarcosine. The obtained molecules were subsequently evaluated in vitro against the cancer cell lines LNCaP, PC3, HCTwt, and HCT(−/−). Several compounds demonstrated a relatively high cytotoxic activity vs. LNCaP cells (IC50 = 1.2–3.5 µM) and a reasonable selectivity index (SI = 3–10). Confocal microscopy revealed that the conjugate of propargyl-substituted dispiro-indolinone with the fluorescent dye Sulfo-Cy5-azide was mainly localized in the cytoplasm of HEK293 cells. P388-inoculated mice and HCT116-xenograft BALB/c nude mice were used to evaluate the anticancer activity of compound 29 in vivo. Particularly, the TGRI value for the P388 model was 93% at the final control timepoint. No mortality was registered among the population up to day 31 of the study. In the HCT116 xenograft model, the compound (170 mg/kg, i.p., o.d., 10 days) provided a T/C ratio close to 60% on day 8 after the treatment was completed. The therapeutic index—estimated as LD50/ED50—for compound 29 in mice was ≥2.5. Molecular docking studies were carried out to predict the possible binding modes of the examined molecules towards MDM2 as the feasible biological target. However, such a mechanism was not confirmed by Western blot data and, apparently, the synthesized compounds have a different mechanism of cytotoxic action.


Introduction
The transcription factor p53-an important tumor suppressor protein-is deeply implicated in a variety of vital intracellular events, especially in apoptosis, DNA repair, and cell-cycle regulation [1,2]. It plays a crucial role in the suppression of tumor growth and progression [1][2][3]. Although rodents with knockout of the TP53 gene live normally, they are prone to the development of various types of cancer [4]. It has been established that, in more than half of the available model tumor cells, this protein is frequently mutated or silenced/deleted. This results in diminished apoptotic capacity and, as a consequence, provokes rapid tumor evolution [5]. This protein has been considered as an attractive biological target due to its unique ability to effectively eradicate many tumors. The functions of patients diagnosed with relapsed or refractory acute myeloid leukemia in combination with cytarabine (a DNA polymerase inhibitor) [51]. The drug was also investigated in early clinical trials (phase I/II studies) for the treatment of relapsed multiple myeloma in combination with ixazomib citrate (a proteasome inhibitor) and dexamethasone (a synthetic adrenocortical steroid), as well as against solid tumors, non-Hodgkin lymphoma (phase I studies), essential thrombocythemia, and polycythemia vera. Among MDM2 inhibitors in clinics, the chemical structures of compounds I-VII ( Figure 1) have been disclosed; these examples clearly demonstrate the paramount clinical potential of MDM2 inhibitors as promising anticancer therapeutics. Spiro-compounds have always been among the most attractive and privileged organic scaffolds in modern medicinal chemistry, due to their wide range of prominent The "nutlin story" has led to the development of RG-7112 [21] and RO-5503781 [45]two compounds produced by Roche, containing this scaffold. RG-7112 was undergoing phase I clinical trials against advanced solid tumors [46] and leukemia [47]. Under MTT tests, this compound demonstrated a relatively high cytotoxicity towards different cancer cell lines, including LNCaP and HCT116, with CC 50 values of 0.18 µM [48] and 0.54 ± 0.12µM [49,50], respectively. However, no recent developments have been reported on this research.
Idasanutlin (RO-5503781) was evaluated in a phase III clinical trial for the treatment of patients diagnosed with relapsed or refractory acute myeloid leukemia in combination with cytarabine (a DNA polymerase inhibitor) [51]. The drug was also investigated in early clinical trials (phase I/II studies) for the treatment of relapsed multiple myeloma in combination with ixazomib citrate (a proteasome inhibitor) and dexamethasone (a synthetic adrenocortical steroid), as well as against solid tumors, non-Hodgkin lymphoma (phase I studies), essential thrombocythemia, and polycythemia vera. Among MDM2 inhibitors in clinics, the chemical structures of compounds I-VII ( Figure 1) have been disclosed; these examples clearly demonstrate the paramount clinical potential of MDM2 inhibitors as promising anticancer therapeutics.
Spiro-compounds have always been among the most attractive and privileged organic scaffolds in modern medicinal chemistry, due to their wide range of prominent physiological activities [52]. In particular, spiro-oxindoles and spiro-hydantoins or isos- teric spiro-thiohydantoins have emerged as attractive synthetic templates because of their prevalence in a considerable number of natural-like products [53]. Alkaloids containing spiro-oxindole motifs were first isolated from plants of the Apocynaceae and Rubiacae families [54]. The basic structural feature of this type of compound is the spiro point at position 3 of the oxindole fragment. This joint can be formed by the attachment of heterocyclic motifs, thereby providing a considerable degree of diversity. As a result, spiro-oxindoles are reasonably regarded as appropriate templates for drug design and development. They can also be readily used as convenient starting points or intermediates in the synthesis of a wide range of structurally diverse natural-like products. A thorough search of the close structural analogues of the synthesized compounds (vide infra) revealed many spiro-indolinones with anticancer activity (compounds VIII-XV, Figure 2). Some compounds were reported as highly active MDM2-p53 PPI inhibitors.
Molecules 2023, 28,1325 4 of 26 physiological activities [52]. In particular, spiro-oxindoles and spiro-hydantoins or isosteric spiro-thiohydantoins have emerged as attractive synthetic templates because of their prevalence in a considerable number of natural-like products [53]. Alkaloids containing spiro-oxindole motifs were first isolated from plants of the Apocynaceae and Rubiacae families [54]. The basic structural feature of this type of compound is the spiro point at position 3 of the oxindole fragment. This joint can be formed by the attachment of heterocyclic motifs, thereby providing a considerable degree of diversity. As a result, spiro-oxindoles are reasonably regarded as appropriate templates for drug design and development. They can also be readily used as convenient starting points or intermediates in the synthesis of a wide range of structurally diverse natural-like products. A thorough search of the close structural analogues of the synthesized compounds (vide infra) revealed many spiro-indolinones with anticancer activity (compounds VIII-XV, Figure 2). Some compounds were reported as highly active MDM2-p53 PPI inhibitors. Using an in silico structure-based de novo design approach, Wang and colleagues at the University of Michigan discovered the novel class of MDM2-p53 PPI inhibitors based on a spiro-oxindole scaffold (MI series) [55]. The compounds developed by Wang's group possessed good PK properties as well as high binding affinity towards MDM2. Several of them, e.g., MI-43 and ISA-27, were evaluated in preclinical trials [56]. However, MI-43 was found to lack the desirable PK profile and was unsuitable for in vivo evaluation against models of human cancer.
Subsequent chemical optimization of MI-43 led to MI-219 (Ki = 5 nM) as a novel analogue with improved in vivo bioavailability (at 50 mg/kg, oral bioavailability was 55-65% in mice), stability, and key PK features. MI-219 was also more effective than iri- Using an in silico structure-based de novo design approach, Wang and colleagues at the University of Michigan discovered the novel class of MDM2-p53 PPI inhibitors based on a spiro-oxindole scaffold (MI series) [55]. The compounds developed by Wang's group possessed good PK properties as well as high binding affinity towards MDM2. Several of them, e.g., MI-43 and ISA-27, were evaluated in preclinical trials [56]. However, MI-43 was found to lack the desirable PK profile and was unsuitable for in vivo evaluation against models of human cancer.
Subsequent chemical optimization of MI-43 led to MI-219 (K i = 5 nM) as a novel analogue with improved in vivo bioavailability (at 50 mg/kg, oral bioavailability was 55-65% in mice), stability, and key PK features. MI-219 was also more effective than irinotecan at its maximum tolerated dose (MTD). Ascenta Therapeutics has purchased the license on MDM2 inhibitors of the MI series proposed by the groups from University of Michigan, and Sanofi-Aventis and Ascenta Therapeutics have signed an exclusive global collaboration and licensing agreement on a number of molecules that inhibit this PPI.
Sanofi suspended the clinical development program for MI-219 and focused on the more potent analogue MI-773. The compound was evaluated in a phase I clinical trial against cancer. MI-888 is another potent inhibitor of p53-MDM2 PPI (K i = 0.44 nM), which was also investigated in preclinical studies in Sanofi. The compound demonstrated a relatively high anticancer activity (IC 50~9 0 nM against SJSA1 human osteosarcoma cells), achieving rapid, complete, and durable tumor regression in two types of xenograft models of human cancer upon oral administration [57]. However, in 2015, Sanofi discontinued the development of MI-773 and MI-888 due to financial reasons.
Compound DS-3032b (IC 50 = 0.183 µM against SJSA1 cells), by Daiichi Sankyo, is under active development in phase I clinical trials for the treatment of advanced solid tumors, lymphoma, leukemia, and multiple myeloma. The number of scientific publications and patents describing this scaffold has continued to increase over the past decade, especially in the field of anticancer drug therapy.
its maximum tolerated dose (MTD). Ascenta Therapeutics has purchased the license MDM2 inhibitors of the MI series proposed by the groups from University of Michiga and Sanofi-Aventis and Ascenta Therapeutics have signed an exclusive global collabor tion and licensing agreement on a number of molecules that inhibit this PPI.
Sanofi suspended the clinical development program for MI-219 and focused on t more potent analogue MI-773. The compound was evaluated in a phase I clinical tr against cancer. MI-888 is another potent inhibitor of p53-MDM2 PPI (Ki = 0.44 nM), whi was also investigated in preclinical studies in Sanofi. The compound demonstrated a r atively high anticancer activity (IC50 ~90 nM against SJSA1 human osteosarcoma cell achieving rapid, complete, and durable tumor regression in two types of xenograft mode of human cancer upon oral administration [57]. However, in 2015, Sanofi discontinued t development of MI-773 and MI-888 due to financial reasons. Compound DS-3032b (IC50 = 0.183 µM against SJSA1 cells), by Daiichi Sankyo, is u der active development in phase I clinical trials for the treatment of advanced solid t mors, lymphoma, leukemia, and multiple myeloma. The number of scientific publicatio and patents describing this scaffold has continued to increase over the past decade, esp cially in the field of anticancer drug therapy.
Previously, we reported a small series of dispiro-indolinone derivatives as possible inhibitors of p53-MDM2 PPI with anticancer activity in vitro [31]. Here, we describe a follow-up study on the synthesis and biological evaluation of a broader series of molecules containing the title scaffold, including in vivo trials.

