Design, Synthesis, and Biological Investigation of Novel Classes of 3-Carene-Derived Potent Inhibitors of TDP1

Two novel structural types of tyrosyl-DNA phosphodiesterase 1 (TDP1) inhibitors with hexahydroisobenzofuran 11 and 3-oxabicyclo [3.3.1]nonane 12 scaffolds were discovered. These monoterpene-derived compounds were synthesized through preliminary isomerization of (+)-3-carene to (+)-2-carene followed by reaction with heteroaromatic aldehydes. All the compounds inhibit the TDP1 enzyme at micro- and submicromolar levels, with the most potent compound having an IC50 value of 0.65 μM. TDP1 is an important DNA repair enzyme and a promising target for the development of new chemosensitizing agents. A panel of isogenic clones of the HEK293FT cell line knockout for the TDP1 gene was created using the CRISPR-Cas9 system. Cytotoxic effects of topotecan (Tpc) and non-cytotoxic compounds of the new structures were investigated separately and jointly in the TDP1 gene knockout cells. For two TDP1 inhibitors, 11h and 12k, a synergistic effect was observed with Tpc in the HEK293FT cells but was not found in TDP1 −/− cells. Thus, it is likely that the synergistic effect is caused by inhibition of TDP1. Synergy was also found for 11h in other cancer cell lines. Thus, sensitizing cancer cells using a non-cytotoxic drug can enhance the efficacy of currently used pharmaceuticals and, concomitantly, reduce toxic side effects.


Introduction
There are a number of unresolved issues regarding chemotherapeutic treatment for oncological diseases, for example, low efficiency, resistance of malignant tumors, side effects, and high toxic load. The cytotoxic effect of chemotherapy is caused by DNA damage, and the ability of cancer cells to repair DNA lesions results in resistance. Therefore, the development of DNA repair enzyme inhibitors is a promising strategy to improve the efficacy of DNA damaging agents in clinical use.
Tyrosyl-DNA-phosphodiesterase 1 (TDP1) is a DNA repair enzyme and a promising therapeutic target to enhance established cancer treatment [1][2][3][4]. TDP1 plays a key role in the removal of DNA lesions, in particular those caused by DNA topoisomerase I (TOP1) inhibitors. TOP1 is an essential enzyme that regulates DNA topology by reducing DNA supercoiling. TOP1 introduces a transient single-strand break in DNA, enabling the broken strand to rotate around the TOP1-bound DNA during fundamental cellular events, such as replication, transcription, and repair [5]. TOP1 inhibitors, for example, topotecan (Tpc) and irinotecan, both important chemotherapeutic agents, stabilize the cleavage complex TOP1-DNA, leading to cell death [6]. However, TDP1 removes the stalled TOP1-DNA complex, reducing the efficacy of the drugs and leading to resistance [7][8][9]. As has been confirmed in gene knockout studies in mice and human cell lines, TDP1 contributes to the development of drug resistance in a number of cancers (see [3] and references therein). Indeed, human cell lines with mutant or missing TDP1 as well as TDP1-knockout mice are both hypersensitive to the well-established TOP1 inhibitor camptothecin and its derivatives. Conversely, at high TDP1 expression levels, camptothecin and etoposide cause limited DNA damage [3]. Thus, a promising approach to increase the selectivity and potency of TOP1 poisons to cancer cells is to combine them with TDP1 inhibitors, which can significantly increase the chemotherapeutic efficacy.
Monoterpene (+)-3-carene 8 (Scheme 1) is one of the main components of turpentine. Nevertheless, this compound is very rarely considered as a molecular platform for the development of new therapeutics and never considered for TDP1 inhibitors. In the current work, we studied the anti-TDP1 activity of (+)-3-carene derivatives for the first time. Two sets of (+)-3-carene derivatives with hexahydroisobenzofuran and 3-oxabicyclo [3.3.1]nonane skeletons were synthesized through preliminary isomerization of 3-carene to 2-carene followed by reaction with aldehydes. As a result, new structural types of potent TDP1 inhibitors were discovered. Furthermore, to elucidate the inhibitors' mechanism of action on the cellular level, a panel of HEK293FT TDP1 knockout isogenic clones was created using the CRISPR-Cas9 approach and the cytotoxic potentials were measured. developed that contain monoterpene moieties; for example, coumarin [16], 4-arylcoumarin [17], and adamantane [18][19][20][21] derivatives as well as octahydro-2H-chromenes 3 (see Figure 1) synthesized from p-menthane monoterpenoid (-)-isopulegol [22]. Their ability to enhance the Tpc cytotoxic potential was demonstrated for several of these derivatives. For example, 7-hydroxycoumarin derivative 4 ( Figure 1) with a pinene-type fragment reduced the CC50 (median cytotoxic concentration causing 50% cell death) of camptothecin eight-fold for the human breast adenocarcinoma cell line MCF-7 [16] while the hybrid molecule 5, comprising of (4-fluorophenyl)-coumarin and geraniol moieties, showed a significant increase in the antitumor effect of Tpc on Krebs-2 ascites in an in vivo tumor model [17]. Among the adamantane derivatives, strong synergism was found for 6 and 7 (Figure 1), both containing the acyclic monoterpene substituents, when lung adenocarcinoma cells A549 were treated in combination with Tpc [18,20].
Monoterpene (+)-3-carene 8 (Scheme 1) is one of the main components of turpentine. Nevertheless, this compound is very rarely considered as a molecular platform for the development of new therapeutics and never considered for TDP1 inhibitors. In the current work, we studied the anti-TDP1 activity of (+)-3-carene derivatives for the first time. Two sets of (+)-3-carene derivatives with hexahydroisobenzofuran and 3-oxabicyclo [3.3.1]nonane skeletons were synthesized through preliminary isomerization of 3-carene to 2-carene followed by reaction with aldehydes. As a result, new structural types of potent TDP1 inhibitors were discovered. Furthermore, to elucidate the inhibitors' mechanism of action on the cellular level, a panel of HEK293FT TDP1 knockout isogenic clones was created using the CRISPR-Cas9 approach and the cytotoxic potentials were measured.