Synthesis
There are a relatively small number of synthetic routes to dispiro-indolinones that have been reported to date. Ouyang and co-workers [70] reported a convenient synthetic pathway to substituted dispiro-oxindole derivatives (yields: 82-90%) via the three-component 1,3-dipolar cycloaddition of azomethine ylides with the dipolarophile 5-benzylideneimidazolidine-2,4dione. An intermediate was generated in the reaction mixture by the decarboxylative condensation of isatin and an α-amino acid. A versatile synthesis of dispiro-oxindole-fused heterocycles (yields 78-89%) using a three-component reaction on the basis of domino 1,4-dipolar addition and the Diels-Alder reaction was also reported [71].
In accordance with the presented synthetic pathway, a series of compoun was readily obtained using a two-step one-pot format. The following synthetic pr turned out to be optimal (Scheme 1): the starting 5-arylmethylene-2-thiohydan equiv.) was synthesized as described in [31], sarcosine 2 (2 equiv.) and ethan loaded into the reaction flask, the mixture was brought with stirring to a boil, a isatin 3 (2 equiv.) was added. The resulting mixture was refluxed for 5-8 h un mogeneous solution was obtained and the initial 2-thiohydantoin was undetect cording to TLC data).
The desired diastereomerically pure dispiro-indolinones 4-69 (Scheme 1) w lated with good yields (up to 89%, Table 1). The obtained compounds were chara by NMR and HRMS spectra (see Supplementary Materials). The supposed mech the performed reaction has been reported previously [73]. It includes the succe actions of isatin with sarcosine that lead to iminium intermediates through the fo of lactone, and subsequent CO2 elimination resulted in azomethine ylide. This 1, then regio-and diastereoselectively attacks the C=C double bo 5-arylidenehydantoin. The formation of diastereomerically pure produc (S*,S*,R*)-configuration was proven on the basis of X-ray crystallographic anal formed for the dispiro-indolinones 29 and 69 (Figure 4), as well as NMR spectral the reaction mixtures. According to the NMR spectra of the reaction mixtures, o stereomers of compounds 4-69 were not formed under the reaction conditions, e those shown in Scheme 1.   CCH 36 In accordance with the presented synthetic pathway, a series of compounds 4-74 was readily obtained using a two-step one-pot format. The following synthetic procedure turned out to be optimal (Scheme 1): the starting 5-arylmethylene-2-thiohydantoin 1 (1 equiv.) was synthesized as described in [31], sarcosine 2 (2 equiv.) and ethanol were loaded into the reaction flask, the mixture was brought with stirring to a boil, and then isatin 3 (2 equiv.) was added. The resulting mixture was refluxed for 5-8 h until a homogeneous solution was obtained and the initial 2-thiohydantoin was undetectable (according to TLC data).
The desired diastereomerically pure dispiro-indolinones 4-69 (Scheme 1) were isolated with good yields (up to 89%, Table 1). The obtained compounds were characterized by NMR and HRMS spectra (see Supplementary Materials). The supposed mechanism of the performed reaction has been reported previously [73]. It includes the successive reactions of isatin with sarcosine that lead to iminium intermediates through the formation of lactone, and subsequent CO 2 elimination resulted in azomethine ylide. This 1,3-dipole then regio-and diastereoselectively attacks the C=C double bond of 5-arylidenehydantoin. The formation of diastereomerically pure products with (S*,S*,R*)-configuration was proven on the basis of X-ray crystallographic analysis performed for the dispiro-indolinones 29 and 69 ( Figure 4), as well as NMR spectral data for the reaction mixtures. According to the NMR spectra of the reaction mixtures, other diastereomers of compounds 4-69 were not formed under the reaction conditions, except for those shown in Scheme 1.
A typical 1 H NMR spectrum of the resulting dispiro-indolinone is shown in Figure S1 in the Supplementary Materials, using the example of compound 27. The characteristic signals are the singlets of the amide protons of the indolinone and imidazolone cycles in the region of~10-11 ppm, the singlet of N-CH 3 group protons at 2.1-2.2 ppm, and the signals of three protons at sp 3 -hybrid atoms of the central pyrrolidine ring in the region of 3.3-4.4 ppm, which are pseudotriplets with J at about 9.0 Hz. In some cases, the structure and stereochemistry of the obtained dispiro-compounds was further confirmed using 2D NMR spectroscopy data (see Figures S2-S4).

Molecular Docking Study
The first high-resolution co-crystal structure of MDM2 in complex with a fragment of p53 peptide responsible for binding (1YCR) was reported by Kussie and colleagues [74]. The PPI interface observed between these proteins is formed by the first ∼120 N-terminal amino acid (AA) residues of MDM2 and the first 30 N-terminal AAs of p53 ( Figure 5) [75]. Since then, more than 50 crystallographic complexes have been published for small-molecule MDM2-p53 inhibitors with high diversity in structure, including spiro-indolinones. The analysis of the MDM2-p53 binding mode indicated that the p53 peptide adopts an αhelical conformation and interacts with MDM2 primarily via the hydrophobic triad of Phe19, Trp23, and Leu26. This anchoring "trident" of i, i + 4, and i + 7 PPI hotspots binds a medium-sized pocket in the structure of MDM2. This site has been used to design a number of different small-molecule MDM2 inhibitors that effectively prevent the MDM2-p53 interaction, thereby blocking tumor growth and progression. To elucidate a possible mechanism of action of the synthesized compounds against the MDM2 protein, we carried out molecular docking studies using MOE software [12]. The model was constructed based on three crystallographic complexes: a p53 peptide bound to MDM2 [76], as well as compounds XX and XXI [48,77,78]. Figure 5b depicts a 2.0 Å resolution structure of the stapled p53 peptide bound to the target protein. The i, i + 7 peptide is conformationally and proteolytically stabilized with all-hydrocarbon staples and remains quite capable of disrupting the described PPI via a tight binding with the conservative MDM2 pocket.
CH3CHPh-4-Cl-C6H4-5-Cl 62 67 Allyl-2-Py H 12 68 Allyl- A typical 1 H NMR spectrum of the resulting dispiro-indolinone is shown in Figure S1 in the Supplementary Materials, using the example of compound 27. The characteristic signals are the singlets of the amide protons of the indolinone and imidazolone cycles in the region of ~10-11 ppm, the singlet of N-CH3 group protons at 2.1-2.2 ppm, and the signals of three protons at sp 3 -hybrid atoms of the central pyrrolidine ring in the region of 3.3-4.4 ppm, which are pseudotriplets with J at about 9.0 Hz. In some cases, the structure and stereochemistry of the obtained dispiro-compounds was further confirmed using 2D NMR spectroscopy data (see Figures S2-S4).

Molecular Docking Study
The first high-resolution co-crystal structure of MDM2 in complex with a fragment of p53 peptide responsible for binding (1YCR) was reported by Kussie and colleagues [74]. The PPI interface observed between these proteins is formed by the first ∼120 N-terminal amino acid (AA) residues of MDM2 and the first 30 N-terminal AAs of p53 ( Figure 5) [75]. Since then, more than 50 crystallographic complexes have been published for small-molecule MDM2-p53 inhibitors with high diversity in structure, including spiro-indolinones. The analysis of the MDM2-p53 binding mode indicated that the p53 peptide adopts an α-helical conformation and interacts with MDM2 primarily via the hydrophobic triad of Phe19, Trp23, and Leu26. This anchoring "trident" of i, i + 4, and i + 7 PPI hotspots binds a medium-sized pocket in the structure of MDM2. This site has been used to design a number of different small-molecule MDM2 inhibitors that effectively prevent the MDM2-p53 interaction, thereby blocking tumor growth and progression. To elucidate a possible mechanism of action of the synthesized compounds against the MDM2 protein, we carried out molecular docking studies using MOE software [12]. The model was constructed based on three crystallographic complexes: a p53 peptide bound to MDM2 [76], as well as compounds XX and XXI [48,77,78]. Figure 5b depicts a 2.0 Å resolution structure of the stapled p53 peptide bound to the target protein. The i, i + 7 peptide is conformationally and proteolytically stabilized with all-hydrocarbon staples and remains quite capable of disrupting the described PPI via a tight binding with the conservative MDM2 pocket. The binding site was reconstructed and compared with the binding mode disclosed for the recombinant p53 peptide [79]. Water molecules were excluded from the modeling procedure. A three-centered 3D pharmacophore model containing three key hydrophobic binding points (see above) was then constructed and used during the docking procedure as a rigid frame to identify and maintain the correct position of the compounds in accordance with the X-ray data. This hybrid model restricts the pool of hypothetical "active" conformations to the most appropriate and reliable ones. The reference compounds XX and XXI were then docked into the developed model, keeping the reported stereo configuration. The obtained results were well correlated with the X-ray structures published previously. For instance, the predicted conformation of compound XX was very similar to that revealed by crystallographic analysis, with an r.m.s.d. value of 1.05 Å. As shown in Figure 5b, tert-butyl and o,m-dihalophenyl moieties of compound XX fit into the Phe19 and Leu26 pockets, respectively. The latter interaction is maintained by a π−π stacking arrangement with His96; however, if this residue is protonated upon a pH value equal or below pKa, π-cation interaction is realized instead. It should be noted that several piperidinone-containing MDM2 inhibitors (for instance, compound XXI) interact with the Phe19 pocket through one chlorophenyl fragment, while the other phenyl moiety occupies the Leu26 sub-pocket, in contrast to compound XX. Thus, there are striking differences in binding mode observed for these two inhibitors [76]. On the other hand, compound XXI, despite bearing the same core, binds the MDM2 pocket in a similar manner to indolinone XX (Figure 5c).  The binding site was reconstructed and compared with the binding mode disclosed for the recombinant p53 peptide [79]. Water molecules were excluded from the modeling procedure. A three-centered 3D pharmacophore model containing three key hydrophobic binding points (see above) was then constructed and used during the docking procedure as a rigid frame to identify and maintain the correct position of the compounds in accordance with the X-ray data. This hybrid model restricts the pool of hypothetical "active" conformations to the most appropriate and reliable ones. The reference compounds XX and XXI were then docked into the developed model, keeping the reported stereo configuration. The obtained results were well correlated with the X-ray structures published previously. For instance, the predicted conformation of compound XX was very similar to that revealed by crystallographic analysis, with an r.m.s.d. value of 1.05 Å. As shown in Figure 5b, tert-butyl and o,m-dihalophenyl moieties of compound XX fit into the Phe19 and Leu26 pockets, respectively. The latter interaction is maintained by a π−π stacking arrangement with His96; however, if this residue is protonated upon a pH value equal or below pKa, π-cation interaction is realized instead. It should be noted that several piperidinone-containing MDM2 inhibitors (for instance, compound XXI) interact with the Phe19 pocket through one chlorophenyl fragment, while the other phenyl moiety occupies the Leu26 sub-pocket, in contrast to compound XX. Thus, there are striking differences in binding mode observed for these two inhibitors [76]. On the other hand, compound XXI, despite bearing the same core, binds the MDM2 pocket in a similar manner to indolinone XX (Figure 5c).
We used the crystal structures of compound 29 and compound 74 solved from singlecrystal X-ray diffraction (see Figure 4) with predefined geometry during the follow-up docking study. As shown in Figure 6c, compound 29 favorably fitted the pharmacophore hypothesis and was positioned within the binding site in a manner similar to that revealed for compound XX. The predicted binding conformation was similar to that published for the MDM2 inhibitors of the MI series. However, in contrast to the template molecule, compound 29 showed an almost twofold lower score value that, in turn, resulted in weaker p53 activation activity (see below). As in the case of morpholinone-based MDM2 inhibitors, we also obtained several alternative conformations with similar binding energy, where the methyl group of compound 29 is located in a deep cavity, similar to the reported 5-halogensubstituted indolinones. As expected, N-substituted indolinone 74 did not show anticancer activity and did not pass in silico evaluation, providing a relatively weak score value (E score over −2 kcal/mol). p53 activation activity (see below). As in the case of morpholinone-based MDM2 inhibitors, we also obtained several alternative conformations with similar binding energy, where the methyl group of compound 29 is located in a deep cavity, similar to the reported 5-halogen-substituted indolinones. As expected, N-substituted indolinone 74 did not show anticancer activity and did not pass in silico evaluation, providing a relatively weak score value (E score over −2 kcal/mol).