Chemistry
It has been shown that carene can be successfully reacted with carbonyl compounds [23,24]. Previously, we showed that condensation of 2-carene 9 with aldehydes 10a-c in the presence of montmorillonite clays led to the formation of hexahydroisobenzofurans 11a-c as a mixture of two isomers at the C(1) position and compounds with 3-oxabicyclo[3 .3.1]nonane structure 12 as minor products (Scheme 1). It was found that hexahydroisobenzofuran 11a obtained with condensation of 2-carene 9 with 4-hydroxy-3-methoxybenzaldehyde 10a demonstrated potent neuroprotective activity in an in vivo Parkinson model combined with low acute toxicity [24]. The reactions of 3carene 8 with aldehydes gave the same products 11 and 12 as it did for 2-carene 9 but in much lower yields (only 3% for the reaction with aldehyde 10a), since 2-carene 9 is more reactive due to the conjugation of the double bond with the cyclopropane ring [23]. Scheme 1. Reaction of carenes with aldehydes 10a-c [23].
Interestingly, 2-carene 9 is widely found in nature albeit in small concentrations. The isolation of 9 from essential oils, or from reaction mixtures after 3-carene 8 isomerization, is difficult, making Scheme 1. Reaction of carenes with aldehydes 10a-c [23].

Chemistry
It has been shown that carene can be successfully reacted with carbonyl compounds [23,24]. Previously, we showed that condensation of 2-carene 9 with aldehydes 10a-c in the presence of montmorillonite clays led to the formation of hexahydroisobenzofurans 11a-c as a mixture of two isomers at the C(1) position and compounds with 3-oxabicyclo[3.3.1]nonane structure 12 as minor products (Scheme 1). It was found that hexahydroisobenzofuran 11a obtained with condensation of 2-carene 9 with 4-hydroxy-3-methoxybenzaldehyde 10a demonstrated potent neuroprotective activity in an in vivo Parkinson model combined with low acute toxicity [24]. The reactions of 3-carene 8 with aldehydes gave the same products 11 and 12 as it did for 2-carene 9 but in much lower yields (only 3% for the reaction with aldehyde 10a), since 2-carene 9 is more reactive due to the conjugation of the double bond with the cyclopropane ring [23].
Interestingly, 2-carene 9 is widely found in nature albeit in small concentrations. The isolation of 9 from essential oils, or from reaction mixtures after 3-carene 8 isomerization, is difficult, making 9 Molecules 2020, 25, 3496 4 of 22 expensive and therefore not attractive for use in organic synthesis [25][26][27][28][29]. Recently, we developed a method for the isomerization of readily available 3-carene 8 to 2-carene 9 using montmorillonite clays as a catalyst [30]. It was demonstrated that a mixture of carenes, limonene, and terpinenes containing 10% 2-carene can be effectively used as a starting material instead of pure 2-carene in a clay-catalyzed reaction with 4-hydroxy-3-methoxybenzaldehyde 10a, resulting in chiral isobenzofurans 11a. The largest yield of 11a was obtained with the 2-carene-containing mixture (60%) [31], which is much higher than using pure 2-carene 9 (33%). At the same time, condensation of carenes with aldehydes has not yet been systematically studied. Here, we synthesized heterocyclic compounds using 2-carene from a mixture containing carenes, limonene, and terpinenes with 2-carene comprising 10%, by condensation with heteroaromatic furan and thiophene aldehydes. The choice of this aldehyde was based on the presence of furan and thiophene moieties in many bioactive compounds, including TDP1 inhibitors [32][33][34][35], as well as preliminary molecular modelling that indicated favourable binding to TDP1.
Molecules 2020, 25,3496 4 of 23 9 expensive and therefore not attractive for use in organic synthesis [25][26][27][28][29]. Recently, we developed a method for the isomerization of readily available 3-carene 8 to 2-carene 9 using montmorillonite clays as a catalyst [30]. It was demonstrated that a mixture of carenes, limonene, and terpinenes containing 10% 2-carene can be effectively used as a starting material instead of pure 2-carene in a clay-catalyzed reaction with 4-hydroxy-3-methoxybenzaldehyde 10a, resulting in chiral isobenzofurans 11a. The largest yield of 11a was obtained with the 2-carene-containing mixture (60%) [31], which is much higher than using pure 2-carene 9 (33%). At the same time, condensation of carenes with aldehydes has not yet been systematically studied. Here, we synthesized heterocyclic compounds using 2-carene from a mixture containing carenes, limonene, and terpinenes with 2carene comprising 10%, by condensation with heteroaromatic furan and thiophene aldehydes. The choice of this aldehyde was based on the presence of furan and thiophene moieties in many bioactive compounds, including TDP1 inhibitors [32][33][34][35], as well as preliminary molecular modelling that indicated favourable binding to TDP1. For the synthesis of heterocyclic compounds, we used a mixture of terpenes obtained by the isomerization of 3-carene 8 in the presence of montmorillonite K10 [31], which consists of 10% 2carene, 30% 3-carene, 20% m-and p-cymols, 12% terpinenes, 5% limonene, and a number of dimeric compounds (m/z = 272) (according to gas chromatography-mass spectrometry (GC-MS) data). The reaction pathways for 3-carene isomerization are shown in Scheme S1, Supplementary Material.