Cytotoxicity
The synthesized compounds were tested for their ability to block the proliferation of the selected model cell lines: LNCaP (p53+), PC3 (p53-), HCT wt (p53+), and HCT −/− (p53-) cells. The assay was carried out using the standard MTT (3-(4,5-dimethylthiazol-2-yl)2,5diphenyl tetrazolium bromide) assay based on the modified approach reported by Ferrari and colleagues [79]. LNCaP and HCT wt cells expressed p53 (positive cells), while PC3 and HCT −/− did not express p53 (negative). Differences in IC50 values against these cell lines were used to assess the selectivity of the compounds. The cytotoxic activity of the evaluated compounds is summarized in Table 2.

Cytotoxicity
The synthesized compounds were tested for their ability to block the proliferation of the selected model cell lines: LNCaP (p53+), PC3 (p53-), HCT wt (p53+), and HCT −/− (p53-) cells. The assay was carried out using the standard MTT (3-(4,5-dimethylthiazol-2yl)2,5-diphenyl tetrazolium bromide) assay based on the modified approach reported by Ferrari and colleagues [79]. LNCaP and HCT wt cells expressed p53 (positive cells), while PC3 and HCT −/− did not express p53 (negative). Differences in IC 50 values against these cell lines were used to assess the selectivity of the compounds. The cytotoxic activity of the evaluated compounds is summarized in Table 2.  17.0 ± 1.0 20.0 ± 6.9 >100 >100 2.7 ± 0.9 28.9 ± 3.7 4.5 ± 1.3 >100 As shown in Table 2, the most active compounds from the series (compounds 13, 18, 29, and 63) exhibited IC 50 values of~1-3.5 µM against LNCaP cells. These molecules contain a halogen atom in position 5 of the isatin fragment (Br for compounds 13, 29, and 63, or Cl for compound 18). Some molecules demonstrated a good selectivity index for LNCaP-PC3 and/or HCT wt -HCT −/− pairs (SI = 3 to 10). Compound 13 showed an SI value similar to that observed for nutlin-3a, with a proven mechanism of cytotoxic action as an inhibitor of the p53-MDM2 interaction.

Intracellular Accumulation
To investigate the intracellular localization, we synthesized propargyl-based compounds 69 and 35, the latter of which is the propargylated analogue of one of the most cytotoxic dispiro-indolinone 29 (Scheme 2), equipped with the fluorescent dye (Sulfo-Cy5azide) via a click reaction inside the cell (no intrinsic fluorescence was observed for the obtained conjugate). Human embryonic kidney cells (HEK293) or prostate cancer cells (22RV1) were treated with N-propargyl derivatives 69 or 70 and then stained with Sulfo-Cy5-azide in the presence of CuSO 4 and sodium ascorbate.
Based on the obtained results, we can conclude that compound 69 mainly localizes in the cytoplasm. Compound 35 similarly does not have a significant distribution pattern in the cell, and it is spread over the cell, but it is somewhat more observed in the cytoplasm than in the nucleus and demonstrates increased cell membrane staining, which is consistent with high hydrophobicity. An example of the fluorescence of the labeled compound (conjugate) in microphotographs obtained by confocal microscopy is shown in Figure 7. Based on the obtained results, we can conclude that compound 69 mainly localizes in the cytoplasm. Compound 35 similarly does not have a significant distribution pattern in the cell, and it is spread over the cell, but it is somewhat more observed in the cytoplasm than in the nucleus and demonstrates increased cell membrane staining, which is consistent with high hydrophobicity. An example of the fluorescence of the labeled compound (conjugate) in microphotographs obtained by confocal microscopy is shown in Figure 7.  Based on the obtained results, we can conclude that compound 69 mainly localize in the cytoplasm. Compound 35 similarly does not have a significant distribution pattern in the cell, and it is spread over the cell, but it is somewhat more observed in the cyto plasm than in the nucleus and demonstrates increased cell membrane staining, which i consistent with high hydrophobicity. An example of the fluorescence of the labeled compound (conjugate) in microphotographs obtained by confocal microscopy is shown in Figure 7.