Scheme 2.
Reaction of 2-carene-containing mixture with aldehydes 10a, c-k.  The reactants for the products with the 3-oxabicyclo[3.3.1]nonane framework 12a in reactions of the 2-carene-containing mixture with aldehydes can be both 2-and 3-carene as well as limonene and possibly terpinenes [15,23]; a proposed reaction mechanism for compounds 11, 12, and 13 is shown in Scheme 3. The yields for 11, 12, and 13 were calculated based on the 2-carene concentration in the mixture and on the total concentration of carenes, limonene, and terpinenes.
Molecules 2020, 25, 3496 5 of 23 possibly terpinenes [15,23]; a proposed reaction mechanism for compounds 11, 12, and 13 is shown in Scheme 3. The yields for 11, 12, and 13 were calculated based on the 2-carene concentration in the mixture and on the total concentration of carenes, limonene, and terpinenes. Scheme 3. The proposed reaction mechanism for compounds 11, 12, and 13.
The reaction of crotonaldehyde 10c with the 2-carene-containing mixture on K10 resulted in isobenzofurans 11c ((S)/(R) = 1.3:1, total yield of 89%). The reaction yielded more 11c than the 2carene's concentration in the mixture, with a higher product yield than for the reaction only using 2carene 9. It is therefore clear that the mixture can be successfully used without isolating the 2-carene 9. We also studied the reactions of the 2-carene-containing mixture with the heteroaromatic aldehydes 10d-k (thiophenecarboxaldehyde, furaldehyde, and their derivatives). Compound (S)-11d with a hexahydroisobenzofuran moiety was synthesized in a 78% yield, reacting the 2-carenecontaining mixture with 2-thiophenecarboxaldehyde 10d catalyzed by montmorillonite K10. In addition, a mixture of two compounds 12d and 13d with a 3-oxabicyclo[3.3.1]nonane skeletone was formed, which are isomers depending on the double bond position (12d/13d = 7:1, total yield 5%) (Scheme 2, Table 1). It should be noted that compound 11d, in contrast to the previously studied reactions with aromatic aldehydes, the crotonaldehyde, formed exclusively as an (S)-isomer [23]. Moreover, although Prins cyclization in reactions of terpenoids with aldehydes, leading to the formation of heterocycles, with the 3-oxabicyclo[3.3.1]nonane framework is well studied [15,23], the only example of isolating a compound with an exocyclic double bond is the reaction of α-terpineol 14 with benzaldehyde in the presence of BF3 * Et2O, leading to 13d [36].
The reaction of crotonaldehyde 10c with the 2-carene-containing mixture on K10 resulted in isobenzofurans 11c ((S)/(R) = 1.3:1, total yield of 89%). The reaction yielded more 11c than the 2-carene's concentration in the mixture, with a higher product yield than for the reaction only using 2-carene 9. It is therefore clear that the mixture can be successfully used without isolating the 2-carene 9.
We also studied the reactions of the 2-carene-containing mixture with the heteroaromatic aldehydes 10d-k (thiophenecarboxaldehyde, furaldehyde, and their derivatives). Compound (S)-11d with a hexahydroisobenzofuran moiety was synthesized in a 78% yield, reacting the 2-carene-containing mixture with 2-thiophenecarboxaldehyde 10d catalyzed by montmorillonite K10. In addition, a mixture of two compounds 12d and 13d with a 3-oxabicyclo[3.3.1]nonane skeletone was formed, which are isomers depending on the double bond position (12d/13d = 7:1, total yield 5%) (Scheme 2, Table 1). It should be noted that compound 11d, in contrast to the previously studied reactions with aromatic aldehydes, the crotonaldehyde, formed exclusively as an (S)-isomer [23]. Moreover, although Prins cyclization in reactions of terpenoids with aldehydes, leading to the formation of heterocycles, with the 3-oxabicyclo[3.3.1]nonane framework is well studied [15,23], the only example of isolating a compound with an exocyclic double bond is the reaction of α-terpineol 14 with benzaldehyde in the presence of BF 3 * Et 2 O, leading to 13d [36].
It should be noted that in the case of the 2-carene-containing mixture reactions with aldehydes 10, the main or even the only way of isobenzofuran synthesis was the formation of (S)-isomers (excluding the interaction with 4-hydroxy-3-methoxybenzaldehyde 10a). Predominant formation of the (S)-isomer of isobenzofuran was clearly observed for the condensation of 2-carene and anisaldehyde [37]. According to denstity functional theory (DFT) calculations, the formation of (S)-isomer is energetically more favorable (by 6.8 kJ/mol) compared to formation of the (R)-isomer [37]. Moreover, the lower stability of the (R)-isomer in comparison with the (S)-isomer due to secondary transformations, leading to resinofication under the reaction conditions compared with (S)-isomer, was also shown [37].
Reactions of 2-and 3-furaldehyde as well as pyrrole-2-carboxaldehyde with the 2-carene-containing mixture did not lead to the formation of the target heterocyclic compounds. According to GC-MS data, after the reactions, 2-carene remained unconverted, but limonene was completely consumed and traces of compounds with m/z values corresponding to the intermolecular products, most likely with the 3-oxabicyclo[3.3.1]nonane skeleton, were observed. Apparently, the intermolecular products formed during the reaction are unstable and undergo degradation in the presence of an acid catalyst.
It should be noted that unreacted monoterpenes and other substances from the 2-carene-containing mixture are easily separated from the more polar condensation products by column chromatography using hexane as an elute. Moreover, compounds with the 3-oxabicyclo[3.3.1]nonane sceletone 12 can also be synthesized based on more available limonene [15]. For example, condensation of limonene with 2-thiophenecarboxaldehyde 10d catalyzed by montmorillonite K10 without solvent at room temperature after 2 h led to the formation of compounds 12d and 13d (total yield 64%, 12d/13d = 14:1).
Thus, we obtained a wide range of hydrogenated isobenzofuran and 3-oxabicyclo[3.3.1]nonane heterocycles, including compounds with an exocyclic double bond.