Western Blot, Mechanism of Cell Death, and 2D Electrophoresis
To verify the mechanism of cytotoxic action of the synthesized dispiro-indolinones, Western blot testing of dispiro-indolinone 29 was performed using the p53-expressing HCTwt and LNCaP cell lines. Both the p53 protein and the proteins involved in the p53dependent process of apoptosis (i.e., BAX, NOXA, and PUMA) were determined. For comparison, similar testing was carried out for a cell line not expressing p53 (HCT (−/−) ).
According to the obtained data ( Figure 8), with an increase in the concentration of compound 29, a slight increase in the concentration of the p53 protein in the HCTwt cell culture was observed, which may include inhibition of the p53-MDM2 interaction by dispiro-indolinone. An increase in the expression of the pro-apoptotic proteins PUMA and CASPASE-3 was also detected. However, the treatment with nutlin-3a, for which the mechanism of action through inhibition of the p53-MDM2 interaction was confirmed, demonstrated a stronger increase in the expression of the p53 protein. Thus, p53 activation is not the main mechanism of 29's cytotoxic action. To verify the mechanism of cytotoxic action of the synthesized dispiro-indolinones, Western blot testing of dispiro-indolinone 29 was performed using the p53-expressing HCTwt and LNCaP cell lines. Both the p53 protein and the proteins involved in the p53-dependent process of apoptosis (i.e., BAX, NOXA, and PUMA) were determined. For comparison, similar testing was carried out for a cell line not expressing p53 (HCT (−/−) ).
According to the obtained data ( Figure 8), with an increase in the concentration of compound 29, a slight increase in the concentration of the p53 protein in the HCTwt cell culture was observed, which may include inhibition of the p53-MDM2 interaction by dispiro-indolinone. An increase in the expression of the pro-apoptotic proteins PUMA and CASPASE-3 was also detected. However, the treatment with nutlin-3a, for which the mechanism of action through inhibition of the p53-MDM2 interaction was confirmed, demonstrated a stronger increase in the expression of the p53 protein. Thus, p53 activation is not the main mechanism of 29's cytotoxic action. The mechanism of cell death (i.e., necrosis/apoptosis) caused by one of the most cytotoxic compounds (13) was investigated by flow cytometry with staining by 7AAD/annexin V. HCT116p53−/− mainly showed necrosis/late apoptosis, HCT116wt predominantly showed early apoptosis, and PC3 also showed early apoptosis ( Figure  S5).
To further study the action of 29, we investigated the changes in the protein landscape with 2D protein gel electrophoresis (Table 1 and Figure S6). As some proteins' expressions patterns differ in HCT116wt and HCT116p53−/− cells after treatment, and 2D protein gel electrophoresis is semi-quantitative, we looked for associations of the pool of proteins whose expression was changed. There are no common edges in the known transcription regulatory networks in accordance with "http://www.licpathway.net/KnockTFv2 (accessed on 10 November 2022)". The pool consisted of proteins, associated with (either) mitochondria and (or) protein folding (in accordance with Gene Ontology by string-db.org). Taken together, we may speculate that The mechanism of cell death (i.e., necrosis/apoptosis) caused by one of the most cytotoxic compounds (13) was investigated by flow cytometry with staining by 7AAD/annexin V. HCT116p53−/− mainly showed necrosis/late apoptosis, HCT116wt predominantly showed early apoptosis, and PC3 also showed early apoptosis ( Figure S5).
To further study the action of 29, we investigated the changes in the protein landscape with 2D protein gel electrophoresis (Table 1 and Figure S6). As some proteins' expressions patterns differ in HCT116wt and HCT116p53−/− cells after treatment, and 2D protein gel electrophoresis is semi-quantitative, we looked for associations of the pool of proteins whose expression was changed. There are no common edges in the known transcription regulatory networks in accordance with "http://www.licpathway.net/KnockTFv2 (accessed on 10 November 2022)". The pool consisted of proteins, associated with (either) mitochondria and (or) protein folding (in accordance with Gene Ontology by string-db.org). Taken together, we may speculate that we saw the secondary effects as responses to stress and apoptosis, which masked the initial changes caused by 29.
Taking into account its high cytotoxicity, sufficient selectivity, and ability to activate pro-apoptotic cascades according to the results of Western blotting, compound 29 was chosen for further biological studies in vivo.

Biological Evaluation In Vivo
All animal procedures were conducted in strict adherence to the European Convention for the Protection of Vertebrate Animals Used for Experimental and other Scientific Purposes (ETS 123), Strasbourg, 1986. To preliminary estimate the antitumor effect of the compound in vivo, transplant P388 and HCT116 xenograft mouse models were used. For intraperitoneal (i.p.) administration, compound 29 was dissolved with Pluronic ® F-127. The sample contained the active compound (17%) and the solubilizing agent (83%). The final 4.8% suspension was prepared at pH = 7.0 in sterile distilled water. For peroral (p.o.) administration, we used a 3.4% suspension of the tested compound in 1% starch-based thickener, which was prepared repeatedly ex tempore.

P388 Model
Prior to the trials, the animals were split into different groups (five mice per group). One group was monitored as a control population, where the animals were kept untreated, while the other groups were subjected to the tested compound. The anticancer effect of the evaluated sample was assessed in model mice at the dose range of 295-1176 mg/kg. Four independent trials were carried out.
For the first evaluation, the sample was initially injected i.p. on day 2 of the study, and the treatment was subsequently continued during the following 9 days (total 10 d course). The typical doses were calculated considering the average body weight of the animals (20 g) and ranged between groups 1 and 3; group 4 was assigned as a control population. Thus, the first group (g1) was daily treated with the investigated sample at a dose of 295 mg/kg (cumulative dose was 2950 mg/kg), the second group was given 882 mg/kg (8820 mg/kg), and the third group was given 1176 mg/kg (11,760 mg/kg). To verify the obtained results, we performed the second evaluation using the following schedule: the first group was administered i.p. with 500 mg/kg (5000 mg/kg), the second group was given 750 mg/kg (7500 mg/kg), and the third group was given 1000 mg/kg (10,000 mg/kg). We also carried out the third trial with an appropriate dose of 1000 mg/kg. For p.o. administration (trial 4), we used a single dose (170 mg/kg) of the unmodified compound; the cumulative dose was 1700 mg/kg.

First Trial
Within the control group (g4), it was shown that at three timepoints (days 12, 15, and 19 after inoculation) the cancer nodes of the P388 cells grew rapidly and normally, reaching 383 ± 93 mm 3 , 591 ± 189 mm 3 , and 1304 ± 676 mm 3 in volume, respectively. The death outcome was registered on days 15-19 ( Figure 9; average life span (ALS) = 16.6 ± 1.8 d). Two mice in the group were still alive on day 8 after the treatment period had ended. It was revealed that at doses of 295 mg/kg (g1) and 882 mg/kg (g2), the cancer growth rate was similar to that observed in the control group. The treatment resulted in a relatively good efficiency: TGRI (tumor growth rate inhibition) = 32-37% (p > 0.05) and T/C < 42%. The lethal outcomes were registered on days 16-24 and 18-22, respectively. Several mice stayed alive for more than 20 days; however, the gain in survival was statistically insignificant: ALS = 19.8 ± 3.0 d and 19.4 ± 1.7 d, respectively, and LSE (lifespan extension) = 17% and 19% (p > 0.05), respectively, vs. ALS = 16.6 ± 1.8 d within the control group. The maximum therapeutic effect was achieved with the daily dose of 1176 mg/kg (11,760 mg/kg, g3) 24 h after the treatment was discontinued (TGRI = 76%, T test = 0.0002, p < 0.05). During the subsequent three days, this effect was reduced to TGRI = 40%, providing a relatively high dispersion value. Deaths were observed on days 14-22; on day 8 after the course was stopped, four mice were feeling well and stayed alive for more than 20 days. The administration of the tested sample at the highest dose of 1263 mg/kg (12,630 mg/kg), close to the MTD value, resulted in one mouse (bw = 19 g) dying on day 3 after the treatment was stopped. The obtained results are depicted in Figures 9 and 10. providing a relatively high dispersion value. Deaths were observed on days 14-22; on day 8 after the course was stopped, four mice were feeling well and stayed alive for more than 20 days. The administration of the tested sample at the highest dose of 1263 mg/kg (12,630 mg/kg), close to the MTD value, resulted in one mouse (bw = 19 g) dying on day 3 after the treatment was stopped. The obtained results are depicted in Figures 9 and 10.

Second Trial
Since we did not obtain statistically significant results during the first trial due to an inappropriate dispersion at a relatively high dose (although the TGRI value reached 76% (p < 0.0002)), we used doses of 500 and 750 mg/kg o.d. for the second trial. At the higher dose, on day 2 after the treatment was completed, we obtained almost identical results reaching TGRI = 68% (p = 0.002) vs. TGRI = 76% (p = 0.0002). Moreover, during the subsequent 3 days, the TGRI rate was kept at a statistically significant level of 62% (T test = 0.02 p < 0.05). Lower doses did not provide any statistically significant results.