Structure-Activity Relationship Analysis
The compounds synthesized starting from (+)-3-carene 8 were tested for their TDP1 inhibitory properties by measuring their IC 50 values using a real-time fluorescent oligonucleotide biosensor [38]. This biosensor demonstrated high sensitivity and specificity for the detection of TDP1 activity. The commercially available TDP1 inhibitor furamidine, a bisbenzamidine derivative belonging to the diamidines family [34], was used as a positive control (IC 50 1.2 ± 0.3 µM). The results for all the derivatives are summarized in Table 2.
The addition of bromine onto the heterocycle significantly increased the inhibition potency of the ligands. Thus, compound 11h with a bromine atom at the 5-position of the thiophene ring had the best activity in the submicromolar range of the hexahydroisobenzofurans (IC 50 0.75 µM). From the compounds with the 3-oxabicyclo[3.3.1]nonane skeleton, the most active was 4-bromine-substituted compound 12g (IC 50 0.65 µM). A bromine shift from position 4 to 5 from 12g to 12h decreased the inhibition potency (IC 50 1.75 µM); nevertheless, 12h is approximately two-fold more active than 12d, which does not have substituents on the thiophene ring.

Cell Growth and Viability
Topotecan Cytotoxicity on HEK293FT Wild Type and TDP1−/− Cells TDP1 plays an essential role in the resistance of cancer cells to antitumour TOP1-inhibiting drugs. It was shown previously for HEK293 [39] and other cell lines that cells lacking TDP1 expression exhibit hypersensitivity to camptothecin and its derivatives, such as Tpc [40]. To examine the potential impact of TDP1 knockout on cell survival and dosing with the synthesized compounds, we generated HEK293FT TDP1-deficient (TDP1−/−) cell line. Using a paired gRNA CRISPR-Cas9 strategy and polymerase chain reaction (PCR) screening of cell clones, we obtained three clones containing deletions in the first protein coding the exon of the TDP1 gene. PCR products obtained from cell clones containing deletions were further cloned in a T-vector and 10 independent plasmid clones were sequenced by the Sanger method. Alignments of these deletions with the wild-type TDP1 gene sequence showed that modifications in all three clones lead to a shift in the reading frame of the gene and should lead to impaired synthesis of TDP1 ( Figure S1). Clone B2 was found to have a 197-bp deletion in one of the TDP1 alleles, as well as a 16-bp deletion and T nucleotide insertion in the other allele; clone B5 carried 198-bp and 196-bp deletions; and the E3 clone had deletions of 197bp and 200bp. We treated TDP1-proficient (wild-type, WT) and TDP1-deficient (TDP1−/−) cells with increasing concentrations of Tpc. The results for all three clones are presented in Figure S2A. Clone E3 was selected for further work as it has a homozigous deletion causing an open reading frame shift in both alleles of the TDP1 gene and demonstrated sensitivity to Tpc. Clone E3 TDP1 mRNA was confirmed to have a 197/200-bp deletion causing the open reading frame shift ( Figures S1, S2B). There is a deletion at the level of genomic DNA ( Figure S2B, lane 4) and only the mutant form was transcribed ( Figure S2B, lane 7). The clone E3 was subsequently screened by the biochemical assay for 3 -phosphotyrosyl cleavage activity to identify detectable TDP1 activity in the cell extract ( Figure S2C). None was found in contrast to the control WT cell extract and purified TDP1.
We analyzed, using a colorimetric test, the relative amount of viable cells after treating HEK293FT WT and TDP1-deficient (TDP1−/−) cells with increasing concentrations of Tpc for 72 h. TDP1 knockout reduced cell viability, where the IC 50 values in H 2 O were 150 nM for TDP1−/− and 250 nM for WT cells (Figure 2A); and in DMSO, they were 30 nM for TDP1−/− and 200 nM for WT cells. Next, we studied WT and mutant cells' live activity after Tpc treatment on the xCELLigence System (ACEA Biosciences, USA) by the impedance-based assay. HEK293FT TDP1−/− cells were more sensitive to Tpc ( Figure 2B). After Tpc administration to the cells, an increase of the slope was seen, characterizing the velocity of the cell growth with Tpc as compared to the control both for WT (red columns) and mutant cells (green columns) ( Figure S1, Supplementary Materials). The analysis of the averaged slope for the time period of~70 h after Tpc treatment showed that the slope decreased depending on the Tpc concentration ( Figure 2CII). The sharper decrease of the slope for the mutant cells than for the WT (compare the histogram for the 0 and 0.05 µM Tpc concentration,~1.5 times for WT cells (red) and~6 times for TDP1−/−cells (green)) also characterizes the higher sensitivity of the mutant to the TOP1 inhibitor. The Tpc IC 50 values determined by the impedance-based assay were 60 nM for WT and 20 nM for TDP1−/− cells in H 2 O. There was a three-fold difference for these IC 50 values, which is nearly the same as for the colorimetric assay. The absolute values slightly differed for these methods and solvents, but still, HEK293FT TDP1−/− cells showed a higher sensitivity to the TOP1 inhibitor Tpc, thus verifying the contribution of TDP1 on cell survival.   