Third Trial
The trial was carried out with a dose of 1000 mg/kg o.d. (cumulative dose of 10,000 mg/kg) under the same conditions. On day 2 after the treatment was completed, we obtained superior results (Figure 11), reaching TGRI = 81.3% vs. TGRI = 76%. During the providing a relatively high dispersion value. Deaths were observed on days 14-22; on day 8 after the course was stopped, four mice were feeling well and stayed alive for more than 20 days. The administration of the tested sample at the highest dose of 1263 mg/kg (12,630 mg/kg), close to the MTD value, resulted in one mouse (bw = 19 g) dying on day 3 after the treatment was stopped. The obtained results are depicted in Figures 9 and 10.

Second Trial
Since we did not obtain statistically significant results during the first trial due to an inappropriate dispersion at a relatively high dose (although the TGRI value reached 76% (p < 0.0002)), we used doses of 500 and 750 mg/kg o.d. for the second trial. At the higher dose, on day 2 after the treatment was completed, we obtained almost identical results, reaching TGRI = 68% (p = 0.002) vs. TGRI = 76% (p = 0.0002). Moreover, during the subsequent 3 days, the TGRI rate was kept at a statistically significant level of 62% (T test = 0.02, p < 0.05). Lower doses did not provide any statistically significant results.

Third Trial
The trial was carried out with a dose of 1000 mg/kg o.d. (cumulative dose of 10,000 mg/kg) under the same conditions. On day 2 after the treatment was completed, we obtained superior results (Figure 11), reaching TGRI = 81.3% vs. TGRI = 76%. During the

Second Trial
Since we did not obtain statistically significant results during the first trial due to an inappropriate dispersion at a relatively high dose (although the TGRI value reached 76% (p < 0.0002)), we used doses of 500 and 750 mg/kg o.d. for the second trial. At the higher dose, on day 2 after the treatment was completed, we obtained almost identical results, reaching TGRI = 68% (p = 0.002) vs. TGRI = 76% (p = 0.0002). Moreover, during the subsequent 3 days, the TGRI rate was kept at a statistically significant level of 62% (T test = 0.02, p < 0.05). Lower doses did not provide any statistically significant results.

Third Trial
The trial was carried out with a dose of 1000 mg/kg o.d. (cumulative dose of 10,000 mg/kg) under the same conditions. On day 2 after the treatment was completed, we obtained superior results (Figure 11), reaching TGRI = 81.3% vs. TGRI = 76%. During the following 3 days, the TGRI value was increased up to 89% (T test = 0.002), maintaining a statistically significant level. On day 8 after the treatment was discontinued, the TGRI value reached 93% (T test = 0.001); the ALS index was also very high, at 27.6 ± 0.5 d (LSE = 33.33%). It should be noted that on day 31 of the study no mortality was observed among the treated population. following 3 days, the TGRI value was increased up to 89% (T test = 0.002), maintainin statistically significant level. On day 8 after the treatment was discontinued, the TG value reached 93% (T test = 0.001); the ALS index was also very high, at 27.6 ± 0.5 d (LS 33.33%). It should be noted that on day 31 of the study no mortality was observed amo the treated population.  Table S2. No statistically sign cant reduction in tumor volume was registered, and no toxic effects were observ Hence, we can conclude that at a dose of 170 mg/kg compound 29 is ineffective after p administration.
As a result, the treatment of P388 mice with the tested substance at a dose of 1 mg/kg (10,000 mg/kg, i.p., cumulative dose) led to a statistically relevant and reprodu ble outcome. The best anticancer effect considering the principal screening criteria (TG ≥ 50%) was achieved at the TGRI level of 93%; however, this dose was close to the M value. It should be noted that at a dose of 500 mg/kg the TGRI value was approx. 50 whereas the administration of the sample at a dose of 1200 mg/kg resulted in lethal o comes.

HCT116 Model
Considering the effective dose indicated previously (1000 mg/kg, o.d., 10 day HCT116 xenograft BALB/c nude mice were treated i.p. with the tested sample using same time and dose schedule. Tolerance and lifespan were not assessed because imm nodeficient animals were used. Within the control group, HCT116 tumor xenogra grew normally, reaching 909 ± 396 mm 3 , 1616 ± 486 mm 3 , and 2554 ± 663 mm 3 in aver volume on days 12, 14, and 16, respectively. In the treated group, after the course w completed, statistically significant results were obtained at a T/C (treatment/control) le of 25% (p = 0.00005, after 24 h) vs. the P388 model (T/C = 133%, TGRI = 93%) at the fi control timepoint (i.e., day 19 of the study). This effect was reduced to T/C = 44% ( 0.0007, after 4 days) and T/C = 58% (p = 0.009, after day 8, Figure 12).  Table S2. No statistically significant reduction in tumor volume was registered, and no toxic effects were observed. Hence, we can conclude that at a dose of 170 mg/kg compound 29 is ineffective after p.o. administration.
As a result, the treatment of P388 mice with the tested substance at a dose of 1000 mg/kg (10,000 mg/kg, i.p., cumulative dose) led to a statistically relevant and reproducible outcome. The best anticancer effect considering the principal screening criteria (TGRI ≥ 50%) was achieved at the TGRI level of 93%; however, this dose was close to the MTD value. It should be noted that at a dose of 500 mg/kg the TGRI value was approx. 50%, whereas the administration of the sample at a dose of 1200 mg/kg resulted in lethal outcomes.

HCT116 Model
Considering the effective dose indicated previously (1000 mg/kg, o.d., 10 days), HCT116 xenograft BALB/c nude mice were treated i.p. with the tested sample using the same time and dose schedule. Tolerance and lifespan were not assessed because immunodeficient animals were used. Within the control group, HCT116 tumor xenografts grew normally, reaching 909 ± 396 mm 3 , 1616 ± 486 mm 3 , and 2554 ± 663 mm 3 in average volume on days 12, 14, and 16, respectively. In the treated group, after the course was completed, statistically significant results were obtained at a T/C (treatment/control) level of 25% (p = 0.00005, after 24 h) vs. the P388 model (T/C = 133%, TGRI = 93%) at the final control timepoint (i.e., day 19 of the study). This effect was reduced to T/C = 44% (p = 0.0007, after 4 days) and T/C = 58% (p = 0.009, after day 8, Figure 12).
Based on the obtained results, it can be rationally concluded that compound 29 demonstrates significantly relevant anticancer activity at a daily dose of 1000 mg/kg (i.p., 10 days) in HCT116-bearing xenograft mice.

Acute Toxicity
This study aimed to estimate the toxicity of compound 29 administered intraperitoneally and by intravenous injection. The objectives of the study included the analysis of the clinical picture of intoxication, the analysis of the reversibility of the revealed toxic effects, and the identification of the most sensitive organs. Acute toxicity tests were performed on mature outbred mice (male and female). The design of the experiment is specified in Table S3 (Supplementary Materials) Based on the obtained results, it can be rationally concluded that compoun demonstrates significantly relevant anticancer activity at a daily dose of 1000 mg/kg 10 days) in HCT116-bearing xenograft mice.