25 μM), 12g (0.65 μM), 12h (1.75 μM), and 12k (1.2 μM). Interestingly, cytotoxicity was absent, or insignificant, in the range of concentrations (0.08-100 μM) for all the   25 µM), 12g (0.65 µM), 12h (1.75 µM), and 12k (1.2 µM). Interestingly, cytotoxicity was absent, or insignificant, in the range of concentrations (0.08-100 µM) for all the compounds in both cell lines ( Figure 3A). The absence of toxicity is very important since it will not lead to undesirable side effects. Next, we checked the influence of the eight most potent TDP1 inhibitors on the cytotoxic effect of Tpc on the WT and mutant cell lines. We chose 30 and 700 nM Tpc concentrations for the TDP1−/− and WT cells, respectively, which are close to the IC50 values for Tpc in DMSO in the colorimetric assay ( Figure 2A). Then, we selected two candidates (11h and 12k) for subsequent studies based on their ability to inhibit TDP1 activity and low cytotoxicity. We tested the HEK293FT cells' viability after treatment with 11h and 12k and in conjunction with Tpc in a concentration close to its IC50 value (30 nM for TDP1−/− and 200 nM for WT cells). Suppressed cell growth was displayed with the TDP1 inhibitors and Tpc on the WT cells; however, no effect was seen for Tpc in TDP1-deficient cells ( Figure  3B). Thus, the synergistic action of Tpc in conjunction with 11h and 12k on HEK293FT WT cells is most likely due to the TDP1 inhibition making it the main target of action. The initial increase in cell proliferation on the graphs up to the 20 μM concentration can be observed because we diluted the inhibitor with the medium before adding it to the well plate. Thus, a stimulation effect at low inhibitor Next, we checked the influence of the eight most potent TDP1 inhibitors on the cytotoxic effect of Tpc on the WT and mutant cell lines. We chose 30 and 700 nM Tpc concentrations for the TDP1−/− and WT cells, respectively, which are close to the IC 50 values for Tpc in DMSO in the colorimetric assay ( Figure 2A). Then, we selected two candidates (11h and 12k) for subsequent studies based on their ability to inhibit TDP1 activity and low cytotoxicity. We tested the HEK293FT cells' viability after treatment with 11h and 12k and in conjunction with Tpc in a concentration close to its IC 50 value (30 nM for TDP1−/− and 200 nM for WT cells). Suppressed cell growth was displayed with the TDP1 inhibitors and Tpc on the WT cells; however, no effect was seen for Tpc in TDP1-deficient cells ( Figure 3B). Thus, the synergistic action of Tpc in conjunction with 11h and 12k on HEK293FT WT cells is most likely due to the TDP1 inhibition making it the main target of action. The initial increase in cell proliferation on the graphs up to the 20 µM concentration can be observed because we diluted the inhibitor with the medium before adding it to the well plate. Thus, a stimulation effect at low inhibitor concentrations could be observed due to the addition of fresh medium.
Activity of 11h and 12k with Topotecan against Tumor Cells Next, we checked the cytotoxicity of the combination of Tpc with 11h and 12k against several cancer cell lines and the results are shown in Figure 4.  The intrinsic cytotoxicity of both 11h and 12k on HeLa (cervical carcinoma), HCT116 (human colon cancer), IMR-32 (human neuroblastoma cell line), and Krebs-2 carcinoma cells was low ( Figure  4A). Due to their low cytotoxicity, these derivatives were accepted as being suitable for further study on their cytotoxic effect with Tpc ( Figure 4B, C). Both compounds showed no enhancement of the Tpc cytotoxicity for HeLa cells (data not shown). We compared HCT and IMR-32 cells' viability after treatment with 11h and 12k alone and in the presence of Tpc in a concentration close to their IC50 values (900 nM, HCT; and 8 nM, IMR-32 cells). We observed suppressed cell growth in the joint presence of Tpc with 11h or 12k on the IMR-32 cells and no effect of Tpc on the HCT cell line ( Figure  4B). We tested HCT and IMR-32 cells' viability after Tpc treatment in the absence and presence of 11h/12k ( Figure 4C). Both 11h and 12k demonstrated synergistic action with Tpc on IMR-32 cell growth. 11h also enhanced Tpc's potency on HCT116 cells. These compounds show promise as candidates for anticancer therapy in conjunction with Tpc. The intrinsic cytotoxicity of both 11h and 12k on HeLa (cervical carcinoma), HCT116 (human colon cancer), IMR-32 (human neuroblastoma cell line), and Krebs-2 carcinoma cells was low ( Figure 4A). Due to their low cytotoxicity, these derivatives were accepted as being suitable for further study on their cytotoxic effect with Tpc ( Figure 4B,C). Both compounds showed no enhancement of the Tpc cytotoxicity for HeLa cells (data not shown). We compared HCT and IMR-32 cells' viability after treatment with 11h and 12k alone and in the presence of Tpc in a concentration close to their IC 50 values (900 nM, HCT; and 8 nM, IMR-32 cells). We observed suppressed cell growth in the joint presence of Tpc with 11h or 12k on the IMR-32 cells and no effect of Tpc on the HCT cell line ( Figure 4B). We tested HCT and IMR-32 cells' viability after Tpc treatment in the absence and presence of 11h/12k ( Figure 4C). Both 11h and 12k demonstrated synergistic action with Tpc on IMR-32 cell growth. 11h also enhanced Tpc's potency on HCT116 cells. These compounds show promise as candidates for anticancer therapy in conjunction with Tpc.

Chemical Space
The calculated molecular descriptors MW (molecular weight), log P (water-octanol partition coefficient), HD (hydrogen bond donors), HA (hydrogen bond acceptors), PSA (polar surface area), and RB (rotatable bonds)) are given in Table S2. The molecular weight of the ligands lies between 206.3 and 327.3 g mol −1 , falling into lead-and drug-like chemical spaces. The log P values range from 3.2 and 5.1, lying mostly in the drug-like but reaching into known drug space whilst the HD, HA, RB, and PSA values are within the lead-like space (for the definition of lead-like, drug-like, and known drug space (KDS) regions (see [41], Table S3).
The known drug index (KDI) of each ligand was calculated to gauge the balance of the molecular descriptors of the ligands. This method is based on the statistical analysis of drugs in clinical use (KDS) and a weighted index for each molecular descriptor. Both the summation (KDI 2a ) and multiplication (KDI 2b ) methods were used [42]; as shown for KDI 2a in Equation (1) and for KDI 2b in Equation (2). Finally, the numerical results are given in Table S4 The KDI 2a values range from 4.15 to 5.15, with a theoretical maximum of 6 and an average of 4.08 for known drugs. KDI 2b ranges from 0.10 to 0.39, with a theoretical maximum of 1 and with a KDS average of 0.18. This indicates that the majority of the ligands are well balanced. 11h has KDI 2a of 4.25 and KDI 2b of 0.11 and 12g has similar values, with KDI 2a of 4.16 and KDI 2b of 0.10. The KDI 2a of both compounds is above the average index value for known drugs; however, the KDI 2b is below the average mainly due to the low PSA values. Nevertheless, these ligands can be considered reasonably well balanced in terms of their molecular descriptors.

Chemistry
All commercially available compounds and solvents were reagent grade and used without further treatment unless otherwise noted. We used montmorillonite K10 clay (Sigma-Aldrich, St. Louis, MO, USA) as the catalyst. The clay was calcinated at 105 • C for 3 h immediately before use. CH 2 Cl 2 was passed through calcined Al 2 O 3 . The 2-carene-containing mixture was synthesized according to Sidorenko et al. [31]  Numeration for the carbon atoms used for assignment in NMR spectra is presented in Figure 5 with compound (S)-11f and 12f as an example.
Molecules 2020, 25,3496 13 of 23 Numeration for the carbon atoms used for assignment in NMR spectra is presented in Figure 6 with compound (S)-11f and 12f as an example.

General Procedure (GP)
To the suspension of the K10 clay (3.0 g) in methylene chloride (15 mL) solution of 0.40 g aldehyde in methylene chloride (10 mL) and 1.0 g of a 2-carene containing mixture (10 wt.% of 2carene., 57 wt.% of the sum of 2-and 3-carenes, limonene, and menthadienes., obtained by isomerization of 3-carene according to [30,31], were added. The solvent was distilled off, and the reaction mixture was kept at rt for 20 h. Thereafter, ethyl acetate was added, the catalyst was filtered off, and the filtrate was evaporated. Isolation of the reaction products was carried out by column chromatography (silica gel (17 g), hexane/Et2O 100:0 → 90:10). Compounds 11 were eluted with a 4-6% solution of Et2O in hexane, and compounds 12 and 13 with a 0.5-2% solution of Et2O in hexane. The yields of compounds 11, 12, and 13 were calculated based on the 2-carene content and on the sum of 2-and 3-carenes, limonene, and menthadienes, respectively.