Acute Toxicity
This study aimed to estimate the toxicity of compound 29 administered intrap toneally and by intravenous injection. The objectives of the study included the analys the clinical picture of intoxication, the analysis of the reversibility of the revealed t effects, and the identification of the most sensitive organs. Acute toxicity tests were formed on mature outbred mice (male and female). The design of the experime specified in Table S3 (Supplementary Materials).
To select the doses, the pilot experiments were carried out on a minimum numb animals. In a preliminary experiment, it was found that with intraperitoneal admini tion of compound 29 at a dose of 2500 mg/kg, there were no deaths among the anim When administered in large doses during autopsy in the abdominal cavities of the m on the inner wall and membranes of the abdominal organs, remnants of non-absorbed drug were found in the form of white layers, indicating incomplete sorption and insufficient solubility of the agent. Therefore, due to the low solubility o test substance in aqueous media, it turned out to be impossible to obtain reliable ex imental data with the introduction of compound 29 at doses > 2500 mg/kg. Accordin the data obtained from the trial experiment, in the main study-with intraperito administration-a dose of 2500 mg/kg was used.
The observation period was 28 days. With intraperitoneal administration of c pound 29 in mice, no deaths were registered. The surviving animals were subjecte euthanasia and necropsy on the 29th day of the experiment to study the effects of the agent on their internal organs. To select the doses, the pilot experiments were carried out on a minimum number of animals. In a preliminary experiment, it was found that with intraperitoneal administration of compound 29 at a dose of 2500 mg/kg, there were no deaths among the animals. When administered in large doses during autopsy in the abdominal cavities of the mice, on the inner wall and membranes of the abdominal organs, remnants of the non-absorbed drug were found in the form of white layers, indicating incomplete absorption and insufficient solubility of the agent. Therefore, due to the low solubility of the test substance in aqueous media, it turned out to be impossible to obtain reliable experimental data with the introduction of compound 29 at doses > 2500 mg/kg. According to the data obtained from the trial experiment, in the main study-with intraperitoneal administration-a dose of 2500 mg/kg was used.
The observation period was 28 days. With intraperitoneal administration of compound 29 in mice, no deaths were registered. The surviving animals were subjected to euthanasia and necropsy on the 29th day of the experiment to study the effects of the test agent on their internal organs.
The clinical picture of intoxication with intraperitoneal administration of compound 29 was characterized by a depressed state, a decrease in motor activity, and the release of porphyrin. Symptoms of intoxication disappeared after 30-60 min. An assessment of the dynamics of body weight in animals that underwent intoxication did not show differences from the control mice. Autopsy of experimental animals on the 29th day after exposure to the drug revealed an increase in the liver, with signs of dystrophy (Table S4).
Thus, it can be concluded that intraperitoneal administration of dispiro-derivative 29 has a toxic effect on the bodies of mice. At the same time, deaths of the animals were not recorded when a dose of 2500 mg/kg was administered. Thus, the therapeutic index (estimated as LD 50 /ED 50 ) for compound 29 in mice is at least 2.5.

Materials and Methods
Commercially available chemicals were purchased from Sigma-Aldrich. The 1 H NMR spectra of all synthesized compounds were recorded on Bruker Advance 400 spectrometer at 400 MHz in DMSO-d 6 . Chemical shifts (δ) are reported in parts per million (ppm) downfield from TMS and referenced from solvent references. Coupling constants are reported in Hz. The elemental compositions of the compounds determined by HRMS were matched within ± 0.4% of the calculated values. The reactions were followed by thin-layer chromatography (TLC) on Merck aluminum silica gel sheets that were visualized under a UV lamp.
The data of compound 29 were collected by using a STOE diffractometer and a Pi-latus100K detector, focusing mirror collimation Cu Kα (1.54086 Å) radiation, in rotation method mode. STOE X-AREA software was used for cell refinement and data reduction. Data collection and image processing were performed using X-Area 1.67 (STOE and Cie GmbH, Darmstadt, Germany, 2013). Intensity data were scaled with LANA (part of X-Area) in order to minimize differences in the intensities of symmetry-equivalent reflections (multi-scan method).
X-ray investigations of compound 69 were performed using a SMART APEX DUO CCD diffractometer (Mo Kα radiation (0.71073 Å), graphite monochromator, ϕand ω-scan). Both structures were solved by direct methods and refined using full-matrix anisotropic approximation for F 2 hkl . The location of hydrogen atoms (CH) was predicted geometrically, and their positions were well adjusted using the "rider" model U iso (H) = 1.5 U eq (C) for methyl groups and 1.2 U eq (X) for the remaining H atoms. The location of hydrogen atoms (NH) was predicted using difference Fourier maps and was isotropically specified. All calculations were performed using SHELX software version 2009-9.13 [80]. CCDC-2217004 (compound 29) and CCDC-2217008 (compound 69) contain the supplementary crystallographic data for this paper. The data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif website (accessed on 10 January 2023).

General Procedure for the Synthesis of Compounds 4-69
Isatin 3 (0.3 mmol) was added to a refluxing solution of 5-arylidene-2-thiohydantoin 1 (0.3 mmol) and sarcosine 2 (0.6 mmol) in ethanol (10-15 mL), and the resulting mixture was boiled for 6 h, monitoring the completion of the reaction by TLC. After cooling, the reaction mixture was poured into 100 mL of cold water, and the precipitate that formed was filtered off and recrystallized from ethanol. The synthesized compounds were characterized by NMR and HRMS spectra (see Supplementary Materials).

Molecular Docking
The crystal structures of small-molecule compounds in complex with MDM2 protein were obtained from the PDB collection [81]. The binding sites were reconstructed and prepared using MOE Software following the routine procedure. Several complexes were then superposed to analyze the conformational dynamics that occurred upon ligand binding. It was considered that water molecules did not significantly influence the docking results. The constructed model was then verified using some MDM2 inhibitors with different structures. Satisfactory results were obtained, providing predicted 3D conformations close to those disclosed by X-ray analysis. After the refinement procedure was completed, two compounds from our series characterized by RSA were docked into the developed model. To restrict the possible conformational space, we designed a 3-centered 3D pharmacophore model, which was integrated into the binding site. Docking was performed in a dynamic mode, where amino acids that organized the active binding site, as along with amino acids that were located in close proximity to the pocket, were also perturbed during the ligand positioning. Several possible active conformations with low energy scores were outputted from each docking simulation. An appropriate conformation was then assessed by a visual inspection and on the basis of energy score values and expert opinion.

In Vitro Assay
Model cell lines were purchased from the in stock collection of the N.N. Blokhin Russian Cancer Research Center (RAMS) [82]. Cells were accurately maintained in the recommended media, supplemented with 10% heat-inactivated FBS (Invitrogen) and 2 mmol/L L-glutamine (Invitrogen).

MTT test
First, 4000 cells per well were plated out in 140 µL of DMEM (with 10% FBS) in 96-well plates and incubated in 5% CO 2 for 24 h at 37 • C without treatment. Then, 15 µL of water-DMSO solutions (several final solutions were opalescent, one looked like a suspension, and the compounds were poorly soluble) of the tested substances were added to the cells (the DMSO concentrations in the media were 1% or less). The cells were treated for 48 h with 45-100 µM (eight dilutions) concentrations of the substances (each in triplicate) or doxorubicin (0.3 nM-600 nM, eight dilutions) as a control substance. After the incubation, 15 µL of MTT solution in PBS (5 mg/mL) was added, and the cells were incubated for 2 h, followed by removal of the media and addition of 120 µL of DMSO. The samples were then incubated for 15 min with shaking to completely solubilize the formazan. The amount of MTT reduced by cells to its blue formazan derivative was measured spectrophotometrically at 565 nm using a plate reader. Cell survival rates were determined by the comparison of optical density with the corresponding untreated control cell culture. Dose-dependent curves and the corresponding IC 50 values were calculated using "GraphPad Prism 6" software (GraphPad Software, Inc., San Diego, CA, USA).