General Procedure (GP)
To the suspension of the K10 clay (3.0 g) in methylene chloride (15 mL) solution of 0.40 g aldehyde in methylene chloride (10 mL) and 1.0 g of a 2-carene containing mixture (10 wt.% of 2-carene., 57 wt.% of the sum of 2-and 3-carenes, limonene, and menthadienes., obtained by isomerization of 3-carene according to [30,31], were added. The solvent was distilled off, and the reaction mixture was kept at rt for 20 h. Thereafter, ethyl acetate was added, the catalyst was filtered off, and the filtrate was evaporated. Isolation of the reaction products was carried out by column chromatography (silica gel (17 g), hexane/Et 2 O 100:0 → 90:10). Compounds 11 were eluted with a 4-6% solution of Et 2 O in hexane, and compounds 12 and 13 with a 0.5-2% solution of Et 2 O in hexane. The yields of compounds 11, 12, and 13 were calculated based on the 2-carene content and on the sum of 2-and 3-carenes, limonene, and menthadienes, respectively.

Real-Time Detection of TDP1 Activity
A fluorophore quencher-coupled oligonucleotide biosensor was used for TDP1 enzyme activity real-time fluorescence detection [38]. The biosensor 5 -FAM-AAC GTC AGG GTC TTC C-BHQ1-3 is a 16-mer single-stranded oligonucleotide with a 5 -fluorophore (FAM), and a 3 -quencher (BHQ1). Recombinant protein TDP1 was expressed in Escherichia coli (pET 16B plasmid containing TDP1 cDNA was provided by Dr. K.W. Caldecott, University of Sussex, United Kingdom) and isolated as described by Lebedeva et al. [43]. The reaction mixture contained TDP1 reaction buffer (50 mM Tris-HCl, 50 mM NaCl, 7 mM β-mercaptoethanol), 50 nM oligonucleotide substrate, varied concentrations of the tested compounds, and 1.5 nM purified TDP1 in a final volume of 200 µL.
The TDP1 reaction mixture was incubated at rt in a POLARstar OPTIMA fluorimeter (BMG LABTECH, GmbH). Fluorescence intensity was measured every 1 min (Ex485/Em520 nm). The efficiency of TDP1 inhibition was evaluated by comparing the rate of increase in the fluorescence of the biosensor in the presence of the compound to that of DMSO (1.5%) control wells. IC 50 values were determined using an 11-point concentration-response curve. The data were imported into MARS Data Analysis 2.0 program (BMG LABTECH) and the slope during the linear phase was calculated. The IC 50 measurements were carried out in at least three independent experiments.

TDP1 Activity by Gel-Based Assay
A fluorophore-labeled oligonucleotide 5 -FAM-AAC GTC AGG GTC TTC C-tyrosine-3 was used as a standard gel-based method for the indication of TDP1 enzyme activity. The oligonucleotide containing a natural phosphotyrosine adduct at the 3 -end is a substrate for TDP1 [33]. The reaction mixture contained reaction buffer (50 mM Tris-HCl, 50 mM NaCl, 7 mM β-mercaptoethanol), 50 nM oligonucleotide substrate, 5 nM purified TDP1, or 1 µg of cell extract in a final volume of 20 µL. The reaction was conducted at 37 • C for 20 min. The reaction products were separated by electrophoresis in a 20% denaturing polyacrylamide gel with 7 M urea. A Typhoon FLA 9500 phosphorimager (GE Healthcare) was used for gel scanning and imaging, and the data were analyzed with QuantityOne 4.6.7 software.

Knockout HEK293FT Clone Generation
First, 5 × 10 5 HEK293FT cells were plated into a well of a 12-well plate and cotransfected with the constructed plasmids TDP1-gRNA1 and TDP1-gRNA2 (0.25 µg of each) using Lipofectamine 3000 Reagent (Thermofisher Scientific). Growth medium contained DMEM/F12 (Gibco, Co Dublin, Ireland) 1:1, 10% fetal bovine serum (FBS) (Gibco), 100 U/mL penicillin-streptomycin (Gibco), and 1x GlutaMAX (Gibco). Then, 48 h after transfection, cells were detached using TrypLE Express (TrypLE, Gibco), and the GFP-positive cell population was enriched by cell sorting using a BD FACSAria III Cell Sorter (BD Biosciences, East Rutherford, NJ, USA). Transfected cells were plated onto a 96-well plate, one cell per well. Single-cell clones grew for two weeks before they were replicated to another 96-well plate, so we obtained two equal 96-well plates with cell clones: One plate was used for PCR analysis of the deletion in the TDP1 gene while the other plate was used for the mutant cell clone multiplicatioin.