Intracellular Accumulation
The studies were carried out according to the protocol described in [83].

In Vivo Cytotoxicity Assay Animals
The P388 cell line was subcutaneously (s.c.) injected in BDF1 hybrid mice (21-26 g females, 29-34 g males) obtained from the nursery of the Russian Academy of Medical Sciences "Stolbovaya" [84]. DBA2 mice were used as donors of cancerous mass; a total of 50 animals were recruited. Twenty-three conventional, immunodeficient female mice (BALB/c nude, 20-23 g) were housed in RAMS and used as donors and recipients for HCT116 cells.
Compound Formulation (a) For i.p. administration, the tested compound 29 was dissolved with Pluronic ® F-127. The sample contained the compound (17%) and the solubilizing agent (83%). The final 4.8% suspension was prepared at pH = 7.0 in sterile distilled water. (b) For p.o. (per os) administration, we used a 3.4% suspension of the tested compound in 1% starch-based thickener, which was prepared repeatedly ex tempore.
For the evaluation of the compound against leucosis in vivo, we used 5 and 8 passages of P388 culture. Inoculum containing 10 6 . of P388 cells immersed in 0.3 mL of nutrient medium 199 was introduced to the animals subcutaneously in the flank zone.
The first implantation of HCT116 tumor cells (1 × 10 6 in a volume of 0.5 mL) was performed subcutaneously into two BALB/c nude mice. The first transplantation of the xenograft (50 mg) was performed after 30 days into 2 mice (the 3rd passage was used in vivo). An ampoule containing the tumor suspension (1 mL) obtained from a cryobank was defrosted ex tempore in a thermostat for 10 min at 37 • C. Then, in a laminar flow hood, the tumor suspension (0.5 mL, the first passage) was implanted subcutaneously into the flank area of two BALB/c nude mice (males). After a volume of 0.8-1.0 cm 3 was achieved, the nudes were passaged using 4 donor mice to obtain the desired grade and mass of the transplants. The donor mice were euthanatized in a CO 2 chamber, in full accordance with the guidance of the RSPCA and LASA (2015). Tumor nests were accurately isolated, placed in plastic Petri dishes, washed with nutrient medium 199, extracted from capsules, and weighed using an electronic balance. The tumor nests were mechanistically disintegrated to obtain 10% of the tumor samples, and then the nutrient medium was added. The inoculation was performed using individual sterile plastic syringes (1 mL) equipped with 18G suture needles. The transplantation was carried out bilaterally in the flank areas. Each mouse was inoculated with 0.5 mL (10%) of the tumor suspension (40 mg of the tumor mass for each animal). Prior to inoculation, the animals' skin was treated with 70% alcohol. Between injections, the Petri dish containing the tumor suspension was tightly covered to avoid any contamination. Before injection, the suspension was vigorously stirred in a syringe. After inoculation, the needle was accurately removed to avoid the tumor escaping. The mice were kept under special conditions in an air-conditioned vivarium equipped with a water flask and a feedbox (Cage Type II-hood of open design, EHRET GNBA, Germany). Two mice were used as donors for tumor samples, while twenty-one mice were recruited to perform the trial. The mice were weighed using an electronic balance (MW-T series, USA, graduation mark 0.01 g). The selected mice were assigned to different groups (10 mice per group). One group was marked as a control series and was not treated with the tested compound. The therapeutic effect was assessed considering the standard tumor growth rate (TGRI ≥ 50%) in the case of subcutaneous administration or T/C ≤ 42% (treatment/control).

Acute Toxicity
Outbred mice (males and females) were used in the experiments. Intraperitoneal injections were applied in the posterior third of the abdomen, slightly departing from the midline. The mice were injected with 1 mL of the solution. Compound 29 was dissolved in a low-molecular-weight polyvinylpyrrolidone (Mw = 8000 ± 200) saline infusion solution (hemodesum). The solution was prepared ex tempore. Doses were calculated for each animal separately, according to body weight data.
The obtained data were statistically processed using Microsoft Office Excel 2010 software. The group arithmetic mean (M) and standard error of the mean (m) were calculated. Statistical significance of the differences was assessed using Student's t-test. The differences between these groups were considered significant at p < 0.05.

Conclusions
In this work, a series of novel small-molecule thiohydantoin-based compounds possessing dispiro-indolinone scaffolds were readily synthesized and evaluated in vitro and in vivo. Cell-based assays revealed several compounds as potent anticancer agents. The most cytotoxic compounds selectively inhibited the growth of androgen-dependent LNCaP cells, providing IC 50 values of~1.2-3.5 µM. Confocal microscopy showed that the product of the intracellular conjugation of propargyl-substituted dispiro-indolinone with the fluorescent dye Sulfo-Cy5-azide was mainly localized in the cytoplasm of HEK293 cells. Animal studies with P388-inoculated and xenograft BALB/c nude mice bearing HCT116 tumor cells demonstrated that, at the sample dose of 1000 mg/kg, TGRI > 90% was achieved after i.p. administration.
On the basis of the performed molecular docking study, MDM2 may be regarded as a possible biological target for the evaluated compounds. However, Western blot data do not confirm this assumption and, apparently, the mechanism of cytotoxic action of the synthesized compounds is different from that of nutlin. Nevertheless, some obtained thiohydantoin-based dispiro-indolinones demonstrated significant cytotoxic effects in vitro and certain antitumor properties in vivo, making this class of compounds promising for further optimization and research. Thus, compound 29 demonstrates significant anticancer activity at a daily dose of 1000 mg/kg (i.p., 10 days) in HCT116-bearing xenograft mice, has a therapeutic index (LD 50 /ED 50 ) of at least 2.5 in mice, and does not have a cumulative effect.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/molecules28031325/s1, General procedure for the synthesis of compounds, NMR spectra, Figure S1: 1 H NMR spectrum of dispiro-indolinone 27, Figure S2: 1 H-1 H COSY NMR spectrum of dispiro-indolinone 27, Figure S3: 1 H-13 C HSQC spectrum of dispiro-indolinone 27, Figure S4: 1 H-13 C HMBC spectrum of dispiro-indolinone 27, Figure S5: The mechanism of cell death (necrosis/apoptosis) caused by compound 12 by flow cytometry staining with 7AAD/AnnexinV, Figure S6: Representative protein gels after 2D electrophoresis, Table S1: Proteins with changed level, determined from 2D-difference gel electrophoresis, Table S2: Anticancer effect of compound 29 achieved after p.o. dose of 170 mg/kg (o.d., 10 days) in P388-inoculated mice, Table S3: Acute toxicity study design for compound 29, Table S4: Dynamics of body weight of mice in the study of acute toxicity of compound 29, Table S5: Average group indicators of the relative mass of the internal organs of mice in the study of acute toxicity of the compound 29. References [85,86]