Analysis of CRISPR/Cas9-Mediated Deletions in the TDP1 Gene
Genome DNA was extracted from cells on one of two 96-well plates using 50 µL of QuickExtract™ DNA Extraction Solution (Epicentre) per well. The DNA extracts were diluted with 200 µL of mQ water. Two microliters of diluted DNA extract were used for PCR amplification of the target region with the primers: TDP1-scF 5 -TCAGGAAGGCGATTATGGGAG-3 and TDP1-scR 5 -TTGATGTGGAGGGCTCCAG-3 . Reactions were run on a S1000 Thermal Cycler (Bio-Rad, Hercules, CA ,USA) using BioMaster HS-Taq PCR-Color (2×) (Biolabmix, Novosibirsk Oblast, Russia) with the following program: 95 • C 3 min; 35 cycles: 95 • C 30 s, 68 • C 30 s, 72 • C 10 s; 72 • C 3 min. Products of the reactions were resolved in 1.5% agarose gel stained with ethidium bromide. PCR products with a lower molecular weight than the wild type were selected for subsequent analysis. PCR products were gel purified and cloned in the pGEM-T Easy vector (Promega, Madison, WI, USA). Ten independent plasmid clones were sequenced using M13 universal primers. Products of Sanger sequencing reactions were analyzed using the ABI 3130xl Genetic Analyzer (SB RAS Genomics Core Facility, http://www.niboch.nsc.ru/doku.php/corefacility). Three clones containing deletions in the TDP1 gene were found: Clone B2 ∆197bp/∆16bp+InT; clone B5 ∆198bp/∆196bp; and clone E3 ∆197bp/∆200bp. To study the expression of the mutant form of TDP1 in clone E3, total RNA was isolated from cells using TRIZOL reagent as per the manufacturer's protocols. The synthesis of the first cDNA strand was performed using SuperScript ™ III Reverse Transcriptase (ThermoFisher Scientific) according to the manufacturer's recommendations. To control DNA contamination of the RNA sample, a reaction was performed without the addition of reverse transcriptase (RT-).

Cell Culture Assay
The cytotoxicity of the compounds was examined against human cell lines HEK293FT (human embryonic kidney), WT and TDP1-deficient (TDP1-/-), HCT116 (human colon cancer), IMR-32 (human neuroblastoma cell line), HeLa (cervical cancer), and Krebs-2 carcinoma ascites using MTT and EZ4U colorimetric tests or an xCELLigence DP Real-Time Cell Analyzer (ACEA Biosciences, San Diego, CA, USA) for HEK293FT in cells exposed to the compounds. The cells were grown in DMEM/F12 or IMDM (for IMR-32 and Krebs-2) medium (Gibco), with 1x GlutaMAX (Gibco), 50 IU/mL penicillin, 50 µg/mL streptomycin (MP Biomedicals, Irvine, CA, USA) and in the presence of 10% fetal bovine serum (Biolot). The cells (~20,000 cells per well) were incubated for 24 h at 37 • C in 5% CO 2 , and then treated with the tested compounds (the volume of the added reagents was 1/100 of the total volume of the culture medium, the amount of DMSO was 1% of the final volume as indicated). Control cells were grown in the presence of 1% DMSO where indicated. After 72 h of cell incubation, the relative amount of living cells was determined using a colorimetric test of the amount of formazan converted from 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT test) or EZ4U Cell Proliferation and Cytotoxicity Assay (Biomedica, Austria), as per the manufacturer's protocols. The real-time detection of the living cells was done by an xCELLigence DP Real-Time Cell Analyzer (ACEA Biosciences). The measurements were carried out in three parallel experiments.

Calculation of Molecular Descriptors
The QikProp 3.2 [45] software was used to calculate the molecular descriptors of the ligands. The reliability of QikProp is established for the calculated descriptors [46]. The known drug indexes (KDI) were calculated from the molecular descriptors as described by Eurtivong and Reynisson [42].

Conclusions
A new class of effective TDP1 inhibitors was discovered with activity in the low micromolar to nanomolar concentrations. All the tested derivatives exhibited low intrinsic cytotoxicity when tested in a panel of cancer cell lines. The most promising compounds with the best combination of TDP1 inhibition, low toxicity, and enhancement of Tpc efficacy were compounds 11h and 12k, with IC 50 values of 0.75 and 1.20 µM, respectively. Thus, these new inhibitors are promising candidates for adjuvant therapy, mainly due to the absence of an additional toxic load.
Furthermore, a panel of isogenic clones of the HEK293FT cell line knockout for TDP1 was created using the CRISPR-Cas9 system. Clone E3 with the ∆197bp/∆200bp deletion in the TDP1 gene was selected for the experiments. We investigated the cytotoxic effect of the Tpc and the TDP1 inhibitors 11h and 12k separately and jointly. HEK293FT TDP1-/-cells were more sensitive to Tpc compared to WT cells. For both 11h and 12k, we observed cell growth suppression in the presence of Tpc only for WT cells but not for the TDP1 knockout cells. It can therefore be stated that the synergistic effect with Tpc on HEK293FT cell growth is only caused by TDP1 activity inhibition by 11h and 12k, with minimal off-target effects. These results make a significant contribution to our understanding of the mechanism of action for the non-cytotoxic 3-carene-derived TDP1 inhibitors. In conclusion, we demonstrated the existence of TDP1 inhibitors, which do not modulate other biomolecular targets, which in turn makes the prospect of developing viable drug candidates more realistic since off-target toxicity often leads to undesirable side effects [47].